The study presents an open‑source multimodal model that classifies bacterial versus fungal microbial keratitis using 17,158 slit‑lamp images from 1,645 patients in the US and India. The model, which integrates cross‑modality contrastive learning, modality‑specific fine‑tuning, and seven clinical variables, achieved a macro‑F1 of 83.06% and an AUROC of 89.9% overall, outperforming five comparator methods. However, performance varied markedly by country—AUROC 83.9% in India versus 67.0% in the US—highlighting geographic confounding and the need for independent geographic validation.
arXiv:2610.33965v1 Announce Type: new Abstract: Automated classification of gastrointestinal (GI) findings in endoscopic images could support early diagnosis, but architectures are often chosen on ac...
arXiv:2606. 13135v1 Announce Type: cross Abstract: Purpose.
arXiv:2609.09189v1 Announce Type: cross Abstract: High classification accuracy alone is insufficient for clinical image analysis, where calibrated confidence and reliable uncertainty estimates are es...
Background: Artificial intelligence (AI)-based glaucoma detection from colour fundus photographs (CFP) offers scalable screening, but performance may decline on external datasets because of difference...
The study developed a Vision Transformer-based deep learning model with uncertainty estimation to detect glaucoma from colour fundus photographs across multi‑ethnic populations, including those with high myopia. Using 56,483 images for training, the model achieved an internal AUROC of 98.7% and maintained high performance (AUROC 86.4–99.6%) on 16 external datasets from eight countries. In high‑myopia eyes, the model outperformed ophthalmologists and matched specialists when full clinical data were available.
The study presents a cost‑sensitive XGBoost model that predicts ESBL‑producing Enterobacteriaceae at the time of culture ordering using 45 pre‑culture EHR features from 132,955 cultures across 12 hospitals. At 90% sensitivity the model achieves a 95.8% negative predictive value, potentially allowing carbapenem‑sparing therapy in non‑ICU settings while avoiding unnecessary broad‑spectrum treatment for 307 out of every 1,000 cultures, at the expense of 14 missed ESBL cases per 1,000. SHAP analysis highlights prior ESBL colonization as the strongest predictor, with minimal performance loss when socioeconomic deprivation features are omitted, supporting equitable bedside deployment.
arXiv:2608.22059v1 Announce Type: cross Abstract: Pretrained image encoders are central to medical image classification, where expert annotation is costly and task-specific cohorts are often limited....
The study evaluated whole‑slide image classifiers for gastric cancer across five Korean hospitals, using foundation‑model features and leave‑one‑hospital‑out validation. A hierarchical patch‑region‑slide graph aggregator (CHIME‑MIL) with site alignment and two‑scale fusion achieved a mean AUC of 0.837, outperforming six conventional attention and transformer aggregators that lacked these steps. However, when matched with the same alignment and fusion, CHIME‑MIL did not surpass simple attention‑based pooling, and global max pooling performed best (AUC 0.855). The alignment step, which subtracts each hospital’s mean feature vector without labels, enabled onboarding of a new hospital from only ten unlabeled slides but yielded only a modest 0.002 AUC improvement, and the model’s accuracy gains were inconsistent. "whyItMatters":"The findings show that adding complex aggregation methods does not necessarily improve cross‑hospital discrimination, highlighting the importance of simpler, label‑free onboarding and calibration checks for deployable pathology systems."
arXiv:2607. 05825v1 Announce Type: cross Abstract: Background.
The study investigates how image‑domain Poisson perturbations affect the classification of ischemic core/penumbra in non‑contrast CT slices. Using a CPAISD cohort, the authors compared direct ResNet‑18 classification with a denoising‑then‑classifying pipeline and found that denoising generally reduced performance. A prospective experiment with joint denoising‑classification models showed no statistically significant advantage over direct noisy classification, indicating limited robustness to Poisson noise.
arXiv:2607. 19864v1 Announce Type: cross Abstract: Diabetic retinopathy is a leading cause of preventable blindness; its early lesions are small, low contrast, and easily missed in manual screening.
The paper introduces a dual cross‑attention deep learning framework for automated grading of Cervical Intraepithelial Neoplasia (CIN) and prediction of Swede scores, using a newly released BUET Multi‑Center Colposcopy Dataset. The architecture fuses paired multimodal cervigrams and employs a custom composite loss to handle class imbalance, achieving 71.85% accuracy and 86.23% AUC‑ROC for three‑class CIN grading, and AUC‑ROC values between 75.7% and 88.4% for individual Swede score components. The total predicted Swede Score has a mean absolute error of 1.489, indicating potential for AI‑assisted colposcopy screening in resource‑limited settings.