Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 5 appraisals
European radiology
Deep learning improves image quality in motion-robust and sedation-free pediatric brain MRI
OBJECTIVES: Motion and limited compliance compromise diagnostic MR image quality, particularly in pediatric patients who frequently require sedation. Single-shot sequences offer a time-efficient alternative but suffer from reduced image quality. This study aimed to evaluate the diagnostic performance of a deep learning (DL) framework combining compressed sensing (CS) and convolutional neural networks (CNNs) to enhance T2-weighted single-shot MRI (T2-SSHDL) compared with conventional CS-based reconstruction (T2-SSHconv) and routinely acquired high-resolution T2-weighted sequences. MATERIALS AND METHODS: This prospective single-center study included 62 pediatric patients (mean age, 7.4 ± 4.9 years; 36 males, 26 females), who underwent T2-weighted single-shot brain MRI (29 sedated, 33 awake). Raw data were reconstructed using a DL-based pipeline and compared with conventional CS-based reconstructions. Quantitative metrics included apparent contrast-to-noise ratio (aCNR), apparent signal-to-noise ratio (aSNR), and edge rise distance (ERD). Two radiologists rated images for artifacts, sharpness, lesion conspicuity, and overall quality on a 5-point Likert scale. RESULTS: T2-SSHDL-sequences showed significantly higher aCNR (29.9 ± 22.6 vs. 26.7 ± 16.5; p < 0.001), aSNR (41.6 ± 27.9 vs. 38.2 ± 20.8; p = 0.003), and improved sharpness (ERD 0.90 ± 0.35 mm vs. 1.35 ± 0.42 mm; p < 0.001). Qualitative assessments confirmed superior image quality, lesion conspicuity, and sharpness (p < 0.001). Compared with high-resolution T2-weighted sequences, T2-SSHDL-sequences showed fewer motion artifacts and comparable lesion conspicuity in non-sedated patients. CONCLUSION: DL-based reconstruction significantly enhances the diagnostic quality of T2-weighted single-shot brain MRI in pediatric patients, enabling clinically usable, ultrafast, motion-robust imaging with potential to reduce the need for sedation. KEY POINTS: Question Can deep learning-based reconstruction elevate motion-robust single-shot T2-weighted pediatric brain MRI to diagnostic image quality levels, enabling reliable imaging without sedation? Findings Both quantitative and qualitative evaluations confirmed significantly improved image quality of deep learning-enhanced single-shot T2-weighted brain MRI compared with conventional reconstruction. Clinical relevance Deep learning-enhanced reconstruction improves image quality in ultrafast, motion-robust single-shot pediatric brain MRI, potentially reducing the need for sedation while preserving diagnostic accuracy. This approach may enhance patient safety and shorten examination time in routine neuroimaging.
2 Aug 2026
Read appraisal →PloS one
Artificial intelligence in the diagnosis of deep vein thrombosis: A scoping review
Deep vein thrombosis (DVT) is the formation of thrombi in the deep venous system, most often in the lower extremities. Although usually not life-threatening, DVT requires timely diagnosis to prevent complications such as pulmonary embolism and post-thrombotic syndrome. The growing demand for image interpretation has generated interest in applying artificial intelligence (AI) to automated DVT detection. This scoping review analyzes the performance of artificial intelligence in diagnosing DVT using computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound (US). We conducted a search across seven databases from inception to May 2025 using terms related to deep vein thrombosis, artificial intelligence, and machine learning. Eligible studies were limited to those evaluating DVT diagnosis using CT, MRI, or ultrasound. Two independent reviewers selected eligible studies, and quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). Eleven studies published between 2021 and 2025 met the inclusion criteria. Some of the AI algorithms included RetinaNet, Deep R-Belief Neural Networks, and Sooty Tern Optimization. US-based models were the most studied algorithms, with sensitivities and specificities ranging from 68 to 100% and 70-100%, respectively. The MRI-based model achieved sensitivities, specificities, and accuracies of 95% to 97%. One CT-based model demonstrated a sensitivity of 83%. Studies evaluated across multiple imaging datasets showed high sensitivities, specificities, and precision of 96% or higher. Future research should prioritize multicenter validation and integration of clinical factors. In addition, explainable frameworks capable of integrating multiple imaging datasets must be developed with attention to workflow efficiency and cost-effectiveness to support clinical translation. The results indicate that AI is best situated as a supplementary tool rather than a replacement for expert interpretation in DVT diagnosis.
23 June 2026
Read appraisal →Science advances
Artificial sparse neuron dendrites for visual information inference
In the human brain, dendrites exhibit nonlinear integration and sparse parallel processing capabilities, which can effectively perform visual tasks by integrating only a small subset of neuronal signals and play a crucial role in high-level information inference. However, conventional neuromorphic devices often ignore these important properties and require all neurons to perceive complete information. This makes it difficult to effectively replicate the efficient spatiotemporal processing capabilities of biological neuron dendrites. In this study, we present an artificial neuron dendrite array that integrates neurons, synapses, and dendrites, emulating the spatiotemporal spike integration properties of biological dendrites for precise parallel computation. Through multigate threshold regulation, the array enables parallel sparse spiking inference with random spatial distribution. This inference process forms a sparse dendritic spiking neural network (SD-SNN) that can perform compression, depth detection, and prediction. As a result, the SD-SNN achieves high-efficiency static and dynamic object processing while using only 0.5% of neuronal activity, slashing the power consumption by 98 and 65%, respectively. Our work reduces neural activity in the perception process by 99.5% while enhancing spatiotemporal computing capabilities and computational efficiency.
21 June 2026
Read appraisal →PloS one
Research on the semantic segmentation of Thangka images via an improved PIDNet.
In the digital preservation, restoration, and research of Thangka paintings, real-time semantic segmentation plays a crucial role in rapid image analysis. However, Thangka images exhibit intricate compositions, where principal figures often blend with backgrounds, ritual objects, and intricate ornaments, leading to blurred boundaries and fine details that challenge conventional segmentation methods in balancing accuracy and efficiency. To address this, we propose an improved PIDNet-based model incorporating ECA-Pag (Efficient Channel Attention-Path Aggregation) and LGFM (Local-Global Feature Fusion Module) modules, along with PConv3 (Partial Convolution 3), enhancing feature extraction and segmentation precision. Experimental results demonstrate that our model achieves 73.28% mIoU and mB-Fscore (mean Boundary F-score) of 40.01% on a custom Thangka dataset while maintaining 109.03 FPS (Frames Per Second), ensuring both high accuracy and real-time performance. Furthermore, evaluations on the Cityscapes benchmark confirm the model's generalization capability, outperforming baseline methods. This work provides an efficient and reliable solution for Thangka image segmentation, with potential applications in cultural heritage preservation and broader computer vision tasks.
17 May 2026
Read appraisal →Magnetic resonance in medicine
Accelerating Multiparametric Quantitative MRI Using Self-Supervised Scan-Specific Implicit Neural Representation With Model Reinforcement
PURPOSE: To develop a self-supervised scan-specific deep learning framework for reconstructing accelerated multiparametric quantitative MRI (qMRI). METHODS: We propose REFINE-MORE (REference-Free Implicit NEural representation with MOdel REinforcement), combining an implicit neural representation (INR) architecture with a model reinforcement module that incorporates MR physics constraints. The INR component enables informative learning of spatiotemporal correlations to initialize multiparametric quantitative maps, which are then further refined through an unrolled optimization scheme enforcing data consistency. To improve computational efficiency, REFINE-MORE integrates a low-rank adaptation strategy that promotes rapid model convergence. We evaluated REFINE-MORE on accelerated multiparametric quantitative magnetization transfer imaging for simultaneous estimation of free water spin-lattice relaxation, tissue macromolecular proton fraction, and magnetization exchange rate, using both phantom and in vivo brain data. RESULTS: Under 4× and 5× accelerations on in vivo data, REFINE-MORE achieved superior reconstruction quality, demonstrating the lowest normalized root-mean-square error and highest structural similarity index compared to baseline methods and other state-of-the-art model-based and deep learning approaches. Phantom experiments further showed strong agreement with reference values, underscoring the robustness and generalizability of the proposed framework. Additionally, the model adaptation strategy improved reconstruction efficiency by approximately fivefold. CONCLUSION: REFINE-MORE enables accurate and efficient scan-specific multiparametric qMRI reconstruction, providing a flexible solution for high-dimensional, accelerated qMRI applications.
6 May 2026
Read appraisal →