analiza aiqbench
This study advances medical imaging AI by optimizing mammogram analysis through transfer learning, resolution reduction techniques, and multi-view classification architectures. The research establishes new state-of-the-art benchmarks for breast cancer detection, demonstrating significant performance gains in both single and multiple-view diagnostic scenarios. These findings provide critical insights into model design and robustness for developing more reliable computer-aided diagnostic tools.
Optimizing Breast Cancer Detection in Mammograms: A Comprehensive Study of Transfer Learning, Resolution Reduction, and Multi-View Classification
arXiv:2503.19945v4 Announce Type: replace-cross Abstract: Mammography, an X-ray-based imaging technique, remains central to the early detection of breast cancer. Recent advances in artificial intelligence have enabled increasingly sophisticated computer-aided diagnostic methods, evolving from patch-based classifiers to whole-image approaches and then to multi-view architectures that jointly analyze complementary projections. Despite this progress, several critical questions remain unanswered. In this study, we systematically investigate these issues by addressing five key research questions: (1) the role of patch classifiers in performance, (2) the transferability of natural-image-trained backbones, (3) the advantages of learn-to-resize over conventional downscaling, (4) the contribution of multi-view integration, and (5) the robustness of findings across varying image quality. Beyond benchmarking, our experiments demonstrate clear performance gains over prior work. For the CBIS-DDSM dataset, we improved single-view AUC from 0.8153 to 0.8343, and multiple-view AUC from 0.8483 to 0.8658. Using a new comparative method, we also observed a 0.0217 AUC increase when extending from single to multiple-view analysis. On the complete VinDr-Mammo dataset, the multiple-view approach further improved results, achieving a 0.0492 AUC increase over single view and reaching 0.8511 AUC overall. These results establish new state-of-the-art benchmarks, providing clear evidence of the advantages of multi-view architectures for mammogram interpretation. Beyond performance, our analysis offers principled insights into model design and transfer learning strategies, contributing to the development of more accurate and reliable breast cancer screening tools. The inference code and trained models are publicly available at https://github.com/dpetrini/multiple-view.
Abstract:Mammography, an X-ray-based imaging technique, remains central to the early detection of breast cancer. Recent advances in artificial intelligence have enabled increasingly sophisticated computer-aided diagnostic methods, evolving from patch-based classifiers to whole-image approaches and then to multi-view architectures that jointly analyze complementary projections. Despite this progress, several critical questions remain unanswered. In this study, we systematically investigate these issues by addressing five key research questions: (1) the role of patch classifiers in performance, (2) the transferability of natural-image-trained backbones, (3) the advantages of learn-to-resize over conventional downscaling, (4) the contribution of multi-view integration, and (5) the robustness of findings across varying image quality. Beyond benchmarking, our experiments demonstrate clear performance gains over prior work. For the CBIS-DDSM dataset, we improved single-view AUC from 0.8153 to 0.8343, and multiple-view AUC from 0.8483 to 0.8658. Using a new comparative method, we also observed a 0.0217 AUC increase when extending from single to multiple-view analysis. On the complete VinDr-Mammo dataset, the multiple-view approach further improved results, achieving a 0.0492 AUC increase over single view and reaching 0.8511 AUC overall. These results establish new state-of-the-art benchmarks, providing clear evidence of the advantages of multi-view architectures for mammogram interpretation. Beyond performance, our analysis offers principled insights into model design and transfer learning strategies, contributing to the development of more accurate and reliable breast cancer screening tools. The inference code and trained models are publicly available at this https URL.
18 pages - Final published Version. Final paper title slightly changed
Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2503.19945 [eess.IV]
(or arXiv:2503.19945v4 [eess.IV] for this version)
https://doi.org/10.1002/ima.70436 DOI(s) linking to related resources
From: Daniel Petrini Mr. [view email] [v1] Tue, 25 Mar 2025 11:51:21 UTC (788 KB) [v2] Sun, 31 Aug 2025 23:36:46 UTC (355 KB) [v3] Tue, 7 Oct 2025 23:35:30 UTC (1,062 KB) [v4] Sat, 3 Oct 2026 20:34:08 UTC (1,537 KB)
Text integral sub licența deschisă a publicației · 315 cuvinte
