Despite numerous deep learning models reporting high accuracy for oral cancer classification, most rely on a single train/test split on a single dataset, with limited statistical validation, external generalization testing, or explainability. This study addresses that gap through a statistically validated, externally tested, and dual-explainable classification framework. We provide a rigorous benchmarking study contrasting 20 hybrid combinations (4 pre-trained backbones × 5 classical classifiers) against fine-tuned end-to-end CNNs. Across repeated stratified fivefold cross-validation with 10 repetitions, the hybrid framework consistently outperformed end-to-end CNNs, with ResNet50 paired with an SVM-RBF classifier achieving top performance: 0.960 accuracy (95
Trabecular bone score (TBS) provides information on bone microarchitecture beyond bone mineral density (BMD) and may be particularly relevant in type 2 diabetes mellitus (T2DM). This exploratory cross-sectional study examined associations of age and body mass index (BMI) with lumbar TBS (L1–L4) and their variation by gender in 240 adults with T2DM. Generalized additive models (GAMs) included overall smooths, gender-specific smooths, and tensor-product surfaces. Between-gender smooth contrasts were evaluated using joint simultaneous 95
Scapular morphology plays a critical role in the diagnosis and treatment of shoulder disorders. However, current assessment methods primarily rely on manual annotation of three-dimensional computed tomography (3D CT) scans by clinicians, which are time-consuming, labor-intensive, and prone to inter-observer variability. The objective of this study is to develop and validate a deep learning–based open-source pipeline for automated scapular morphology assessment, aiming to improve accuracy, reproducibility, and clinical efficiency. Descriptive Laboratory Study; Level of evidence, 2. We retrospectively collected 793 CT images of the shoulder from 774 patients (mean age, 60.00 ± 18.62 years). A 2D U-Net model performed high-resolution glenoid segmentation (native-resolution axial CT slices, no down-sampling), and a 3D nnUNet model was employed to localize five anatomical scapular landmarks: trigonum spinae (TS), angulus inferior (AI), processus coracoideus (PC), acromion (AC), and angulus acromialis (AA). Model performance was evaluated using the Dice coefficient and precision-recall metrics for glenoid segmentation and the Euclidean distance between the predicted and ground truth landmark position. This pipeline was designed as an automated open-source framework that can be readily integrated into clinical workflows or research environments through clinician-preferred user interface. The glenoid segmentation model achieved a 3D Dice coefficient of 98.46
Imaging informatics is becoming more central to clinical practice with the expansion of artificial intelligence, telehealth, patient portals, and diagnostic imaging. Despite these technological advances, formal education in imaging informatics is typically not part of the traditional medical school curriculum. This paper describes the creation and implementation of a hands-on active learning module for medical students focused on foundational concepts in imaging informatics. The active learning module for second-year medical students was developed and led by faculty with expertise in imaging informatics. Students were instructed to view a brief prerecorded video prior to the session. During the in-person 50-min session, students were divided into groups of 6–10 and asked to take a photograph of a body part and send it to another team member. This process was repeated multiple times using a different image-sharing method and partner each time, followed by discussion of challenges encountered. Post-session evaluations yielded a mean rating of 4.33/5 for being conducive to learning and engaging the audience; opportunities to apply foundational concepts and activities that promoted problem solving each received mean ratings of 4.00/5. Qualitative feedback indicated that some students found the material new and applicable to clinical practice, while others desired more advanced content, clearer key takeaways, and additional information about physicians’ roles in informatics. Qualitative feedback informed iterative modifications to the session. This educational session facilitated a rich discussion of core imaging informatics concepts including the importance of image standards, metadata to accompany images, workflow automation, and interoperability.
Reliable linkage of chest X-ray examinations to corresponding reports is fundamental to large radiology archives. Yet heterogeneous view composition and similar image–report content across examinations, particularly longitudinal records, make exact content-based linkage difficult. Global vision–language representations support archive-wide search but can obscure view-specific and section-level correspondences, whereas exhaustive fine-grained interaction is computationally costly. We propose CXR-CoRe, a two-stage view- and section-aware framework. Stage 1 combines learned view embeddings with residual gated multi-view fusion, achieving 89.3
