analiza aiqbench
This research addresses a critical gap in deploying medical AI by introducing a compression framework that prioritizes uncertainty preservation alongside standard accuracy metrics like Dice score. The proposed SGAV-MP method ensures that reduced model sizes do not compromise safety-critical reliability signals, enabling safer deployment of compressed models in MRI and CT segmentation tasks.
Reliability-Preserving Compression for Deployable Medical Image Segmentation
arXiv:2610.25460v1 Announce Type: cross Abstract: Post-training compression of pretrained medical segmentation models commonly uses quantization, pruning, or distillation to reduce model size and computational cost. Existing medical segmentation quantization studies mainly evaluate Dice, model size, memory, and latency, but this is incomplete for safety-critical use because compressed models should also preserve uncertainty signals that indicate unreliable predictions. We study this problem in compressed MRI/CT segmentation and formalize why Dice can remain stable while uncertainty changes. Motivated by this failure mode, we propose score-guided aggregate-validated mixed precision (SGAV-MP), a post-training compression framework that selects layer precisions under reliability constraints. SGAV-MP ranks quantization actions with a compact benefit-risk score and accepts them only after full-model reliability validation on a calibration set. Across five 2D backbones and three 3D benchmarks, SGAV-MP preserves FP32-level Dice and high uncertainty-map agreement, achieving $3.59\times$--$7.8\times$ compression in 2D and $4.73\times$--$7.82\times$ in 3D. We also report backend-supported TensorRT deployment measurements separately to distinguish runtime execution from reliability-preserving storage compression.
Abstract:Post-training compression of pretrained medical segmentation models commonly uses quantization, pruning, or distillation to reduce model size and computational cost. Existing medical segmentation quantization studies mainly evaluate Dice, model size, memory, and latency, but this is incomplete for safety-critical use because compressed models should also preserve uncertainty signals that indicate unreliable predictions. We study this problem in compressed MRI/CT segmentation and formalize why Dice can remain stable while uncertainty changes. Motivated by this failure mode, we propose score-guided aggregate-validated mixed precision (SGAV-MP), a post-training compression framework that selects layer precisions under reliability constraints. SGAV-MP ranks quantization actions with a compact benefit-risk score and accepts them only after full-model reliability validation on a calibration set. Across five 2D backbones and three 3D benchmarks, SGAV-MP preserves FP32-level Dice and high uncertainty-map agreement, achieving $3.59\times$--$7.8\times$ compression in 2D and $4.73\times$--$7.82\times$ in 3D. We also report backend-supported TensorRT deployment measurements separately to distinguish runtime execution from reliability-preserving storage compression.
Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Signal Processing (eess.SP)
arXiv:2610.25460 [eess.IV]
(or arXiv:2610.25460v1 [eess.IV] for this version)
Text integral sub licența deschisă a publicației · 183 cuvinte