Radiology-Artificial Intelligence最新文献

{"title":"Let's Agree to Be Wrong: Transforming Alzheimer Disease Diagnosis.","authors":"Ilya Nasrallah","doi":"10.1148/ryai.250686","DOIUrl":"10.1148/ryai.250686","url":null,"abstract":"","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"7 6","pages":"e250686"},"PeriodicalIF":13.2,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145201639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiologist and AI Concordance in Screening Mammography and Association with Future Breast Cancer Risk.","authors":"Eun Young Kim, Eun Kyung Park, Mi-Ri Kwon, Minjeong Kim, Junhee Park, Jeonggyu Kang, Yoosun Cho, Sanghyup Lee, Minji Song, Ki Hwan Kim, Tae Soo Kim, Hyeonsoo Lee, Ria Kwon, Ga-Young Lim, JunHyeok Choi, Soo-Youn Ham, Shin Ho Kook, Yoosoo Chang, Seungho Ryu","doi":"10.1148/ryai.240804","DOIUrl":"10.1148/ryai.240804","url":null,"abstract":"<p><p>Purpose To evaluate the association between radiologist and stand-alone AI concordance in screening mammography interpretations and future breast cancer risk and to assess how breast density influences this relationship. Materials and Methods This retrospective study included Korean women who underwent digital screening mammography between January 2009 and December 2018. Incidental breast cancers, defined as those diagnosed within 1 year of screening, were excluded using the National Cancer Registry data. A commercial AI system retrospectively analyzed mammograms, with a 10% malignancy probability threshold. Patients were categorized into four groups based on AI-radiologist concordance: Concordant-Negative, Radiologist-Positive/AI-Negative, Radiologist-Negative/AI-Positive, and Concordant-Positive. Breast density was categorized using Breast Imaging Reporting and Data System classification. Cox proportional hazards models estimated adjusted hazard ratios (HRs) and 95% CIs. Results Over a median 7.3-year follow-up, 1011 breast cancers occurred among 82 899 women (mean age, 43.4 years ± 8.6 [SD]). Breast cancer incidence was the highest in the Concordant-Positive group (5-year cumulative incidence, 37.4 per 1000 person-years) and the lowest in the Concordant-Negative group (5.9 per 1000 person-years). Compared with the Concordant-Negative group, adjusted HRs for incidental breast cancer were 2.30 (<i>P</i> < .001) for the Radiologist-Negative/AI-Positive, 1.15 (<i>P</i> = .17) for the Radiologist-Positive/AI-Negative, and 4.51 (<i>P</i> < .001) for the Concordant-Positive groups. Risk increase in the Concordant-Positive group was consistent across breast densities; in dense breasts, elevated risk occurred only with positive AI. Conclusion Positive screening mammography findings identified by both radiologists and stand-alone AI (Concordant-Positive group) were associated with he highest 5-year breast cancer incidence of 3.74%, exceeding the 3% threshold for considering chemoprevention or supplemental imaging. <b>Keywords:</b> Mammography, Breast Neoplasms, Artificial Intelligence, Screening, Concordance <i>Supplemental material is available for this article.</i> © RSNA, 2025.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e240804"},"PeriodicalIF":13.2,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145293781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DLMUSE: Robust Brain Segmentation in Seconds Using Deep Learning.","authors":"Vishnu M Bashyam, Guray Erus, Yuhan Cui, Di Wu, Gyujoon Hwang, Alexander Getka, Ashish Singh, George Aidinis, Kyunglok Baik, Randa Melhem, Elizabeth Mamourian, Jimit Doshi, Ashwini Davison, Ilya M Nasrallah, Christos Davatzikos","doi":"10.1148/ryai.240299","DOIUrl":"10.1148/ryai.240299","url":null,"abstract":"<p><p>Purpose To introduce an open-source deep learning brain segmentation model for fully automated brain MRI segmentation, enabling rapid segmentation and facilitating large-scale neuroimaging research. Materials and Methods In this retrospective study, a deep learning model was developed using a diverse training dataset of 1900 MRI scans (patients aged 24-93 years, with a mean of 65 years ± 11.5 [SD]; 1007 female, 893 male) with reference labels generated using a multi-atlas segmentation method with human supervision. The final model was validated using 71 391 scans from 14 studies. Segmentation quality was assessed using Dice similarity and Pearson correlation coefficients with reference segmentations. Downstream predictive performance for brain age and Alzheimer disease was evaluated by fitting machine learning models. Statistical significance was assessed using Mann-Whitney <i>U</i> and McNemar tests. Results The DLMUSE model achieved high correlation (<i>r</i> = 0.93-0.95) and agreement (median Dice scores, 0.84-0.89) with reference segmentations across the testing dataset. Prediction of brain age using DLMUSE features achieved a mean absolute error of 5.08 years, similar to that of the reference method (5.15 years, <i>P</i> = .56). Classification of Alzheimer disease using DLMUSE features achieved an accuracy of 89% and F1 score of 0.80, which were comparable to values achieved by the reference method (89% and 0.79, respectively). DLMUSE segmentation speed was over 10 000 times faster than that of the reference method (3.5 seconds vs 14 hours). Conclusion DLMUSE enabled rapid brain MRI segmentation, with performance comparable to that of state-of-the-art methods across diverse datasets. The resulting open-source tools and user-friendly web interface can facilitate large-scale neuroimaging research and wide utilization of advanced segmentation methods. <b>Keywords:</b> Convolutional Neural Network (CNN), Deep Learning Algorithms, Machine Learning Algorithms, Segmentation, Application Domain, Supervised Learning, MRI, Brain/Brain Stem <i>Supplemental material is available for this article.</i> © RSNA, 2025.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e240299"},"PeriodicalIF":20.1,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12661376/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145076191","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}

{"title":"Segmenting Whole-Body MRI and CT for Multiorgan Anatomic Structure Delineation.","authors":"Hartmut Häntze, Lina Xu, Christian J Mertens, Felix J Dorfner, Leonhard Donle, Felix Busch, Avan Kader, Sebastian Ziegelmayer, Nadine Bayerl, Nassir Navab, Daniel Rueckert, Julia Schnabel, Hugo J W L Aerts, Daniel Truhn, Fabian Bamberg, Jakob Weiss, Christopher L Schlett, Steffen Ringhof, Thoralf Niendorf, Tobias Pischon, Hans-Ulrich Kauczor, Tobias Nonnenmacher, Thomas Kröncke, Henry Völzke, Jeanette Schulz-Menger, Klaus Maier-Hein, Alessa Hering, Mathias Prokop, Bram van Ginneken, Marcus R Makowski, Lisa C Adams, Keno K Bressem","doi":"10.1148/ryai.240777","DOIUrl":"10.1148/ryai.240777","url":null,"abstract":"<p><p>Purpose To develop and validate MRSegmentator, a retrospective cross-modality deep learning model for multiorgan segmentation of MRI scans. Materials and Methods This retrospective study trained MRSegmentator on 1200 manually annotated UK Biobank Dixon MRI sequences (50 participants), 221 in-house abdominal MRI sequences (177 patients), and 1228 CT scans from the TotalSegmentator CT dataset. A human-in-the-loop annotation workflow used cross-modality transfer learning from an existing CT segmentation model to segment 40 anatomic structures. The model's performance was evaluated on 900 MRI sequences from 50 participants in the German National Cohort (NAKO), 60 MRI sequences from the Multi-Modality Abdominal Multi-Organ Segmentation Challenge 2022 dataset, and 29 MRI sequences from TotalSegmentator MRI. Reference standard manual annotations were used for comparison. Metrics to assess segmentation quality included Dice similarity coefficient (DSC). Statistical analyses included organ- and sequence-specific means ± SDs reporting and two-sided <i>t</i> tests for demographic effects. Results One-hundred thirty-nine participants were evaluated; demographic information was available for 70 (mean age, 52.7 years ± 14.0 [SD], 36 female participants). Across all test datasets, MRSegmentator demonstrated high class-wise DSC for well-defined organs (lungs: 0.81-0.96, heart: 0.81-0.94) and organs with anatomic variability (liver: 0.82-0.96, kidneys: 0.77-0.95). Smaller structures showed lower DSC (portal or splenic veins: 0.64-0.78, adrenal glands: 0.56-0.69). The average DSC on the external testing using NAKO data ranged from 0.85 ± 0.08 for T2-weighted half-Fourier acquisition single-shot turbo spin echo to 0.91 ± 0.05 for in-phase sequences. The model generalized well to CT, achieving a mean DSC of 0.84 ± 0.12 on abdominal multiorgan segmentation CT data. Conclusion MRSegmentator accurately segmented 40 anatomic structures on MRI and generalized to CT, outperforming existing open-source tools. <b>Keywords:</b> MR-Imaging, Segmentation, Vision, Application Domain, Supervised Learning, Type of Machine Learning © The Author(s) 2025. Published by the Radiological Society of North America under a CC BY 4.0 license. See also commentary by Tao in this issue.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e240777"},"PeriodicalIF":13.2,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144790144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of Privacy-preserving Deep Learning Model with Homomorphic Encryption: A Technical Feasibility Study in Kidney CT Imaging.","authors":"Sang-Wook Lee, Jongmin Choi, Min-Je Park, Hajin Kim, Soo-Heang Eo, Garam Lee, Sulgi Kim, Jungyo Suh","doi":"10.1148/ryai.240798","DOIUrl":"10.1148/ryai.240798","url":null,"abstract":"<p><p>Purpose To evaluate the technical feasibility of implementing homomorphic encryption in deep learning models for privacy-preserving CT image analysis of kidney masses. Materials and Methods A privacy-preserving deep learning system was developed through three sequential technical phases: a reference convolutional neural network (CNN) model (Ref-CNN) based on ResNet architecture, modification for encryption compatibility (Approx-CNN) by replacing rectified linear unit with polynomial approximation and max-pooling with average-pooling, and implementation of fully homomorphic encryption (HE-CNN). The Cheon-Kim-Kim-Song encryption scheme was used for its capability to perform arithmetic operations on encrypted real numbers. Using 12 446 CT images from a public dataset (3709 images of renal cysts, 5077 images of normal kidneys, and 2283 images of kidney tumors), authors evaluated model performance using area under the receiver operating characteristic curve (AUC) and area under the precision-recall curve (AUPRC). Results All models demonstrated high diagnostic accuracy, with AUC ranging from 0.89 to 0.99 and AUPRC from 0.67 to 0.99. The diagnostic performance trade-off was minimal from Ref-CNN to Approx-CNN (AUC, 0.99-0.97 for normal category), with no evidence of differences between models. However, encryption substantially increased storage and computational demands: A 256 × 256-pixel image expanded from 65 KB to 32 MB, requiring 50 minutes for central processing unit inference but only 90 seconds with graphics processing unit acceleration. Conclusion This technical development demonstrates that privacy-preserving deep learning inference using homomorphic encryption is feasible for classifying kidney masses on CT images, achieving diagnostic performance similar to that of nonencrypted models while maintaining data privacy through end-to-end encryption. <b>Keywords:</b> Privacy-preserving AI, Homomorphic Encryption, Kidney Cancer <i>Supplemental material is available for this article.</i> © RSNA 2025. See also commentary by Busch and Adams in this issue.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e240798"},"PeriodicalIF":13.2,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144971947","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Privacy-preserving Deep Learning in Medical Imaging: Feasibility and Challenges.","authors":"Felix Busch, Lisa C Adams","doi":"10.1148/ryai.250682","DOIUrl":"10.1148/ryai.250682","url":null,"abstract":"","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"7 6","pages":"e250682"},"PeriodicalIF":13.2,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145252934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Completing the Baby Album: AI Synthesizing Infant Brain MRI for Missing Time Points.","authors":"Gunvant Chaudhari, Andreas Rauschecker","doi":"10.1148/ryai.250779","DOIUrl":"10.1148/ryai.250779","url":null,"abstract":"","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"7 6","pages":"e250779"},"PeriodicalIF":13.2,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145606000","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Distinguishing between Rigor and Transparency in FDA Marketing Authorization of AI-enabled Medical Devices.","authors":"Abdul Rahman Diab, William Lotter","doi":"10.1148/ryai.250369","DOIUrl":"10.1148/ryai.250369","url":null,"abstract":"<p><p>The increasing prevalence of artificial intelligence (AI)-enabled medical devices presents significant opportunities for improving patient outcomes. However, recent studies based on public U.S. Food and Drug Administration (FDA) summaries have raised concerns about the extent of validation that such devices undergo before FDA marketing authorization and subsequent clinical deployment. Here, the authors clarify key concepts of FDA regulation and provide insights into the current standards of performance validation, focusing on radiology AI devices. The authors distinguish between two fundamentally different but often conflated concepts: validation rigor (ie, the quality and comprehensiveness of the evidence supporting a device's performance) and validation transparency (ie, the extent to which this evidence is publicly accessible). The authors begin by describing the inverse relationship between the amount of performance data contained and the transparency of specific components of an FDA submission. Drawing on FDA guidelines and on experience developing authorized AI devices, the authors outline current validation standards and present a mapping from common radiology AI device types to their typical clinical study designs. This article concludes with actionable recommendations, advocating for a balanced approach tailored to specific use cases while still enforcing certain universal standards. These measures will help ensure that AI-enabled medical devices are both rigorously evaluated and transparently reported, thereby fostering greater public trust and enhancing clinical utility.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e250369"},"PeriodicalIF":13.2,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145132123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bridging Radiologic Reasoning and Artificial Intelligence: Explainable Deep Learning for Focal Liver Lesions.","authors":"Lisa C Adams, Keno K Bressem","doi":"10.1148/ryai.250806","DOIUrl":"10.1148/ryai.250806","url":null,"abstract":"","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"7 6","pages":"e250806"},"PeriodicalIF":13.2,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145446158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Characterizing the Impact of Training Data on Generalizability: Application in Deep Learning to Estimate Lung Nodule Malignancy Risk.","authors":"Bogdan Obreja, Joeran Bosma, Kiran Vaidhya Venkadesh, Zaigham Saghir, Mathias Prokop, Colin Jacobs","doi":"10.1148/ryai.240636","DOIUrl":"10.1148/ryai.240636","url":null,"abstract":"<p><p>Purpose To investigate the relationship between training data volume and performance of a deep learning artificial intelligence (AI) algorithm developed to assess the malignancy risk of pulmonary nodules detected on low-dose CT scans in lung cancer screening. Materials and Methods This retrospective study used a dataset of 16 077 annotated nodules (1249 malignant, 14 828 benign) from the National Lung Screening Trial (NLST) to systematically train an AI algorithm for pulmonary nodule malignancy risk prediction across various stratified subsets ranging from 1.25% to the full dataset. External testing was conducted using data from the Danish Lung Cancer Screening Trial (DLCST) to determine the amount of training data at which the performance of the AI was statistically noninferior to the AI trained on the full NLST cohort. A size-matched cancer-enriched subset of DLCST, in which each malignant nodule had been paired in diameter with the closest two benign nodules, was used to investigate the amount of training data at which the performance of the AI algorithm was statistically noninferior to the average performance of 11 clinicians. Results The external testing set included 599 participants (mean age ± SD, 57.65 years ± 4.84 for female participants and 59.03 years ± 4.94 for male participants) with 883 nodules (65 malignant, 818 benign). The AI achieved a mean area under the receiver operating characteristic curve (AUC) of 0.92 (95% CI: 0.88, 0.96) on the DLCST cohort when trained on the full NLST dataset. Training with 80% of the NLST data resulted in noninferior performance (mean AUC, 0.92; 95% CI: 0.89, 0.96; <i>P</i> = .005). On the size-matched DLCST subset (59 malignant, 118 benign), the AI reached noninferior clinician-level performance (mean AUC, 0.82; 95% CI: 0.77, 0.86) with 20% of the training data (<i>P</i> = .02). Conclusion The deep learning AI algorithm demonstrated excellent performance in assessing pulmonary nodule malignancy risk, achieving clinical level performance with a fraction of the training data and reaching peak performance before using the full dataset. <b>Keywords:</b> Convolutional Neural Network (CNN), CT, Lung, Screening, Diagnosis, Supervised Learning, Lung Cancer Screening, Pulmonary Nodule Malignancy Risk, Deep Learning, Pulmonary Nodule Management <i>Supplemental material is available for this article.</i> © RSNA, 2025 See also commentary by Archer in this issue.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e240636"},"PeriodicalIF":13.2,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144971996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}