Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 41 appraisals
ACS biomaterials science & engineering
A Review of Biomass-Derived Nanomaterials as Functional Interfaces for Advanced Electrochemical Hormone Detection
The study of high-performance sensing platforms is essential to achieve a high sensitivity for hormonal analytes, short response times, and cost-effectiveness. Nevertheless, standard analytical procedures, such as high-performance liquid chromatography, inductively coupled plasma mass spectrometry, and gas chromatography mass spectrometry (HPLC, ICP-MS, and GC-MS), are rather limited because their sensitivity is low at pM/nM concentrations, and the costs of these processes are too high. Moreover, modern sensor technology can be ill-equipped to selectively respond to intricate biological sample types, and there is no single standard used to quantify sensor signals. This review addresses these knowledge gaps by providing a comprehensive discussion on the role of biomass-derived materials as useful interfaces in electrochemical hormone sensing systems. To achieve improved results, we critically investigated the recent advances in the electrode modification process through biomass-based nanomaterials to promote electron transfer kinetics and interfacial engineering. The present article also discusses contemporary issues such as problems with stability, reproducibility, and selectivity in complex matrices. Furthermore, we highlighted some of the new approaches, such as the development of artificial intelligence, Internet of Things, and wearable point-of-care devices, to create the next generation hormone sensors. Overall, this current review is a comprehensive overview of biomass-derived materials, which have the prospect of enhancing surface modification at the advanced phase, AI integration, and the development of point-of-care diagnostics.
14 July 2026
Read appraisal →Molecular biology reports
Update of Mexican population databases for complex kinship interpretation with twelve and seven X-STRs
BACKGROUND: X-chromosome short tandem repeats (X-STRs) are valuable for resolving complex kinship cases. Most population databases are based on the ArgusX12 kit (Qiagen), using PCR and capillary electrophoresis (CE) to identify length-based (LB) alleles, and are formatted for FamlinkX software. However, haplotype diversity is often underrepresented by population studies, compelling database updates ( https://famlink.se/fx_databases.html ). The ForenSeq DNA Signature Prep kit (Verogen) uses massive parallel sequencing (MPS) to analyze 7 X-STRs and report sequence-based (SB) alleles, for which population databases are currently unavailable. Because these 7 markers are a subset of the ArgusX12 panel, existing LB-based databases can support kinship interpretation of MPS-derived data. METHODS AND RESULTS: We therefore updated a Mexican population database for 12 X-STRs and generated a corresponding 7-marker subset with FamlinkX-compatible files. Using 500 haplotypes from the Investigator Argus X-12 kit QS (Qiagen), we estimated forensic parameters and evaluated the impact of the updated database by calculating likelihood ratios (LRs) in four representative kinship scenarios. The updated database (n = 1433) increased LRs by an average of 55% compared to the previous version (n = 933). CONCLUSIONS: These results support improved forensic interpretation in Mexican and Latin American populations lacking comprehensive national X-STR databases based on ArgusX12 and ForenSeq DNA Signature kits.
29 June 2026
Read appraisal →Analytica chimica acta
Rapid detection of drug-resistant leukemia cell using an optofluidic chip and machine learning
BACKGROUND: Rapid detection of drug-resistant leukemia played a crucial role in formulating appropriate treatment plans for patients and improving their prognosis. In this research, an integrated optofluidic platform was developed to detect and analyze leukemia and drug-resistant leukemia cell. The detection technique was designed by embedding optical fibers coupled with photosensors and a laser source into optofluidic chip. The scattered light signals were detected when the cells pass through the detecting area. RESULTS: The platform was first validated by classifying 1 μm polystyrene microparticles and 1 μm polystyrene microparticles coated with spherical 10 nm Fe3O4 nanoparticles. After validation, the method was applied to classify leukemia and drug-resistant leukemia cells by injecting the testing sample and obtaining. The SVM classifier demonstrated the highest classification accuracy of 91.1% compared with LR, RF, and KNN classifiers for analyzing leukemia cells and drug-resistant leukemia cells. The proposed method can perform detection within 10 min with a total experimental timeframe of 20 min. SIGNIFICANCE: The presented results demonstrate the feasibility of applying microfluidics and machine learning approaches to detect and classify biological entities with slight variations based on scattered light signals. This platform holds significant potential for clinical diagnostics, offering a rapid, cost-effective, and efficient method for detecting drug-resistant leukemia cells, potentially aiding in personalized treatment strategies.
16 June 2026
Read appraisal →Croatian medical journal
Artificial intelligence enables scale, consistency, and rigor in forensic identity inference.
Advances in forensic genomics, which include massively parallel sequencing, dense single nucleotide polymorphism testing, and forensic genetic genealogy (FGG), have greatly expanded the range of cases in which DNA evidence can generate investigative leads. As a result, the primary limitations in modern forensic DNA analysis are no longer analytical sensitivity or marker availability, but the ability to reason consistently, transparently, and at scale over increasingly complex genetic, genealogical, and contextual information. Current forensic workflows remain predominantly human-centered, relying on manual reasoning that is difficult to standardize, reproduce, document, or scale across growing case inventories. The potential of artificial intelligence (AI) is described as an enabling layer for forensic identity inference, defined here as computational decision-support systems that structure, prioritize, and document reasoning over genetic associations, genealogical structures, and investigative context during identity hypothesis development. While formal statistical inference quantifies evidentiary weight, identity inference governs how evidence is explored, combined, and acted upon during investigations. AI-assisted systems can augment, but not replace, expert judgment by supporting scalable prioritization, relational reasoning, and systematic documentation of analytical decisions, while reducing bias. Properly designed AI-enabled systems offer a path to sustainably scaling FGG while supporting scientific rigor, accountability, and public trust.
15 June 2026
Read appraisal →Developmental cell
A machine learning-based megakaryocyte identification system uncovers resident organs, markers, and functional diversity
The organ- and stage-specific diversity of megakaryocytes (MKs) has prompted a reassessment of their distribution and functions. By integrating single-cell transcriptomic data across multiple organs and developmental stages, we identified previously unreported MK and platelet markers, including Tnik, a key regulator of MK function and platelet production. Using these markers alongside established ones, we developed a machine learning-based MK identification system (MKIDS) that enables MK detection in the brain, heart, and placenta in mice and humans. Functional studies demonstrated that brain-resident MKs are essential for neural development, underscoring organ-specific roles of MKs in regulating tissue development and function. Transcriptomic integration of MKs across organs and stages, with functional validation, revealed a developmental shift in platelet production-from a mitochondria-low to a mitochondria-enriched subpopulation. Our findings offer a transformative perspective on the MK system, highlighting its cellular diversity, functional complexity, and developmental dynamics.
12 June 2026
Read appraisal →European respiratory review : an official journal of the European Respiratory Society
Artificial intelligence-enabled histological analysis in pre-clinical respiratory disease models: a scoping review
Histological analysis is a cornerstone of pre-clinical respiratory disease research. It enables assessment of pathology, therapeutic effects, and mechanisms. However, conventional approaches rely on manual scoring, which is subjective, time-consuming, and difficult to scale. Artificial intelligence (AI), particularly deep learning, offers potential to automate histology workflows. To date, its use in pre-clinical respiratory models has not been synthesised.We conducted a scoping review following the Joanna Briggs Institute guidelines. We searched MEDLINE and Embase (inception - January 2025) for pre-clinical studies using AI to analyse histology in respiratory disease models. Screening, full-text review, and data extraction were performed in duplicate.Of 6271 studies screened, 29 met inclusion criteria. Most used murine models (76%) and investigated lung cancer (28%), pulmonary fibrosis (24%), or tuberculosis (17%). Haematoxylin and eosin was the most common stain (48%), with others targeting collagen or immune markers. AI tasks included image classification (n=20), segmentation (n=10), and object detection (n=4), predominantly using convolutional neural networks (69%). Pre-processing methods (e.g. stain normalisation) were common, but annotation and training practices were inconsistently reported. AI model performance was generally high (accuracy ≥90%; seven studies); however, validation metrics varied, and external validation was absent. Most studies used "black box" models, with minimal application of explainability techniques. Reproducibility measures, such as sharing datasets or code were rarely reported.AI tools are poised to transform histological analysis in pre-clinical respiratory research. The field will be able to further harness AI to automate pre-clinical respiratory histological analysis by addressing gaps that we have identified in validation, transparency, and standardisation.
12 June 2026
Read appraisal →Clinical chemistry and laboratory medicine
T-cell receptor beta constant 1 flow cytometry for diagnosing clonality in mature T-cell neoplasms: a systematic review and meta-analysis of diagnostic accuracy
INTRODUCTION: Accurate assessment of T-cell clonality is key for diagnosing mature T-cell neoplasms. TRBC1-based flow cytometry provides a rapid, robust, and cost-efficient approach. This systematic review and meta-analysis assessed the diagnostic accuracy of TRBC1 flow cytometry (TRBC1-FC) for detecting T-cell clonality in mature T-cell neoplasms. CONTENT: We systematically searched Scopus, PubMed (MEDLINE), and Google Scholar for articles on TRBC1-FC diagnostic accuracy published up to July 1, 2025. Pooled sensitivity and specificity were estimated, between-study heterogeneity was evaluated and small-study effects were examined. SUMMARY: This meta-analysis included 10 studies. The pooled sensitivity was 97.6 % (95 % CI, 95.1-99.4 %) and specificity 90.7 % (95 % CI, 76.0-99.3 %). The pooled LR+ was 10.9 (95 % CI, 4.1-28.9), LR-was 0.053 (95 % CI, 0.025-0.12), and DOR 339 (95 % CI, 64-1,788). The HSROC curve demonstrated an AUC of 0.974 (partial AUC=0.970), confirming excellent global discriminatory capacity. Between-study heterogeneity was substantial (I2=83.3 %), mainly affecting specificity and DOR, while sensitivity remained highly consistent. No evidence of a threshold effect was found. Deeks' test showed significant small-study effects (p<0.001), and sensitivity analyses identified one influential study whose exclusion markedly reduced heterogeneity. These results confirm the high diagnostic performance and robustness of TRBC1-FC for T-cell clonality assessment. OUTLOOK: TRBC1-FC demonstrates high sensitivity and low LR-, supporting its role as a rule-out test. Variability in specificity, LR+ and DOR, mainly due to small-study effects, advises caution for rule-in use. Standardized protocols and cost-effectiveness analyses are needed before broad clinical adoption.
26 May 2026
Read appraisal →Clinical chemistry and laboratory medicine
Automation in value-based laboratory medicine: driving precision, capacity, and better outcomes
Laboratory medicine lies at the core of modern healthcare, enabling timely diagnosis, effective patient monitoring, and increasingly personalized therapeutic strategies. Over the past decades, automation has profoundly reshaped the role of clinical laboratories, substantially enhancing their contribution to clinical outcomes, operational efficiency, and the overall sustainability of healthcare systems. More recently, laboratory automation has emerged as a cornerstone of value-based laboratory medicine, representing not merely a technological upgrade but a strategic transformation of laboratory practice aimed at delivering measurable value to patients and healthcare stakeholders. Although automation has long been established in clinical chemistry and immunoassays, its scope is now expanding to molecular diagnostics and mass spectrometry - two disciplines that are central to precision medicine. Looking ahead, the convergence of automation, digitalization, and artificial intelligence is driving the emergence of hyperautomation in laboratory medicine. Within this paradigm, laboratories evolve from isolated testing units into integrated diagnostic hubs, in which results from multiple laboratory disciplines are harmonized and contextualized to effectively support clinical decision-making.
26 May 2026
Read appraisal →ACS sensors
Urine-Based Noninvasive Detection of Prostate Cancer Using Human Olfactory Receptor-Embedded Nanodiscs
The diagnosis of prostate cancer (PCa) is limited by the low specificity of prostate-specific antigen (PSA) testing, which contributes to overdiagnosis and patient exposure to unnecessary invasive prostate biopsies. Here, we report a urine-based diagnostic platform that combines a six-member human olfactory receptor-embedded nanodisc (OR-ND) sensor array with fluorescence sensing and machine learning to detect PCa-associated volatile organic compounds (VOCs). Six ORs were reconstituted into lipid nanodiscs and evaluated using urine samples from a strictly curated subcohort of 40 PCa patients and 33 healthy controls (n = 73) to establish baseline sensor responsiveness. Subsequently, to ensure diagnostic robustness, machine learning classifiers were developed and cross-validated using an expanded data set of 290 samples, achieving an accuracy of 0.890 and an AUC of 0.964. Analysis of the full receptor data set revealed distinct response patterns, and statistical screening with correlation filtering identified OR2W1, OR51E1, and OR51E2 as the most informative features. Using these selected receptors, a random forest classifier achieved an accuracy of 0.890 and an AUC of 0.964, demonstrating high sensitivity and specificity. OR response patterns showed a closer association with Gleason score than with serum PSA levels, indicating that urinary VOC signatures capture tumor-related metabolic information complementary to conventional biomarkers. These results demonstrate that an OR-ND sensor array coupled with machine learning enables accurate and noninvasive detection and classification of PCa from urine samples. Our study provides a modular framework extensible to other VOC-associated diseases.
23 May 2026
Read appraisal →Biomedical microdevices
Automated blood group classification using a digital microfluidics chip and vision transformer-based image analysis
Identifying blood groups accurately is critical for safe medical practices, especially in emergencies, surgeries, and prenatal care. Conventional methods often depend on visual inspection of agglutination reactions, which can be error-prone, particularly when using small sample volumes. In this work, we introduce an efficient and intelligent system for blood type detection that combines digital microfluidics with advanced deep learning and antigen-based decision logic. A Vision Transformer (ViT) model was trained to recognize agglutination patterns in droplet-based blood images, followed by an automated mapping of standard ABO/Rh rules to assign the corresponding blood group based on antigen presence. The proposed system achieved 100% accuracy specifically in detecting antigen-antibody agglutination reactions, with perfect scores across precision, recall, specificity, and F1-score. Our method reduces the need for large reagent volumes and minimizes testing cost, while also improving reliability. In future work, our aim is to embed this system into a compact, microfluidics paper-based device, enabling low-cost, AI-assisted blood typing in portable point of care settings.
21 May 2026
Read appraisal →Journal of clinical pathology
AI triage of duodenal biopsies improves workflow
AIMS: To develop, deploy and evaluate artificial intelligence (AI) for triaging duodenal biopsies within a National Health Service (NHS) histopathology laboratory, with the aim of improving reporting turnaround times for clinically significant diagnoses. METHODS: The pathway was developed in the UK in an NHS laboratory. Rule-based automation software was used to find all newly scanned duodenal biopsy slides. Those with case numbers ending in an odd number were exported for AI triage and if they had significant AI predicted abnormalities they were prioritised for reporting. The cases with even numbers followed the routine reporting pathway. RESULTS: 313 cases (517 duodenal slides) were processed by the routine pathway, and 329 cases (533 duodenal slides) were processed by the AI triage pathway. AI processing took about 70 s per slide. The AI classifier had a sensitivity and positive predictive value (PPV) as follows: normal small bowel: 99.6%, 95.7%; coeliac disease: 86.7%, 100%; gastric heterotopia: 84.6%, 95.7%; adenoma: 88.9%, 88.9%; adenocarcinoma: 50.0%, 100%. In the AI triage workstream coeliac disease, and non-neoplastic abnormalities as a group, were reported quicker than in the standard workstream (6 days vs 10 days and 7 days vs 10 days, respectively, both p<0.005), but neoplastic lesions were not reported quicker. The cost of deployment and operation was reasonable. CONCLUSIONS: An NHS histopathology laboratory successfully developed and implemented an AI-based triage system for duodenal biopsies, achieving high diagnostic accuracy and significantly improving turnaround times for coeliac disease and non-neoplastic abnormalities as a group. This study demonstrates the feasibility and clinical value of locally developed AI tools within routine diagnostic practice.
21 May 2026
Read appraisal →Genome medicine
DNA methylation biomarkers-based pan-cancer classifier: predictive modeling for cancer classification
BACKGROUND: Machine-learning (ML) driven molecular diagnostics based on omics data has a potential to revolutionize personalized medicine. However, implementation of ML into diagnostic protocols is hindered by methodological challenges which often lead to inflated performance assessment of models during development followed by poor performance of these models in implementation phase. Here, we aimed to develop and validate a pan-cancer classification framework based on DNA methylation data, that addresses methodological challenges of omics data powered ML. METHODS: We curated a primary dataset of DNA methylation profiles for 10756 samples, that included 54 healthy and cancer tissue types and validation dataset comprising data for 2306 samples from 28 independent studies. The classification framework was build using custom biomarkers selection strategy based on effect size metric that considers variance and class imbalance. The ML models were trained, tuned and evaluated using nested cross-validation approach. Local outlier factor algorithm was built into the inference pipelines to identify and filter samples displaying technical or biological anomalies. Additionally, for methodological validation of our framework we used methylation profiles for 3905 central nervous system (CNS) tumors. RESULTS: We found that relatively simple ML models outperformed complex algorithms such as deep neural network. A logistic regression classifier achieved a balanced accuracy (BACC) of 0.90 to classify 54 cancer and healthy tissue types using methylation levels at 1208 CpG sites. Similarly, our CNS tumor classifier also based on logistic regression algorithm reached a BACC of 0.94 across 59 CNS tumor subtypes. The anomaly filtering improved performance across all categories of samples tested. CONCLUSIONS: Our study demonstrates that DNA methylation profiling, when combined with carefully controlled ML practices allows for development of robust solutions that might substantially increase the efficacy of oncological diagnosis. Finally, we deployed our inference pipelines for public access via secure web platform - https://opp.pum.edu.pl/ .
21 May 2026
Read appraisal →Clinica chimica acta; international journal of clinical chemistry
Exploring clinical feasibility of zero-shot learning in a large language model for immunofixation electrophoresis image interpretation
INTRODUCTION: Immunofixation electrophoresis (IFE) is the gold standard method for detecting and typing monoclonal immunoglobulins, but its interpretation requires expert evaluation and remains susceptible to subjectivity and variability. Recent advances in multimodal large language models (LLMs) have enabled zero-shot visual reasoning without task-specific training. This study aimed to evaluate the feasibility and diagnostic performance of zero-shot multimodal LLMs for automated interpretation of IFE images. MATERIALS AND METHODS: A dataset of 487 immunofixation electrophoresis images representing ten semantic classes (AK, AL, GK, GL, K, L, LGL, MK, ML, and NONE) was retrospectively collected and annotated by expert laboratory specialists. Two multimodal LLMs, ChatGPT-5.2 and Gemini3 Pro, were evaluated using a zero-shot visual question answering (ZS-VQA) framework without additional training or fine-tuning. Each image was analyzed using two prompt configurations: a simple prompt and a detailed prompt with structured diagnostic guidance. Model performance was assessed using precision, recall, F1-score, and confusion matrix analysis. RESULTS: Gemini outperformed ChatGPT across most classes, particularly with detailed prompts, achieving high F1-scores in clinically relevant monoclonal categories such as MK (89.80), AK (84.75), and GK (71.54). Detailed prompting consistently improved recall and F1-scores, underscoring the importance of prompt design. In contrast, ChatGPT showed lower recall, F1-scores, and classification consistency. Both models performed less effectively in visually subtle patterns. CONCLUSIONS: Zero-shot multimodal LLMs, particularly Gemini, show promising potential for interpreting IFE images without task-specific training. However, performance variability and limitations in certain classes indicate that further optimization and validation are required before clinical implementation.
18 May 2026
Read appraisal →Talanta
SERS technology helps bacterial inactivation in vivo: including detection, identification, and evaluation.
The problem of bacterial infection and antimicrobial resistance has become a major challenge threatening to human health. Detection, identification, and real-time evaluation of bacteria during the sterilization process are of great significance for interpreting the sterilization mechanism and ensuring the effectiveness of sterilization. With the advantages of high sensitivity, fast detection speed, and real-time monitoring, Surface-enhanced Raman Scattering (SERS) technology provides a revolutionary solution to in vivo bacterial detection, identification, and antibacterial effect evaluation through functional substrate optimization, microfluidic chip integration, and deep collaboration of machine learning algorithms. This review focuses on the prominent applications of SERS technology in chemical, biological, and physical sterilization processes. By analyzing the changes in bacterial SERS fingerprint patterns, precise detection and identification of sensitive/resistant bacteria have been achieved, deepening the interpretation of drug resistance mechanisms at the gene, cell, and biofilm levels. Meanwhile, SERS technology provides an ideal tool for monitoring and evaluating the biological sterilization process. In the process of physical sterilization, enhancing the substrate material not only helps to improve SERS signals, but also promotes bacterial inactivation. By monitoring real-time changes in bacterial metabolism, the sterilization process can be precisely regulated. This review summarizes the innovative application of SERS technology in accurately guiding the sterilization process in vivo, providing important support for optimizing antibacterial treatment strategies.
18 May 2026
Read appraisal →Clinica chimica acta; international journal of clinical chemistry
Artificial intelligence algorithm optimization and application in patient-based real-time quality control (PBRTQC)
Patient-based real-time quality control (PBRTQC) serves as a vital supplement to quality management in clinical laboratories. Its core principle is to monitor the testing process in real time and continuously through patient test data. As artificial intelligence (AI) technology develops rapidly, AI has provided novel pathways for the innovation of PBRTQC algorithms and drives its transition from a traditional statistics-driven model to intelligent monitoring. This review systematically summarizes the progress of AI-driven PBRTQC algorithm optimization. Meanwhile, it provides a detailed account of the clinical applications of the AI-PBRTQC monitoring platform. These applications encompass timely quality control early warning, homogeneous monitoring across multiple settings, precise quality control in complex clinical settings, anomaly traceability and subsequent correction. In addition, this review offers an in-depth analysis of the challenges that arise during the practical implementation of AI-PBRTQC. These include technical limitations, shortage of professional talents, system compatibility barriers, and lagging standardization and regulation. It also explores future development trends and provides valuable references for the intelligent upgrade of PBRTQC.
16 May 2026
Read appraisal →PloS one
Analyzing the performance of deep learning splice prediction algorithms
SpliceAI is the leading tool for predicting splice-altering variants, but restrictive licensing limits clinical adoption. While open-source implementations have been published with author-reported comparisons, independent benchmarking across diverse datasets is needed to establish equivalence. We compared the original SpliceAI with two open-source implementations (OpenSpliceAI and CI-SpliceAI) and a legacy ensemble baseline across six datasets: a curated set of 1,316 validated variants, 213 variants with splice-assay data, 99,601 variants from the SPiP splicing prediction study, 242 manually curated deep intronic pathogenic variants, and two ClinVar-derived datasets comprising 53,600 intronic variants and 58,064 variants spanning all genomic contexts. The deep learning models were also evaluated against an ensemble of four legacy splice-prediction tools. Across all datasets, the deep learning algorithms outperformed the legacy ensemble. All three deep learning algorithms showed similar performance on the larger datasets dominated by canonical splice site variants (balanced accuracies 0.889-0.977). On the deep intronic benchmark, the original SpliceAI achieved the highest balanced accuracy (0.940), outperforming both CI-SpliceAI (0.890) and OpenSpliceAI (0.841). Critically, optimal thresholds for deep intronic variants were an order of magnitude lower than standard recommendations, indicating that default thresholds would miss the majority of pathogenic deep intronic variants. A correlation analysis showed that CI-SpliceAI maintained balanced concordance across event types, whereas OpenSpliceAI showed stronger correlation for loss events than gain events. Both implementations showed high positional agreement with SpliceAI, with exact splice-site match rates exceeding 90% across event types. Together, these results demonstrate that both open-source reimplementations of SpliceAI successfully reproduce the predictive behavior of the original algorithm across multiple evaluation contexts, while consistently outperforming traditional splice prediction methods. However, performance diverges on deeply intronic variants, and standard score thresholds are poorly calibrated for this variant class regardless of algorithm choice.
15 May 2026
Read appraisal →JMIR formative research
Automatic Speech Recognition and Large Language Models for Multilingual Pathology Report Generation: Proof-of-Concept Study
BACKGROUND: Accurate transcription of pathology gross examination dictation is important for clinical documentation, but multilingual dictation remains challenging in settings where clinicians mix Chinese and English while final pathology reports are written in English. OBJECTIVE: This study aimed to evaluate whether a Whisper-based automatic speech recognition (ASR) pipeline guided by contextual system messages and combined with open-source large language models (LLMs; Qwen2:72b, Llama3.1:70b, Gemma2:27b) could improve multilingual (Chinese-English) pathology dictation transcription accuracy and generate clinically appropriate English gross description reports. METHODS: We conducted a controlled proof-of-concept study using 125 simulated mixed Chinese-English pathology gross examination audio recordings created by physicians or pathologists. Audio recordings were transcribed using Whisper ASR with and without a contextual system message. The ASR transcripts were then converted into English gross description reports using 3 open-source LLMs: Qwen2:72b, Llama3.1:70b, and Gemma2:27b. Outcomes included character error rate, Bilingual Evaluation Understudy, Recall-Oriented Understudy for Gisting Evaluation (ROUGE)-1, ROUGE-2, ROUGE-L, Metric for Evaluation of Translation with Explicit Ordering, pathologist Win-Tie-Lose rankings, report-level error categories, inference time, and interrater agreement. RESULTS: The ASR contextual system message reduced the mean character error rate from 0.344 (SD 0.176; 95% CI 0.313-0.375) to 0.066 (SD 0.100; 95% CI 0.048-0.084; P<.001). Qwen2:72b achieved the highest automated metric scores, including a Bilingual Evaluation Understudy of 0.644 (SD 0.307), ROUGE-1 of 0.866 (SD 0.163), ROUGE-2 of 0.771 (SD 0.235), ROUGE-L of 0.842 (SD 0.178), and Metric for Evaluation of Translation with Explicit Ordering of 0.805 (SD 0.214). Pathologist-coded total error rates were 16.8% (21/125) for Qwen2:72b, 45.6% (57/125) for Llama3.1:70b, and 92.8% (116/125) for Gemma2:27b. The exact agreement between the 2 pathologists across full ranking categories was 76.8% (96/125; Cohen κ=0.668), and agreement on the top-ranked model or tied top group was 81.6% (102/125; Cohen κ=0.722). CONCLUSIONS: In this proof-of-concept evaluation, contextual prompting improved ASR transcription accuracy, and Qwen2:72b generated the most accurate English pathology reports among the evaluated LLMs. However, the study used simulated audio recordings, a local vocabulary prompt, and report-level rather than term-level clinical annotation. LLM-generated reports should therefore be considered draft documentation requiring pathologist verification, and prospective validation in real clinical workflows is needed before clinical deployment.
15 May 2026
Read appraisal →Journal of clinical microbiology
Diagnostic performance of β-(1→3)-D-glucan, two Candida antigen, and five anti-Candida antibody assays in ICU patients with sepsis and high risk for invasive candidiasis: a secondary endpoint of the CandiSep randomized clinical trial
Invasive Candida infection (ICI) is the most common fungal infection in critically ill patients. This study analyzed the performance of various biomarkers in sera from the CandiSep trial, a randomized, multicenter trial including 342 sepsis patients at high risk for ICI across 18 German intensive care units. ICI and candidemia were diagnosed in 48 (14.0%) and 14 (4.1%) patients, respectively. Sera collected on 2 consecutive days at sepsis onset were analyzed for β-(1→3)-D-glucan (BDG; Fungitell), mannan (Platelia-Candida-Ag-Plus [Platelia-Mn] and Serion-ELISA-antigen-Candida [Serion-Mn]), and anti-Candida antibodies (Platelia-Candida-Ab-Plus, Serion-ELISA-Candida albicans-IgA/IgM/IgG, and Virclia-Candida albicans-germ-tube-antibody-IgG-Monotest). Only antigen levels (BDG, Platelia-Mn, Serion-Mn), but not anti-Candida antibody levels, were significantly elevated in ICI patients. Antigen levels were unaffected by Candida colonization, while antibody levels were significantly increased. Sensitivity and specificity at the manufacturer's cutoffs were unsatisfactory. Performance improved by adjusting cutoffs: BDG required a 3-fold increase, while all others required lowering. At 80% specificity, sensitivities (95% confidence intervals) for the diagnosis of ICI and candidemia were as follows: BDG (cutoff >280 pg/mL), 46% (31.4-60.8) and 64% (35.1-87.2); Platelia-Mn (cutoff >50 pg/mL), 38% (24.0-52.6) and 64% (35.1-87.2); Serion-Mn (cutoff >0.7 U/mL), 38% (24.0-52.6) and 50% (23.0-77.0), and below 28% and 43% for all antibody assays. Area under the receiver operating characteristic curve comparisons showed no significant differences between antigen assays. Combining biomarkers, either simultaneously or sequentially, offered no diagnostic advantage over single-biomarker use. In conclusion, anti-Candida antibody assays are likely influenced by colonization and have limited diagnostic utility. Antigen assays offer similar diagnostic value but require cutoff optimization. Combining biomarkers did not enhance diagnostic yield in our cohort.IMPORTANCEOur main findings based on our specific ICU cohort were as follows: (i) only Candida antigen levels, not anti-Candida antibody levels, were significantly elevated in ICI and candidemia. (ii) In patients without ICI, Candida colonization was not associated with altered antigen levels, but with elevated antibody levels. (iii) Sensitivity and specificity at the manufacturer's cutoffs were unsatisfactory, and cutoffs should be significantly adjusted. Potential optimal cutoff values are proposed by us. (iv) The evaluated antigen assays demonstrated overall comparable diagnostic performance in this study. (v) Anti-Candida antibody assays did not provide a meaningful diagnostic contribution. (vi) The combination of biomarkers offered no diagnostic advantage over the use of individual biomarkers, neither simultaneously nor sequentially. Our results suggest that the use of a single Candida antigen test with a cohort-adapted cutoff value may be sufficient. Furthermore, avoiding biomarker combinations could potentially reduce healthcare costs. The clinical implications of these findings should be interpreted with caution, given the limited data with associated large confidence intervals for candidemia and the cohort-specific nature of the study. Our results should therefore be confirmed in larger studies and additionally with other patient groups.
15 May 2026
Read appraisal →Journal of clinical microbiology
Prediction of KPC-producing Klebsiella pneumoniae by MALDI-TOF MS, ensemble learning, and spectral peak annotation
The increasing prevalence of carbapenem-resistant Klebsiella pneumoniae represents a serious global health challenge. Rapid and accurate detection methods are essential to inform early and correct use of antimicrobial therapy to idealize patient outcomes and limit the spread of resistant isolates. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has been evaluated for resistance detection, but its clinical application remains limited. In this study, MALDI-TOF MS was integrated with machine learning (ML) to improve detection of Klebsiella pneumoniae carbapenemase (KPC) among K. pneumoniae isolates harboring KPC carbapenemase. Ensemble learning strategies were applied to 435 clinical isolates from different geographic locations (Spain, United States, South America, and Asia), with validation performed using external data sets. By constructing two peak matrices and applying four ML algorithms, as well as combinations of two and three in ensemble models, 92 different classifiers were tested. Ensemble combinations increased the specificity of classifiers to over 95%, while sensitivity reached 72%, being significantly higher than that of the MALDI Biotyper KPC module. Low sensitivity may be affected by technical variability during spectral acquisition. The first annotated MALDI-TOF MS spectrum for K. pneumoniae by in silico prediction of protein masses was developed to enable peak identification, including peaks potentially related to antimicrobial resistance. Overall, the results of this study show that combining MALDI-TOF MS with ensemble learning can enhance KPC detection performance.IMPORTANCERapid and accurate detection of Klebsiella pneumoniae carbapenemase (KPC)-producing Klebsiella pneumoniae informs early and correct use of antimicrobial therapy to idealize patient outcomes and limit the spread of antimicrobial resistance. In this study, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) and ensemble machine learning were employed to detect KPC-producing isolates, using diverse isolate collections, obtaining excellent specificity. An annotated K. pneumoniae MALDI-TOF MS profile is provided with peak annotation, providing a resource for peak analysis.
14 May 2026
Read appraisal →Journal of chromatography. A
Microfluidics in separation science: Fabrication, applications and intelligent systems
The emergence of advanced, innovative microfluidic devices has reshaped analytical separation methodologies and biomolecular diagnostics by introducing highly integrated, miniaturized systems that enable next-generation fluid manipulation, enhanced automation, precision, cost efficiency, and high-throughput capabilities. Advances in fabrication methods, such as soft lithography, photolithography, and 3D printing, have enabled the production of customizable microfluidic platforms from materials such as PDMS, glass, and thermoplastics. These microfluidic systems support a range of analytical functionalities of the technologies, including gas chromatography (GC), high-performance liquid chromatography (HPLC), and electrophoresis, delivering rapid, high-resolution analyses with minimal sample and reagent requirements, quicker analysis, and enhanced uniformity and stability, facilitating high productivity screening, improved sensitivity and portability. Integration of micro-GC systems enables rapid, portable detection of volatile compounds, while microfluidic HPLC and electrophoresis provide precise biomolecular separations and real-time diagnostics. Recent advancements in artificial intelligence (AI) have now brought significant innovations in microfluidic devices, resulting in a new era of intelligent analytical separations, where machine learning algorithms and neural networks are applicable not only for data processing and pattern recognition but also for real-time optimization, anomaly detection, innovative read-out systems, and autonomous system control. The fusion of AI and microfluidics has enabled new approaches to dynamic experimentation, reshaping traditional microfluidics and personalized diagnostics and health care systems, where rapid, reliable analysis is paramount. In this review article, we comprehensively explain the progressive convergence of microfluidics and AI across various electrophoretic and chromatographic analytical separation techniques, elaborating on the latest technological innovations, fundamental fabrication strategies, and emergent applications. The article also presents a forward-looking perspective on how AI-enabled microfluidics enhances automation, scalability, and application-specific customization, which could shape the future of intelligent analytical separation systems.
12 May 2026
Read appraisal →Analytical chemistry
Quantitative Analysis of Multiple Serum Tumor Biomarkers by an Interpretable Stacked Ensemble Model
Surface-enhanced Raman spectroscopy (SERS) offers exceptional sensitivity and specificity for biomolecular detection, particularly in analyzing serum tumor biomarkers. However, its application is hindered by spectral complexity and matrix interference. Although machine learning (ML) has shown promise in SERS data analysis, existing quantitative models for biomarker detection lack generalizability and interpretability. To address these challenges, we developed an interpretable stacked ensemble model (ISEM) that integrates feature selection, ensemble learning, and explainable artificial intelligence (XAI) for quantifying 12 serum tumor biomarkers. To ensure reliable quantitative analysis, we validated the acquired SERS data using uniform and reproducible Au nanoparticle substrates. Preprocessing, least absolute shrinkage, and selection operator (LASSO) feature selection were employed to establish a foundation for accurate quantification. Subsequently, 12 serum tumor biomarkers were quantified, demonstrating the superior performance of ISEM, which achieved a high predictive accuracy, with R2 values exceeding 0.9 for all biomarkers. Crucially, based on the quantitative results, we provided molecular-level interpretability for structure-spectrum correlations using Shapley additive explanations (SHAP), revealing how glycosylation reactions, matrix interference, and spectral overlap influence prediction accuracy. Furthermore, the capability of ISEM to quantify multiple biomarkers in unseen samples was confirmed through validation of its generalizability. Our study establishes an ensemble-driven, interpretable framework for quantitative biomarker analysis in a complex biological matrix, demonstrating significant potential for early cancer diagnosis and screening.
11 May 2026
Read appraisal →Analytical chemistry
Precise Artificial Intelligence Nanomotor Tracking-Powered Amplification-Free Digital MicroRNA Sensing
Digital assays have become a state-of-the-art technology for the sensing of nucleic acid molecules. Nevertheless, most of the prevalent digital nucleic acid assays heavily rely on sealed microchambers and sophisticated nucleic acid amplification strategies to achieve an end-point digital signal readout. Herein, we proposed an artificial intelligence (AI)-facilitated, digital fluorescence nanomotor tracking-powered assay (AI-dFNA) for one-step, amplification-free, and highly sensitive detection of microRNAs (miRNAs). In this design, even a single target miRNA-templated amplification-free ligation can introduce a sufficient driving force on the surface of a fully open fluorescence nanoparticle to cause remarkable motion behaviors. With the assistance of a self-developed AI algorithm, clear discrimination of positive and negative fluorescent nanomotors can be achieved, enabling real-time digital miRNA sensing. The AI-dFNA avoids the use of enclosed microchambers and intricate signal amplification techniques, which could even identify varying fluorescence colors of nanomotors to allow for the analysis of different kinds of miRNA molecules, pioneering a new direction toward next-generation digital nucleic acid sensing.
11 May 2026
Read appraisal →Annals of laboratory medicine
Evaluation of the Performance of Advanced Large Language Models in Laboratory Medicine Using Residency Examinations
BACKGROUND: Recent advancements in large language models (LLMs) have accelerated their integration into clinical domains, including laboratory medicine. The performance of LLMs in answering board-level laboratory medicine questions has not been comprehensively evaluated. Given the importance of diagnostic accuracy in this field, rigorous and objective evaluations of LLM capabilities are essential. METHODS: We assessed 12 LLMs from OpenAI, Anthropic, and Google using 320 Korean Residency Examination questions (2021-2024) spanning six laboratory medicine subspecialties. Standardized prompts were provided via their application programming interfaces under deterministic settings (temperature=0). Questions were administered thrice to assess response reproducibility. Outputs were compared with validated answers and analyzed for accuracy, reasoning quality, and error typology. RESULTS: Google's Gemini 2.0 Pro achieved the highest accuracy (80.0%), followed by OpenAI's GPT-4.5 (77.2%) and Anthropic's Claude 3.7 Sonnet (74.1%). Accuracy decreased as the difficulty of questions increased (78.0% for easy vs. 45.1% for challenging). Subspecialty performance varied. Al models underperformed on questions on transfusion medicine (mean accuracy: 38.8%), primarily because of limitations in domain-specific and regional knowledge representations. Incorrect answers primarily resulted from reasoning errors. Reproducibility exceeded 95% for most models; however, some residual non-determinism appeared even with greedy decoding (temperature=0). CONCLUSIONS: LLMs demonstrated substantial potential for integration into laboratory medicine, particularly in clinical chemistry and immunology. Performance inconsistencies (particularly for high-difficulty questions) and knowledge gaps (notably for transfusion medicine) highlight the necessity for further development-potentially including domain-specific fine-tuning and retrieval-augmented generation integration-and robust expert oversight before clinical application.
6 May 2026
Read appraisal →Microscopy research and technique
Deep Learning Integration in Optical Microscopy: Advancements and Applications
Optical microscopy is a cornerstone imaging technique in biomedical research, enabling visualization of subcellular structures beyond the resolution limit of the human eye. However, conventional optical microscopy faces challenges such as optical aberrations, diffraction-limited resolution, low signal-to-noise ratio (SNR), and poor contrast. The exponential growth of bioimaging data further underscores the need for advanced computational tools. Deep learning (DL) is a subset of machine learning that has emerged as a transformative approach to address these limitations, offering enhanced precision, reduced manual intervention, and diminished reliance on domain-specific expertise for image reconstruction, enhancement, and analysis. This review explores the integration of DL into optical microscopy, focusing on key applications including image classification, segmentation, and computational reconstruction. We examine prominent DL architectures such as convolutional neural networks (CNNs), U-Nets, residual networks (ResNets), and generative adversarial networks (GANs)-and their role in advancing diverse microscopy modalities. These frameworks enhance image quality, improve quantitative analysis, and democratize access to high-performance microscopy. Additionally, we discuss persisting challenges, including the demand for large, annotated datasets, dynamic sample variability, model interpretability, and potential data biases. Collectively, DL is poised to revolutionize optical microscopy, shaping its future developments in biomedical imaging.
6 May 2026
Read appraisal →Cytopathology : official journal of the British Society for Clinical Cytology
Comparative Evaluation of Transfer Learning Models for Detecting Malignant Cells in Urinary Cytology
AIMS AND OBJECTIVES: In the present paper, we compared the efficiency of six transfer learning models to detect malignant cells in urine cytology. We also applied an ensemble learning with weighted soft voting to assess its importance in the diagnostic accuracy in urine cytology. MATERIALS AND METHODS: The voided urine samples of total 104 cases of urothelial cell carcinoma (UCC) and 86 cases with no malignancy (benign) were selected. All the 104 cases of UCC were histopathology proven high grade urothelial cell carcinoma (UCC). Urine with negative cytology reports were followed up clinically. There were 446 images of benign samples and 1369 images from malignant samples on 100 x objective. We applied six transfer learning models (DenseNet121, inception_v3, ResNet50, MobileNetV2, VGG16 and Xception) to detect malignant cells in urine. To compare the performance of different models, dynamic training optimization was done and each model was auto stopped after their maximum performance. In addition, an ensemble learning with soft voting was used including the top three models to enhance the diagnostic accuracy. RESULT: Xception transfer model showed the highest sensitivity (88.57%), accuracy (86.55%), precision (80.52%) and FI score (84.35%). It was the best performing model. The other two top performing models were InceptionV3 and ResNet50. The area under curve of receiver operating characteristic (AUCROC) was ≤ 90 in all the transfer learning models. The accuracy, sensitivity, specificity, precision, F1-Score and AUCROC of the ensemble transfer learning model were as follows: 92.10%, 95.41%, 85.51%, 91.23%, 93.24% and 0.977 respectively. CONCLUSIONS: First time, we evaluated a large number of transfer learning models in urine cytology to detect malignant cells. All the models showed high sensitivity, specificity and accuracy. In addition, the ensemble learning technique with soft voting showed remarkable superior performance than individual top three transfer learning models. The techniques transfer learning and ensemble models have high potential to use in routine screening of urine.
6 May 2026
Read appraisal →Nano letters
Seeing and Feeling DNA Methylation: Single-Molecule Biophysics Meets Machine Learning
DNA methylation at 5-methylcytosine (5mC) is crucial for embryonic development and cellular function, while aberrant patterns strongly drive disease onset and progression. Its reversible nature offers substantial therapeutic potential, emphasizing the need for precise, context-specific genome wide 5mC mapping. Conventional techniques such as bisulfite sequencing and ensemble biosensor assays are hindered by DNA degradation, amplification bias, high cost, and inability to resolve single-molecule structural and mechanical effects of methylation. This review examines advances in single-molecule biophysical methods (nanopore sensing, smFRET, optical/magnetic tweezers, and AFM) that provide direct, label-free/minimally invasive 5mC detection, along with quantitative insights into DNA conformation, mechanics, and protein-DNA interactions. These techniques complement traditional methylome mapping by linking genomic localization to molecular mechanisms. Emerging machine-learning approaches are revolutionizing analysis, particularly in nanopore sensing, while promising applications in smFRET, tweezers, and AFM address throughput and reproducibility challenges. Their convergence promises scalable, high-resolution epigenetic profiling, advancing precision epigenomics toward clinical application.
4 May 2026
Read appraisal →Lab on a chip
MaGIC-OT: an AI-guided optical tweezers platform for autonomous single-cell isolation in microfluidic devices
Automating the isolation of rare cells such as circulating tumour cells (CTCs) within crowded microfluidic environments remains a bottleneck in liquid biopsy workflows. Optical tweezers offer contact-free, selective manipulation but traditionally rely on expert operators. We present MaGIC-OT (machine-guided isolation of cells using optical tweezers), a platform that integrates classical path planning and deep reinforcement learning (DRL) to automate single-cell manipulation inside a microfluidic chip. We built a high-fidelity simulation to train and benchmark control policies and show that cooperative, human-in-the-loop training improves DRL performance. Trained agents outperform expert users in speed and isolation success in silico, and we demonstrate proof-of-concept isolation of a cancer cell from a spiked blood sample on-chip. MaGIC-OT provides a flexible framework for intelligent optical manipulation, aligning microfluidic device design with autonomous control strategies and offering a pathway toward high-purity, label-free single-cell workflows.
26 Apr 2026
Read appraisal →Lab on a chip
Machine learning-driven single-cell phenotyping in size-controlled microenvironments via parallel deterministic droplet microfluidics
Understanding how individual cells respond to distinct physical microenvironments is critical for mechanobiology, cell therapy, and tissue engineering. Current single-cell encapsulation methods are often limited by Poisson loading and fixed droplet sizes, preventing parallel generation of multiple, size-specific microenvironments and constraining high-resolution phenotypic analyses. Here, we present a droplet microfluidic platform that enables deterministic single-cell encapsulation within microgels of multiple sizes from a single precursor stream, achieved through parallelized flow-focusing combined with cell-selective gelation. This system produces distinct microgel size regimes simultaneously, minimizing empty compartments and enabling direct, side-by-side comparisons of cellular behavior under controlled yet variable confinement. Using machine learning to analyze 3D morphological and cytoskeletal features, we reveal heterogeneous, size-dependent phenotypic responses and demonstrate that cellular phenotypes alone can predict microgel confinement across time. Together, these results establish a data-driven framework for mapping single-cell responses across engineered microenvironments and provide a scalable platform for predictive studies of mechanosensitive behavior in heterogeneous niches.
24 Apr 2026
Read appraisal →mSystems
Short-chain fatty acid-producing microbes differentiate non-infectious and infectious neutropenic fever in leukemia.
UNLABELLED: Neutropenic fever (NF) is often the first sign of infection in patients with hematologic malignancies, but its cause is frequently unknown, leading to broad-spectrum antibiotic use without confirmed infections. Although research links gut microbiome disruptions to treatment-related infections, it typically examines NF as the outcome, leaving a gap in understanding how the microbiome and metabolic factors distinguish infectious from non-infectious cases. Stool samples from acute myeloid leukemia patients were analyzed to characterize gut microbiome composition and fecal metabolites at baseline and at fever onset. Machine learning models, network analyses, and functional profiling were used to differentiate infectious NF vs non-infectious NF at baseline and at fever onset. The baseline model (area under the receiver operating characteristic [AUROC] = 0.769) identified higher levels of Akkermansia, Enterobacter, Escherichia-Shigella, and Flavonifractor as predictors of infectious NF, while Collinsella, Lachnospiraceae, Coprococcus, and acetic acid were linked to non-infectious cases. At fever onset, Enterococcus was enriched in infectious NF, whereas Gemella, Butyrate, Lachnospiraceae, Ruminococcaceae, and Eisenbergiella abundances favored non-infectious NF outcomes (AUROC = 0.752). Network analyses also revealed greater functional diversity and microbiome-metabolome connectivity in non-infectious cases at fever onset. This study suggests that gut microbiota and metabolites may serve as biomarkers for distinguishing infectious from non-infectious neutropenic fever, warranting further validation in larger cohorts. IMPORTANCE: Our study tackles the challenge of managing neutropenic fever (NF) in immunocompromised patients whose numbers have increased due to various immunodeficiencies and treatments that suppress immune function. Fever is often the only sign of a serious infection in these patients, yet there are neither clear patterns linking risk factors to infection nor biomarkers reliable for ruling out non-infectious causes. As a result, febrile patients are typically empirically treated for major pathogens, even in the absence of confirmed infections, which propagates antimicrobial resistance and gut dysbiosis. Our research utilizes gut microbiome and targeted metabolomic profiling from two cohorts of patients with acute myeloid leukemia undergoing chemotherapy and employs a machine learning framework to distinguish between infectious and non-infectious NFs at baseline and upon fever onset.
23 Apr 2026
Read appraisal →Medicine
Integration of the ISO 15189 quality management system in the undergraduate internship teaching of medical laboratory technology
BACKGROUND: The International Organization for Standardization (ISO) 15189 standard is essential for quality management in clinical laboratories, yet its integration into undergraduate medical laboratory technology education remains limited. This study aimed to evaluate the impact of a simulated laboratory teaching model based on the ISO 15189 standard on the quality management competencies of undergraduate interns and to compare it with traditional teaching methods. METHODS: A parallel-group, randomized controlled trial was conducted in the Department of Clinical Laboratory at the Second People's Hospital of Hunan Province (Brain Hospital of Hunan Province). A total of 60 undergraduate medical laboratory technology students scheduled to commence their clinical clerkship between July 2023 and May 2025 were enrolled and randomly allocated to an intervention group (n = 30) or a control group (n = 30). The intervention group received simulated laboratory teaching based on the ISO 15189 standard, including the construction of a quality management system and full-process simulated accreditation practice. The control group received traditional clinical clerkship teaching. Primary outcome measures were comprehensive assessment scores at the end of the internship (covering pre-examination, intra-examination, and post-examination quality control knowledge) and problem-solving ability scores in quality management roles. Secondary outcome measures included understanding of the quality management system, quality improvement awareness, and teaching satisfaction. RESULTS: Compared with the control group, the intervention group showed significantly higher scores in pre-examination quality control, intra-examination quality control, post-examination quality control, understanding of the quality management system, quality improvement awareness, and problem-solving ability in quality management roles (P < .01). Teaching feedback questionnaires also indicated higher satisfaction with the teaching model in the intervention group. CONCLUSION: For undergraduate medical laboratory technology interns, the simulated laboratory teaching model based on the ISO 15189 standard is significantly more effective than traditional teaching methods in enhancing their knowledge, skills, and competency in quality management.
19 Apr 2026
Read appraisal →Platelets
Neural network reveals platelet age from fluorescence microscopy images
Platelets are small, anucleate cells with a primary physiological role in vascular damage repair (hemostasis) and initiation of thrombus formation in response to vascular injury. Platelets circulate approximately 7-10 days, slowly undergoing age-related changes in molecular composition, morphology, activation capacity, function, and surface receptor density. As older platelets are associated with poor clinical outcome, no in vitro tests are available to predict platelet age, or to determine the fitness of platelet transfusion products. In this study, we developed a convolutional neural network model that could determine platelets' chronological age from confocal microscopic images. The model was trained using platelets stored in platelet-rich plasma up to 8 hours and using routine platelet concentrates up to 10 days. The model predicted chronological age of stored platelets with >97% accuracy. To test our model in vivo, we analyzed a cohort of patients with acute myeloid leukemia, experiencing thrombocytopenia due to chemotherapy. Our model could reliably distinguish in vivo between samples with younger and older platelets during the course of treatment. This study demonstrates the ability to predict platelets' chronological age both in vitro during storage and in vivo, which may impact clinical transfusion medicine and the diagnosis and treatment of patients with platelet disorders. What is the context?Platelets circulate approximately 7–10 days, slowly undergoing age-related changes in molecular composition, morphology, activation capacity, function, and surface receptor density. During storage, proinflammatory changes occur in platelets over time and platelet age may be associated with poorer clinical outcomes. Therefore, understanding platelet aging and developing novel tools to interrogate platelet age will be useful. However, no platelet tests are currently available that can predict platelet age accurately based on platelet characteristics without keeping record of the (storage) time.What is new?We developed a convolutional neural network model that could determine platelets’ chronological age from confocal microscopic images. The model was trained using platelets stored in platelet-rich plasma up to 8 hours and using routine platelet concentrates up to 10 days. The model predicted chronological age of stored platelets with very high accuracy. Furthermore, our model was in vivo tested in a cohort of patients with acute myeloid leukemia, experiencing thrombocytopenia due to chemotherapy. Our model could reliably distinguish between in vivo samples with younger and older platelets during the course of treatment.What is the impact?Our study demonstrates the utility of a convolutional neural network model using confocal images of platelets to determine storage time-associated platelet phenotype in vitro as well as physiologic platelets’ chronological age in vivo. This model may be useful in several clinical settings, including clinical transfusion medicine, testing the quality of stored platelet concentrates, and diagnosing and treatment of patients with platelet disorders, thrombocytopenia, or thrombosis.
18 Apr 2026
Read appraisal →Computers in biology and medicine
A systematic review and benchmarking of modern metagenomic tools for taxonomic classification
Advancements in metagenomics have been driven by the continuous development of bioinformatic tools, particularly taxonomic classification software, which are central to the accurate characterization of microbial communities. However, establishing direct comparisons between these tools remains challenging due to variations in evaluation metrics, reference databases, and input data types. In this study, we present a systematic review of recently developed metagenomic taxonomic classification tools. Of the 31 identified tools, nine satisfied all functional and methodological criteria for the benchmark analysis. We evaluated their accuracy and computational performance using a standardized dataset derived from the NCBI RefSeq database. Our analysis revealed that most of these tools are domain-specific, each excelling in particular areas. Tools like TAMA, CAMITAX and PhyloFlash achieved higher accuracy for prokaryotic organisms, while ViWrap and PhaBOX achieved higher accuracy for viral classifications. SqueezeMeta achieved high F1 scores across most domains, though its assembly-based architecture limits effectiveness on highly diverse samples. MegaPath-Nano was least affected by increased mutation rates. The execution time varied widely among the tools, with domain-specific and machine learning-based tools generally being faster, while tools like BASTA had longer runtimes and lower accuracy. This review synthesizes performance results for current tools, providing an overview of their strengths and computational methodologies.
17 Apr 2026
Read appraisal →Journal of clinical microbiology
Identification of Aspergillus at section and species levels by artificial intelligence-based microscopic morphology image recognition
Rapid and accurate identification of Aspergillus species in clinical microbiology laboratories is crucial for aspergillosis diagnosis and antifungal therapy. However, traditional methods still face challenges in distinguishing phylogenetically related species due to their morphological similarities. This study presents FungalNet, a deep learning model integrating ResNet-50 architecture with Focal Loss algorithm, specifically designed to enhance feature extraction for Aspergillus identification. A total of 12,000 high-resolution images were obtained from lactophenol cotton blue-stained slide preparations under a 100× oil immersion objective, among which 311 images were excluded through a novel quality control approach combining fivefold cross-validation and expert manual review. The performance of four deep learning models (FungalNet and three established models) for identifying Aspergillus species and sections was evaluated using the remaining 11,689 qualified images. FungalNet demonstrated superior classification performance, achieving overall accuracies of 98.45% and 97.85% at the section and species levels, respectively. These results indicate that FungalNet shows significant promise for rapid and accurate identification of Aspergillus species. With further optimization and multicenter validation, this tool could potentially be integrated into routine diagnostic workflows to enhance the efficiency and reliability of fungal identification in clinical settings.IMPORTANCEThis study integrates microscopic morphology identification with deep learning to address the challenge of accurate Aspergillus species identification. Twelve clinically isolated Aspergillus species belonging to eight different sections were included. From touch-tape slide preparations with lactophenol cotton blue staining under a 100× oil immersion objective, 11,689 qualified images were collected and analyzed using FungalNet (our proposed model) along with three established models (GoogLeNet, ResNet-50, and Xception). The results showed that FungalNet demonstrated superior performance in Aspergillus identification, achieving the highest classification accuracy at both section (98.45%) and species (97.85%) levels. Given its rapid turnaround time and cost-effectiveness, this AI-based image analysis approach shows promising potential for the rapid and accurate identification of Aspergillus species in clinical microbiology laboratories.
9 Apr 2026
Read appraisal →Journal of the American Heart Association
Development and Validation of a Machine Learning Model to Predict Oral Anticoagulant Use in Stroke From Prothrombin Time-International Normalized Ratio and Activated Partial Thromboplastin Time
BACKGROUND: Timely recognition of oral anticoagulant use is critical in acute stroke but is often hampered by impaired consciousness and unavailable medication history. We investigated whether routinely available coagulation tests, prothrombin time-international normalized ratio and activated partial thromboplastin time, paired with machine learning can identify anticoagulant exposure at presentation. METHODS: We performed a multicenter diagnostic modeling study using the Japan Stroke Data Bank. Adults with ischemic or hemorrhagic stroke admitted in 2016 to 2021 comprised the derivation cohort; those in 2022 formed a temporally distinct validation cohort. Three models (random forest, extreme gradient boosting, and neural network) used prothrombin time-international normalized ratio and activated partial thromboplastin time as predictors to classify anticoagulant status: vitamin K antagonist, direct oral anticoagulant, or none. Discrimination (area under the receiver operating characteristic curve and area under the precision-recall curve), calibration, and decision curve analysis were assessed. Model outputs were implemented as bedside probability heat maps. RESULTS: Among 30 767 patients (derivation n=24 857; validation n=5910), vitamin K antagonist/direct oral anticoagulant use was 4.5%/8.2% in derivation and 2.8%/9.5% in validation. In the validation cohort, the neural network yielded the best performance: area under the receiver operating characteristic curve, 0.933 (95% CI, 0.908-0.955) and area under the precision-recall curve, 0.52 (95% CI, 0.44-0.61) for vitamin K antagonist; area under the receiver operating characteristic curve, 0.841 (95% CI, 0.822-0.858) and area under the precision-recall curve, 0.39 (95% CI, 0.35-0.43) for direct oral anticoagulant, with preserved calibration and highest net benefit across clinically relevant thresholds. Heat maps summarized class probabilities over the prothrombin time-international normalized ratio×activated partial thromboplastin time space to support rapid inference. CONCLUSIONS: A simple machine learning model using only prothrombin time-international normalized ratio and activated partial thromboplastin time classified vitamin K antagonist and direct oral anticoagulant exposure with high accuracy and practical clinical utility. This tool assists emergency decision-making when medication history is unknown, supporting the safety and timeliness of acute stroke care.
8 Apr 2026
Read appraisal →Analytical chemistry
Multicomponent Simultaneous Identification Network (MSINet): An Advanced Deep Learning Model for Boosting Multiplex SERS Detection in Untreated Real Samples
Surface-enhanced Raman spectroscopy (SERS) represents unique advantages for on-site analysis due to its high sensitivity, high specificity, and applicability to gaseous, liquid, and solid samples. However, the complex and unknown components in matrices, as well as competitive adsorptions between targets, will cause signal fluctuation and spectral overlap, which make it challenging to detect multiple targets in untreated real samples. Herein, we designed a deep learning model named the multicomponent simultaneous identification network (MSINet) for analyzing the SERS spectra of untreated samples and output qualitative and content evaluation results of multiple inside targets. The hierarchical feature extraction and multiple feature integration made MSINet robust to signal fluctuations and spectral overlaps while requiring only a small amount of training to quantify target contents in any samples due to its threshold judgment rule. Using the MSINet-assisted SERS detection scheme, the presence and content levels of multiple UV absorbers in seawater, food additives in ready-to-drink cocktails, and biomarkers in human serum could all be determined with ≥0.90 accuracy when facing testing databases. But in contrast, the accuracy in evaluating the target contents in the above untreated samples by the standard curve method was as low as 0.05. The powerful ability to process spectral data, an easy-to-use graphical user interface, and "zero" extra cost make MSINet a universal tool to assist arbitrary SERS sensors in performing high-throughput detection of real samples, leading to great potential in on-site analysis.
8 Apr 2026
Read appraisal →Clinical chemistry
Discovery of an Artificial Intelligence Label Feedback Loop: How the Success of a Clinically Implemented Artificial Intelligence Algorithm Has Created an Unforeseen Challenge to Algorithm Retraining
BACKGROUND: Artificial intelligence (AI) augmented laboratory tests can improve quality, efficiency, cost-effectiveness, and staff satisfaction. However, the clinical success of these tests can introduce unforeseen challenges for model retraining. This study describes the discovery of an "AI label feedback loop" in a clinically implemented AI-augmented kidney stone composition test. METHODS: An AI-augmented kidney stone composition test (V1) has been previously deployed for clinical kidney stone characterization. After several years of clinical use, a retrained model (V2) was developed using 6 times more data. Model performance of both V1 and V2 were evaluated across 3 datasets: a recent production validation (hold-out) set (mostly AI-influenced labels), the original V1 validation set (pre-AI, entirely human-labeled), and a subset of recent cases with exclusively human-generated or human-corrected labels. RESULTS: V2 demonstrated a 10% lower concordance rate than V1 when evaluated on the recent production hold-out set, despite a much larger training dataset. Performance between V1 and V2 was similar when applied to the pre-AI validation set. Notably, V2 outperformed V1 on the recent subset of cases with human-only or human-corrected labels, particularly for less-common stone types. These findings revealed an AI label feedback loop, confounding retraining and evaluation. CONCLUSIONS: The integration of AI into clinical practice can potentially influence reported test results, complicating the development and evaluation of future models. To mitigate AI label feedback loops, ongoing human annotation and careful validation set construction are essential. These strategies can ensure reliable performance assessment and support the safe evolution of clinical AI systems.
4 Apr 2026
Read appraisal →Clinical chemistry
Towards Clinical Integration of Deep Learning-Based Classification of Urinary Sediment Particles from Digital Microscopy Images: A Prospective Study
BACKGROUND: Urinalysis is a standard clinical test that includes the microscopic examination of urinary sediment to identify formed elements. Manual evaluation by laboratory technicians is time-intensive and subject to human error. Automated analysis using digital microscopy images presents a potential alternative. This study evaluates the integration of a deep learning approach to automatically classify urinary sediment images in the clinical laboratory, including independent prospective validation of its performance. METHODS: An annotated data set comprising 13 classes of urinary sediment elements was created from a database of Sysmex UD-10 digital microscope images. An EfficientNet-based model was trained and tested across three experimental scenarios to evaluate the effects of data collection strategies on performance. Uncertainty calibration was examined. The model's robustness and interpretability were examined using gradient-weighted class activation mapping (Grad-CAM) to visualize influential image regions and t-distributed stochastic neighbor embedding (t-SNE) to analyze learned feature embeddings. Lastly, a graphical user interface was developed for a prospective evaluation in the laboratory. RESULTS: The model achieved approximately 97% overall accuracy on the test set. Experiments revealed sensitivity to data set variability, suggesting that performance may improve by integrating additional training examples. Confidence scores aligned with accuracy, and interpretability analyses showed that the model focused on relevant image regions and learned embeddings demonstrated clear class separation. In the prospective evaluation, top 1 and top 3 accuracies decreased to approximately 78% and 92%, respectively. CONCLUSIONS: Our results indicate that a lightweight deep learning model can achieve high performance in classifying urine particles. Analysis of discrepancies between retrospective and prospective evaluations provides important insights toward reliable clinical application.
4 Apr 2026
Read appraisal →Clinical genetics
Clinical Feasibility of Long-Read WGS for DNA Methylation Signature Analysis
DNA methylation (DNAm) signatures have emerged as valuable diagnostic biomarkers for rare genetic disorders. To date, the most widely used approach for establishing and validating these signatures has relied on array-based technologies. However, in clinical diagnostics, there is a growing shift from short-read sequencing (SRS) toward long-read sequencing (LRS) technologies. Recent advances in platforms such as Pacific Biosciences (PacBio) and Oxford Nanopore Technologies (ONT) enable direct assessment of DNAm from native DNA, offering improved resolution and reduced technical bias compared to array-based technologies. In this study, we compared DNAm profiles generated by LRS with those obtained from DNAm arrays. DNAm profiles of two individuals with pathogenic KMT2D variants were analyzed using DNAm arrays, LRS using PacBio and ONT, and ONT multiplexed sequencing with adaptive sampling. A support vector machine (SVM) classifier trained on array data, as well as the public classification platform EpigenCentral, yielded correct predictions for all LRS samples, underscoring the potential of LRS platforms in DNAm-based diagnostics. Our results suggest that DNAm profiles generated by LRS align well with DNAm signatures established using DNAm arrays, supporting their feasibility in clinical and research applications with the added benefit of simultaneous methylation and sequence analysis.
4 Apr 2026
Read appraisal →IEEE transactions on medical imaging
Self-Paced Learning for Images of Antinuclear Antibodies
Antinuclear antibody (ANA) testing is a critical method for diagnosing autoimmune disorders such as Lupus, Sjögren's syndrome, and scleroderma. Despite its importance, manual ANA detection is slow, labor-intensive, and demands years of training. ANA detection is complicated by over 100 coexisting antibody types, resulting in vast fluorescent pattern combinations. Although machine learning and deep learning have enabled automation, ANA detection in real-world clinical settings presents unique challenges as it involves multi-instance, multi-label (MIML) learning. In this paper, a novel framework for ANA detection is proposed that handles the complexities of MIML tasks using unaltered microscope images without manual preprocessing. Inspired by human labeling logic, it identifies consistent ANA sub-regions and assigns aggregated labels accordingly. These steps are implemented using three task-specific components: an instance sampler, a probabilistic pseudo-label dispatcher, and self-paced weight learning rate coefficients. The instance sampler suppresses low-confidence instances by modeling pattern confidence, while the dispatcher adaptively assigns labels based on instance distinguishability. Self-paced learning adjusts training according to empirical label observations. Our framework overcomes limitations of traditional MIML methods and supports end-to-end optimization. Extensive experiments on one ANA dataset and three public medical MIML benchmarks demonstrate the superiority of our framework. On the ANA dataset, our model achieves up to +7.0% F1-Macro and +12.6% mAP gains over the best prior method, setting new state-of-the-art results. It also ranks top-2 across all key metrics on public datasets, reducing Hamming loss and one-error by up to 18.2% and 26.9%, respectively. The source code can be accessed at https://github.com/fletcherjiang/ANA-SelfPacedLearning.
4 Apr 2026
Read appraisal →J Allergy Clin Immunol
Proteomic profiling and machine learning for endotype prediction in chronic rhinosinusitis
BACKGROUND: Chronic rhinosinusitis (CRS) is a common, heterogeneous upper airway inflammatory disorder, affecting approximately 12% of the general population. The disease is clinically stratified into CRS without nasal polyps and CRS with nasal polyps, including the most severe subtype of nonsteroidal anti-inflammatory drug (NSAID)-exacerbated respiratory disease (N-ERD). OBJECTIVE: To identify molecular signatures and biomarkers allowing for the distinction between different disease endotypes and controls, we used targeted proteomics combined with bioinformatics and machine learning analyses. METHODS: Nasal secretions and serum from 80 patients (20 each of CRS without nasal polyps, CRS with nasal polyps, N-ERD, and disease controls) were subjected to high-throughput targeted proteomics (Olink). The expression patterns of 161 and 2677 proteins, for nasal secretions and serum, respectively, were analyzed alongside clinical evaluations of nasal polyp and smell test scores. RESULTS: Two distinct expression patterns were identified in nasal secretions: proteins associated with macrophage recruitment and type 2 inflammation were increased in CRS with nasal polyps and N-ERD, whereas proteins associated with innate immunity, particularly Toll-like receptor 4 signaling, were gradually downregulated from disease control to N-ERD. Furthermore, using machine learning, we confirmed 2 potential biomarkers for nasal polyposis: the glial cell line-derived neurotrophic factor in nasal secretions and Charcot-Leyden crystal protein in serum. CONCLUSIONS: Our findings provide unique insights into CRS pathophysiology and highlight potential biomarkers for precision diagnosis and treatment, particularly in severe cases such as N-ERD.
1 Jan 2026
Read appraisal →Blood
Artificial intelligence in hematology
Artificial intelligence (AI) and its subdiscipline, machine learning (ML), have the potential to revolutionize health care, including hematology. The diagnosis and treatment of hematologic disorders depend on the integration of diverse data sources, such as imaging, pathology, omics, and laboratory parameters. The increasing volume and complexity of patient data have made clinical decision-making more challenging. AI/ML hold significant potential for enhancing diagnostic accuracy, risk stratification, and treatment response prediction through advanced modeling techniques. Generative AI, a recent advancement within the broader field of AI, is poised to have a profound impact on health care and hematology. Generative AI can enhance the development of novel therapeutic strategies, improve diagnostic workflows by generating high-fidelity images or pathology reports, and facilitate more personalized approaches to patient management. Its ability to augment clinical decision-making and streamline research represents a significant leap forward in the field. However, despite this potential, few AI/ML tools have been fully implemented in clinical practice due to challenges related to data quality, equity, advanced infrastructure, and the establishment of robust evaluation metrics. Despite its promise, AI implementation in hematology faces critical challenges, including bias, data quality issues, and a lack of regulatory frameworks and safety standards that keep pace with rapid technological advancements. In this review, we provide an overview of the current state of AI/ML in hematology as of 2025, identify existing gaps, and offer insights into future developments.
1 Nov 2025
Read appraisal →