Intelligent oncology最新文献

{"title":"Assessing Quantitative Performance and Expert Review of Multiple Deep Learning-Based Frameworks for Computed Tomography-based Abdominal Organ Auto-Segmentation.","authors":"Udbhav S Ram, Joel A Pogue, Michael Soike, Neil T Pfister, Rojymon Jacob, Carlos E Cardenas","doi":"10.1016/j.intonc.2025.03.003","DOIUrl":"10.1016/j.intonc.2025.03.003","url":null,"abstract":"<p><strong>Background: </strong>Segmentation of abdominal organs in computed tomography (CT) images within clinical oncological workflows is crucial for ensuring effective treatment planning and follow-up. However, manually generated segmentations are time-consuming and labor-intensive in addition to being subject to inter-observer variability. Many deep learning (DL) and Automated Machine Learning (AutoML) frameworks have emerged as a solution to this challenge and show promise in clinical workflows.</p><p><strong>Objective: </strong>This study presents a comprehensive evaluation of existing AutoML frameworks (Auto3DSeg, nnU-Net) against a state-of-the-art non-AutoML framework, the Shifted Window U-Net Transformer (SwinUNETR).</p><p><strong>Methods: </strong>Each framework was trained on the same 122 training images, taken from the Abdominal Multi-Organ Segmentation (AMOS) grand challenge. Frameworks were compared using Dice Similarity Coefficient (DSC), Surface DSC (sDSC), and 95th Percentile Hausdorff Distances (HD95) on an additional 72 holdout-validation images. The perceived clinical viability of 30 auto-contoured test cases was assessed by three physicians in a blinded evaluation.</p><p><strong>Results: </strong>Comparisons show significantly better performance by AutoML methods. nnU-Net (average DSC: 0.924, average sDSC: 0.938, average HD95: 4.26, median Likert: 4.57), Auto3DSeg (average DSC: 0.902, average sDSC: 0.919, average HD95: 8.76, median Likert: 4.49), and SwinUNETR (average DSC: 0.837, average sDSC: 0.844, average HD95: 13.93). AutoML frameworks were quantitatively preferred (13/13 OARs <i>P</i> >0.0.5 in DSC and sDSC, 12/13 OARs <i>P</i>>0.05 in HD95, comparing Auto3DSeg to SwinUNETR, and all OARs <i>P</i>>0.05 in all metrics comparing SwinUNETR to nnU-Net). Qualitatively, nnU-Net was preferred over Auto3DSeg (<i>P</i>=0.0027).</p><p><strong>Conclusion: </strong>The findings suggest that AutoML frameworks offer a significant advantage in the segmentation of abdominal organs, and underscores the potential of AutoML methods to enhance the efficiency of oncological workflows.</p>","PeriodicalId":521072,"journal":{"name":"Intelligent oncology","volume":"1 2","pages":"160-171"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12462693/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145188275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}