The paper introduces CalSAM, a lightweight adaptation framework that fine‑tunes only the mask decoder of the Segment Anything Model (SAM) while keeping its encoders frozen. CalSAM employs a Feature Fisher Information Penalty (FIP) to reduce encoder sensitivity to domain shift and a Confidence Misalignment Penalty (CMP) to curb overconfident voxel‑wise errors. Experiments on cross‑center, scanner‑shift, and motion‑corrupted brain MRI datasets show significant gains in Dice similarity coefficient, Hausdorff distance, and expected calibration error, with only a modest training‑time overhead.
arXiv:2410.18321v3 Announce Type: replace Abstract: Confidence calibration matters wherever a classifier's probabilities, not just its labels, are consumed downstream. We study Focal Calibration Loss...
arXiv:2606. 19300v1 Announce Type: cross Abstract: Glioma segmentation in multiparametric MRI is a critical component of treatment planning.
The paper proposes a lightweight residual refiner that post‑processes the outputs of a two‑model ensemble for brain‑MRI inpainting. By training the refiner with an λ‑weighted combination of α loss and SSIM, the authors achieve a modest but statistically significant SSIM improvement (from 0.8767 to 0.8780 on a held‑out set) without altering MSE. Ablations show that adding a third model or using classical unsharp masking does not yield similar gains, indicating the improvement comes from learned sharpening rather than generic post‑processing.
MVC-Bench is a calibration-focused benchmark for medical vision‑language models, evaluating how well these models express confidence across different modalities, backbones, and domain shifts. It tests robustness to modality, backbone, and domain changes, the effectiveness of calibration and prompt‑tuning strategies, and stability under prompt‑template and random‑seed variations. The benchmark includes 1638 experiments, reporting accuracy and Expected Calibration Error (ECE) along with other calibration metrics, and introduces a simple train‑time calibration method, Multi‑Class Margin (MCM) regularization, that achieves the lowest ECE in most settings.
MVC-Bench is a new benchmark designed to evaluate the calibration of vision‑language models (VLMs) and medical VLMs (Medical‑VLMs) for medical image classification. It tests calibration across robustness to modality, backbone, and domain shift; effectiveness of calibration strategies and prompt‑tuning methods; and stability under prompt‑template and random‑seed variations. The benchmark includes eight backbones, three medical modalities (fundus imaging, histopathology, chest X‑ray), and compares post‑hoc, train‑time, and zero‑shot calibration approaches, reporting accuracy, Expected Calibration Error (ECE), Maximum Calibration Error (MCE), and Adaptive Calibration Error (ACE) over 1,638 experiments, while also proposing a Multi‑Class Margin (MCM) regularization technique that improves ECE in most settings.
arXiv:2609.15888v1 Announce Type: cross Abstract: Deep networks trained on structural MRI for Alzheimer's disease (AD) staging often reach reasonable accuracy while attending to anatomically irreleva...
arXiv:2608. 16377v1 Announce Type: cross Abstract: Instance-level lesion detection has been an increasingly larger focal point in medical image segmentation besides the more standard voxel-level overlap.
arXiv:2607. 16317v1 Announce Type: cross Abstract: Deep networks now subtype brain tumors on MRI about as well as specialist readers, yet accuracy is not what keeps them out of the clinic.
arXiv:2603. 04795v2 Announce Type: replace-cross Abstract: Medical image analysis depends on accurate segmentation and controllable synthesis, but both tasks face severe spatial imbalance: lesions occupy small regions against large backgrounds.
The paper presents a method for the BIC-MAC challenge, aiming to generate pseudo‑CT images from NAC‑PET, DIXON MRI, and a topogram, and to evaluate both the pseudo‑CT and the resulting attenuation‑corrected PET. The authors improve upon a 3D U‑Net baseline by focusing on loss design: they compute an L1 error in the Carney attenuation‑coefficient space, weighted by anatomical region, and incorporate DIXON MRI as additional input only after this loss was applied. Finally, they fuse two independently trained models via a fixed convex combination, achieving better performance than either model alone and topping the public validation leaderboard.
arXiv:2608.28681v1 Announce Type: new Abstract: Probability calibration aligns model confidence with predictive accuracy, enabling clinicians to identify unreliable segmentation regions. This alignme...
The paper introduces SWIFT, a Swin V2‑based model pretrained on 10,444 3D CT volumes and fine‑tuned for rectal cancer segmentation on T2‑weighted MRI. Four configurations—full fine‑tuning (SWIFT), decoder compression (SWIFTe), low‑rank adaptation (SWIFTe‑LoRA), and a LoRA‑decoder ensemble (SWIFTe‑LDE4)—were evaluated on 247 cases, showing that SWIFTe reduces parameters by 70.1% while improving tumor detection and radiomic agreement. The study also demonstrates a trade‑off between detection and boundary agreement, and highlights that SWIFTe‑LDE4 achieves the lowest calibration errors after temperature scaling.