
We have a new low-frequency email list. Please consider signing up if you want to hear about our important announcements. melba-journal.org/contact.ht…

Open-access, independent, minimum fee journal. Co-founded by @arbtal,@ja_schnabel,@mertrory,@wmwells3,@MarcNiethammer,@adriandalca

We have a new low-frequency email list. Please consider signing up if you want to hear about our important announcements. melba-journal.org/contact.ht…

🎯 Authors evaluate uncertainty quantification methods for multi-task glioma diagnosis, combining MRI tumor segmentation and classification. Reliability varies across settings, highlighting the importance of task-specific evaluation. 🔎 Free to read: doi.org/10.59275/j.melba.202…

🚨 New publication alert: 📢 ”Towards Trustworthy AI for Glioma Diagnosis: A Task-Aware Evaluation of Uncertainty Quantification.” 🖊️ G E Mosquera Rojas, S R van der Voort, C M Pirkl, S Kaushik, M Smits, S Klein. ⬇️

🎯 Authors evaluate ensemble methods for hippocampal segmentation across dementia and epilepsy datasets. Ensemble-based uncertainty estimates help flag unreliable segmentations for expert review, supporting safer clinical use. 🔎 Free to read: doi.org/10.59275/j.melba.202…

🚨 New publication alert: 📢 ”Evaluation of Uncertainty-Aware Multi-Software Ensembles for Hippocampal Segmentation.” 🖊️ G Oliveira-Stahl, A Schroder, J Moggridge, H A Salhab, C Micallef, J Barnes, M J Cardoso, C H Sudre, M Grech-Sollars. ⬇️

🎯 Authors investigate privacy–fairness trade-offs in diffusion models for medical imaging, showing that rare anatomical features are either forgotten or memorized rather than generalized. 🔎 Free to read: doi.org/10.59275/j.melba.202…

🚨 New publication alert: 📢 ”A Data-Interventional Framework for Auditing Privacy and Fairness in Generative Medical Imaging.” 🖊️ M Dombrowski, B Kainz. ⬇️

🎯 Authors introduce a public dataset of nearly 8,000 retinal images acquired with portable fundus cameras in Brazil and with detailed image quality and degradation annotations. 🔎 Free to read: doi.org/10.59275/j.melba.202…

🚨 New publication alert: 📢 ”M-FIQ: Mobile Fundus Image Quality Assessment Dataset for Teleophthalmology Screening.” 🖊️ J M M Moreira, J D S Almeida, E P F Costa, J S Soares, ..., I G S C Santos, A C Silva, D B P Quintanilha, A C Paiva, G B Junior, M A G Campos. ⬇️

🎯 Authors introduce a public mammography dataset featuring over 12,000 images from 3,219 patients in India, addressing the underrepresentation of South Asian populations. 🔎 Free to read: doi.org/10.59275/j.melba.202…

🚨 New publication alert: 📢 ”The Indian Mammography DataBase (IMDB): A Versioned Open Mammography Resource from a Screening-Naive Indian Population for Artificial Intelligence Research.” 🖊️ O Shivom Nagpal, V Holla, K Chaturvedi,... , T Vyas, N Chavan, A Gupta, K Rangarajan. ⬇️

🎯 Authors introduce SAGE, a South Asian gastrointestinal endoscopy dataset supporting image captioning, classification, and visual question answering. Benchmark results highlight the need for more geographically diverse datasets. 🔎 Free to read: doi.org/10.59275/j.melba.202…

🚨 New publication alert: 📢 ”SAGE: An Expert-Annotated South Asian GI Endoscopy Dataset for Multimodal Learning and Hallucination Analysis.” 🖊️ N Oli, S Acharya, S Pokhrel, S Bhandari, R Rana, N Mani Shrestha, R Bahadur Gurung, Y Raj Shrestha, P K Gyawali, B Bhattarai. ⬇️

🎯 Authors introduce a bilingual English–Urdu benchmark for leukemia and white blood cell analysis, featuring 110,000 question–answer pairs across 20,000 cell images, and supporting the development of multilingual vision-language models. 🔎 Free to read: doi.org/10.59275/j.melba.202…

🚨 New publication alert: 📢 ”Multilingual Hematology Visual Question Answering Dataset.” 🖊️ H Malik, H Tooba Aftab, A Rehman, M Ali, W Sultani. ⬇️

🎯 Authors introduce a dataset of 72 dual-energy CT scans with expert-validated vertebral bone segmentations, supporting AI-based segmentation, quantitative bone analysis, and multiple myeloma research. 🔎 Free to read: doi.org/10.59275/j.melba.202…