Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 8 appraisals
PloS one
Deep learning based two-way feature depiction model for brain tumor detection
Brain tumors are one of the most fatal disorders that cause one of the highest mortalities in the world. Gliomas are the most common primary brain tumors originating from glial cells in the central nervous system. Traditionally, a tissue sample is extracted and examined for its genetic and characteristic properties. This method is invasive, painful, and takes a longer period to produce results. Various automatic Deep learning (DL) based schemes have been presented for the brain glioma detection, but they lack due to poor explainability, lower generalization, poor feature depiction, class imbalance problem and lower detection rate. This paper presents a deep learning based brain tumor detection using two way feature depiction model (TWFDM) that combines the 2D-Deep Convolution Neural Network (DCNN) and 1D-DCNN. The 2D-DCNN accepts the raw MRI images and the 1D-DCNN accepts the handcrafted local binary pattern (LBP), gray level cooccurrence matrix (GLCM), and Histogram of Oriented Gradient (HOG) features. Furthermore, improved particle swarm optimization (IPSO) is used for feature selection to minimize the computational complexity of the TWFDM system. The proposed TWFDM achieves an overall accuracy of 96.25%, a recall of 96.34%, a precision of 96.31%, and an F1-score of 96.32% on the Brain MRI dataset for four-class classification, representing an important improvement over traditional techniques.
3 July 2026
Read appraisal →Journal of neuro-oncology
AI for prognosis and treatment stratification in glioblastoma neurosurgery: a systematic review.
BACKGROUND: Glioblastoma (GBM) remains one of the most lethal adult primary brain tumors, and neurosurgical decision-making increasingly depends on integrating imaging, molecular, perioperative, and post-treatment data. Artificial intelligence (AI) methods have been proposed for several clinically relevant GBM tasks, but the literature remains heterogeneous and difficult to translate into practice. METHODS: We performed a PROSPERO-registered systematic review of AI, machine learning, and deep learning studies using MRI-derived and/or multimodal perioperative data in GBM for prognosis, risk stratification, treatment-response assessment, post-treatment classification, recurrence/progression prediction, and molecular prediction. Risk of bias was assessed using PROBAST-informed criteria. RESULTS: Thirty studies were included. Survival-focused tasks predominated (20/30, 66.7%), with radiomics plus conventional machine learning as the most common model family (13/30, 43.3%), followed by deep learning (8/30, 26.7%) and hybrid deep learning plus radiomics approaches (4/30, 13.3%). Validation was predominantly internal, and external validation was uncommon (5/30, 16.7%). CONCLUSIONS: AI shows promise for prognosis and treatment stratification in GBM neurosurgery, but current evidence is limited by heterogeneity, incomplete external validation, and inconsistent methodological reporting. CLINICAL TRIAL NUMBER: Not applicable.
20 June 2026
Read appraisal →Progress in biomedical engineering (Bristol, England)
Advancement of deep learning models with whole slide image in diagnosis, subtyping and prognosis for glioma.
Gliomas are heterogeneous primary central nervous system tumors with diverse molecular subtypes and variable prognosis. A paradigm shift in histopathology is underway as samples are digitized into whole-slide images (WSIs). Deep learning (DL) shows great potential for glioma diagnosis, subtyping, grading, and prognosis. This review summarizes recent advances in WSI-based DL models, covering data preprocessing, model architectures, and translational performance. Model evolution has progressed from CNNs (e.g. ResNet, capturing local features via convolution) to Transformers (e.g. Vision transformer, modeling global dependencies via self-attention), hybrid architectures, and large language model. Typical pipelines include quality control, stain normalization, patch extraction, feature integration, and prediction. Public datasets such as The Cancer Genome Atlas and Clinical Proteomic Tumor Analysis Consortium serve as key resources. On various datasets (public and institutional), a glioma-specific model ResNet-50 achieved AUC 0.983 for subtype classification; a Vision Transformer model reached AUC 0.960 for molecular typing; a hybrid model, ROAM, attained AUC 0.990; and the pan-cancer hybrid model called CHIEF achieved AUC 0.9397 for diagnosis. Attention heatmaps and Shapley Additive exPlanations provide interpretability by linking model outputs to histologic regions. Novel multimodal fusion integrates genomic, proteomic, radiologic, and clinical data to enhance prediction and uncover biological relevance. Future research may focus on enhancing model generalizability and predictive accuracy via advanced architectures, developing lightweight models for resource-limited settings to facilitate clinical translation.
20 June 2026
Read appraisal →Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology
Concurrent de novo glioblastoma and meningiomatosis: a case report and systematic review of clinical, molecular, and topographical characteristics
INTRODUCTION: The co-occurrence of glioblastoma (GBM) and meningioma in a single patient is an exceptionally rare clinical phenomenon, often associated with prior irradiation or genetic syndromes. This study presents a unique case of concurrent de novo GBM and meningiomatosis and provides a systematic review of the literature to characterize this rare association. CASE PRESENTATION AND METHODS: A 78-year-old female with a known history of untreated meningiomatosis presented with refractory seizures and aphasia. Imaging revealed multiple stable extra-axial lesions and a new, distinct intra-axial right temporal lesion. The patient underwent a temporal lobectomy with gross total resection. Histopathology confirmed WHO grade 4 GBM (IDH-wildtype). Following multidisciplinary review, she completed adjuvant radiotherapy. At 6-month follow-up, she remained clinically stable with no recurrence. A systematic review was conducted, identifying 34 cases of concurrent GBM and meningioma. Data regarding demographics, topography, molecular profiles, and survival were analyzed. RESULTS: The median age was 63 years (range 30-86), with a slight male predominance (52.9%). Tumors were synchronous in 88.2% of cases. Topographically, 47.1% were "collision" or adjacent tumors, while 32.3% occurred in different hemispheres. Most meningiomas were WHO grade I (79.4%), while 97.1% of glial tumors were grade 4 GBM. Molecular analysis revealed classic independent drivers: 22q loss/NF2 mutations in meningiomas and 10q loss/EGFR amplification in GBMs. Median overall survival was 7 months (range 0.5-24 months), with modern trimodal therapy (Stupp protocol) showing a trend toward extended survival compared to older cohorts (Fig. 1). CONCLUSION: The coexistence of GBM and meningioma appears mostly coincidental, driven by distinct molecular pathways rather than a common progenitor. Despite the benign nature of the associated meningioma, prognosis is dictated by the aggressive GBM component. Early surgical intervention and modern adjuvant therapy remain essential for optimizing survival in these complex cases.
31 May 2026
Read appraisal →European radiology
Predicting molecular types of adult-type diffuse gliomas based on MRI reports with large language models
OBJECTIVES: To evaluate the performance of large language models (LLMs) in predicting molecular types of adult-type diffuse gliomas according to the 2021 WHO classification using MRI radiology reports. MATERIALS AND METHODS: This retrospective study included 2169 patients diagnosed with adult-type diffuse gliomas (294 oligodendrogliomas, 295 IDH-mutant astrocytomas, and 1580 IDH-wildtype glioblastomas) between July 2005 and March 2024 from four hospitals in Asia and Europe. Seven proprietary and open-source LLMs were assessed: GPT-4o-mini, GPT-4.1-mini, Llama 3.1 8B, Llama 3.1 70B, Qwen2.5 7B, Deepseek-r1 8B, and Mistal 7B. The performance of LLMs in classifying molecular types was compared based on the provision of relevant knowledge of glioma imaging findings (knowledge-based vs. naïve prompt). The impact of radiologists' subspecialization in neuro-oncology, report quality, and reporting language on LLMs' performance was also evaluated. RESULTS: LLMs achieved significantly higher (naïve vs. knowledge-based; GPT-4o-mini, 77.0% vs. 79.1%, p < 0.001; Qwen2.5 7B, 75.9% vs. 79.5%, p < 0.001; Deepseek-r1 8B, 66.0% vs. 73.2%, p < 0.001) or comparable accuracy (GPT-4.1-mini, 78.7% vs. 78.6%; Llama 3.1 70B, 78.0% vs. 78.1%; Mistral 7B, 58.4% vs. 57.4%) using knowledge-based prompt compared to naïve prompt, except for Llama 3.1 8B (65.4% vs. 44.6%, p < 0.001). Differences in accuracy were more pronounced in smaller-sized LLMs. Additionally, the accuracy was significantly higher with reports by neuro-oncology specialists and high-quality reports in all LLMs (p < 0.001). CONCLUSIONS: LLMs may provide preoperative information on the tumor types of adult-type diffuse gliomas from MRI reports by providing relevant knowledge in the prompt. Informative and descriptive reports could further enhance LLMs' performance. KEY POINTS: Question Our study aimed to evaluate large language models' (LLMs) ability to efficiently predict molecular types of adult-type diffuse gliomas according to the 2021 WHO classification. Findings Larger models generally showed better accuracy and were less sensitive to domain-specific knowledge. Their performance improved when using high-quality, longer reports or reports by neuro-oncology specialists. Clinical relevance These findings highlight the potential role of LLMs in predicting glioma molecular types, underscoring the importance of informative and descriptive reports in enhancing their performance.
4 May 2026
Read appraisal →Neurosurgery
Congress of Neurological Surgeons Systematic Review and Evidence-Based Guidelines for the Treatment of Adults With WHO Grade II Diffuse Glioma: Update
BACKGROUND: The management of World Health Organization (WHO) grade II diffuse glioma is an important facet of all physicians involved in neuro-oncology. OBJECTIVE: This is an update of the evidence-based guidelines for management of WHO grade II diffuse gliomas published by the Congress of Neurological Surgeons and American Association of Neurological Surgeons in 2015. METHODS: The medical literature from January 1, 2013, through January 31, 2020, was searched to determine if information was available to update, modify, or create new recommendations related to imaging, surgical approaches, neuropathology and molecular markers, radiotherapy, chemotherapy, and management of tumor recurrence. RESULTS: The writing group used the information from the updated literature search to formulate recommendations based on this evidence and not simply built on biased consensus or expert opinion. CONCLUSION: This series of guideline documents provides an update of the information and recommendations provided in the 2015 version. It sets a benchmark as to the published information we have to support the management of this difficult disease. It also provides clues to key investigations that are necessary to move us toward effective control of WHO grade II diffuse gliomas.
16 Apr 2026
Read appraisal →European radiology
Attention-based deep learning network for predicting World Health Organization meningioma grade and Ki-67 expression based on magnetic resonance imaging
OBJECTIVES: Preoperative assessment of World Health Organization (WHO) meningioma grading and Ki-67 expression is crucial for treatment strategies. We aimed to develop a fully automated attention-based deep learning network to predict WHO meningioma grading and Ki-67 expression. MATERIALS AND METHODS: This retrospective study included 952 meningioma patients, divided into training (n = 542), internal validation (n = 96), and external test sets (n = 314). For each task, clinical, radiomics, and deep learning models were compared. We used no-new-Unet (nn-Unet) models to construct the segmentation network, followed by four classification models using ResNet50 or Swin Transformer architectures with 2D or 2.5D input strategies. All deep learning models incorporated attention mechanisms. RESULTS: Both the segmentation and 2.5D classification models demonstrated robust performance on the external test set. The segmentation network achieved Dice coefficients of 0.98 (0.97-0.99) and 0.87 (0.83-0.91) for brain parenchyma and tumour segmentation. For predicting meningioma grade, the 2.5D ResNet50 achieved the highest area under the curve (AUC) of 0.90 (0.85-0.93), significantly outperforming the clinical (AUC = 0.77 [0.70-0.83], p < 0.001) and radiomics models (AUC = 0.80 [0.75-0.85], p < 0.001). For Ki-67 expression prediction, the 2.5D Swin Transformer achieved the highest AUC of 0.89 (0.85-0.93), outperforming both the clinical (AUC = 0.76 [0.71-0.81], p < 0.001) and radiomics models (AUC = 0.82 [0.77-0.86], p = 0.002). CONCLUSION: Our automated deep learning network demonstrated superior performance. This novel network could support more precise treatment planning for meningioma patients. KEY POINTS: Question Can artificial intelligence accurately assess meningioma WHO grade and Ki-67 expression from preoperative MRI to guide personalised treatment and follow-up strategies? Findings The attention-enhanced nn-Unet segmentation achieved high accuracy, while 2.5D deep learning models with attention mechanisms achieved accurate prediction of grades and Ki-67. Clinical relevance Our fully automated 2.5D deep learning model, enhanced with attention mechanisms, accurately predicts WHO grades and Ki-67 expression levels in meningiomas, offering a robust, objective, and non-invasive solution to support clinical diagnosis and optimise treatment planning.
4 Apr 2026
Read appraisal →Neuro Oncol
EANS-EANO guidelines on the extent of resection in gliomas
Intense efforts are employed to optimize neurosurgical resection of gliomas in different clinical scenarios. However, the level of evidence employed toward personalized treatment decisions is still moderate. Therefore, a joint task force was created by the European Association for Neurosurgical Societies (EANS) and the European Association of Neuro-Oncology (EANO) in order to provide interdisciplinary guidance on the extent of resection (EOR) in glioma surgery. This task force assessed the data on the EOR in gliomas, evaluated the level of evidence, and composed recommendations. The EOR is determined routinely by postoperative magnetic resonance imaging. Since EOR may be associated with survival and may have an impact on epilepsy, neurocognition, quality of life, neurological status, and also on planning of radiation and pharmacotherapy (evidence classes I-IV), recommendations of different levels can be made for the resection of newly diagnosed glioblastomas (level A), recurrent glioblastomas (level C), newly diagnosed (level B) and recurrent (good practice point) IDH mutant gliomas, ependymomas in children (level B) and adults (good practice point) as well as World Health Organization CNS grade 1 tumors (good practice point). We hereby report the outcome of this collaborative effort by the joint task force between EANS and EANO.
1 Jan 2026
Read appraisal →