Radiology-Artificial Intelligence最新文献

{"title":"Semi-supervised Learning for Generalizable Intracranial Hemorrhage Detection and Segmentation.","authors":"Emily Lin, Esther L Yuh","doi":"10.1148/ryai.230077","DOIUrl":"10.1148/ryai.230077","url":null,"abstract":"<p><p>Purpose To develop and evaluate a semi-supervised learning model for intracranial hemorrhage detection and segmentation on an out-of-distribution head CT evaluation set. Materials and Methods This retrospective study used semi-supervised learning to bootstrap performance. An initial \"teacher\" deep learning model was trained on 457 pixel-labeled head CT scans collected from one U.S. institution from 2010 to 2017 and used to generate pseudo labels on a separate unlabeled corpus of 25 000 examinations from the Radiological Society of North America and American Society of Neuroradiology. A second \"student\" model was trained on this combined pixel- and pseudo-labeled dataset. Hyperparameter tuning was performed on a validation set of 93 scans. Testing for both classification (<i>n</i> = 481 examinations) and segmentation (<i>n</i> = 23 examinations, or 529 images) was performed on CQ500, a dataset of 481 scans performed in India, to evaluate out-of-distribution generalizability. The semi-supervised model was compared with a baseline model trained on only labeled data using area under the receiver operating characteristic curve, Dice similarity coefficient, and average precision metrics. Results The semi-supervised model achieved a statistically significant higher examination area under the receiver operating characteristic curve on CQ500 compared with the baseline (0.939 [95% CI: 0.938, 0.940] vs 0.907 [95% CI: 0.906, 0.908]; <i>P</i> = .009). It also achieved a higher Dice similarity coefficient (0.829 [95% CI: 0.825, 0.833] vs 0.809 [95% CI: 0.803, 0.812]; <i>P</i> = .012) and pixel average precision (0.848 [95% CI: 0.843, 0.853]) vs 0.828 [95% CI: 0.817, 0.828]) compared with the baseline. Conclusion The addition of unlabeled data in a semi-supervised learning framework demonstrates stronger generalizability potential for intracranial hemorrhage detection and segmentation compared with a supervised baseline. <b>Keywords:</b> Semi-supervised Learning, Traumatic Brain Injury, CT, Machine Learning <i>Supplemental material is available for this article.</i> Published under a CC BY 4.0 license. See also the commentary by Swimburne in this issue.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e230077"},"PeriodicalIF":9.8,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140498/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140040505","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":"Faster, More Practical, but Still Accurate: Deep Learning for Diagnosis of Progressive Supranuclear Palsy.","authors":"Bahram Mohajer","doi":"10.1148/ryai.240181","DOIUrl":"10.1148/ryai.240181","url":null,"abstract":"","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"6 3","pages":"e240181"},"PeriodicalIF":9.8,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140513/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140858206","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":"A Semiautonomous Deep Learning System to Reduce False Positives in Screening Mammography.","authors":"Stefano Pedemonte, Trevor Tsue, Brent Mombourquette, Yen Nhi Truong Vu, Thomas Matthews, Rodrigo Morales Hoil, Meet Shah, Nikita Ghare, Naomi Zingman-Daniels, Susan Holley, Catherine M Appleton, Jason Su, Richard L Wahl","doi":"10.1148/ryai.230033","DOIUrl":"10.1148/ryai.230033","url":null,"abstract":"<p><p>Purpose To evaluate the ability of a semiautonomous artificial intelligence (AI) model to identify screening mammograms not suspicious for breast cancer and reduce the number of false-positive examinations. Materials and Methods The deep learning algorithm was trained using 123 248 two-dimensional digital mammograms (6161 cancers) and a retrospective study was performed on three nonoverlapping datasets of 14 831 screening mammography examinations (1026 cancers) from two U.S. institutions and one U.K. institution (2008-2017). The stand-alone performance of humans and AI was compared. Human plus AI performance was simulated to examine reductions in the cancer detection rate, number of examinations, false-positive callbacks, and benign biopsies. Metrics were adjusted to mimic the natural distribution of a screening population, and bootstrapped CIs and <i>P</i> values were calculated. Results Retrospective evaluation on all datasets showed minimal changes to the cancer detection rate with use of the AI device (noninferiority margin of 0.25 cancers per 1000 examinations: U.S. dataset 1, <i>P</i> = .02; U.S. dataset 2, <i>P</i> < .001; U.K. dataset, <i>P</i> < .001). On U.S. dataset 1 (11 592 mammograms; 101 cancers; 3810 female patients; mean age, 57.3 years ± 10.0 [SD]), the device reduced screening examinations requiring radiologist interpretation by 41.6% (95% CI: 40.6%, 42.4%; <i>P</i> < .001), diagnostic examinations callbacks by 31.1% (95% CI: 28.7%, 33.4%; <i>P</i> < .001), and benign needle biopsies by 7.4% (95% CI: 4.1%, 12.4%; <i>P</i> < .001). U.S. dataset 2 (1362 mammograms; 330 cancers; 1293 female patients; mean age, 55.4 years ± 10.5) was reduced by 19.5% (95% CI: 16.9%, 22.1%; <i>P</i> < .001), 11.9% (95% CI: 8.6%, 15.7%; <i>P</i> < .001), and 6.5% (95% CI: 0.0%, 19.0%; <i>P</i> = .08), respectively. The U.K. dataset (1877 mammograms; 595 cancers; 1491 female patients; mean age, 63.5 years ± 7.1) was reduced by 36.8% (95% CI: 34.4%, 39.7%; <i>P</i> < .001), 17.1% (95% CI: 5.9%, 30.1%: <i>P</i> < .001), and 5.9% (95% CI: 2.9%, 11.5%; <i>P</i> < .001), respectively. Conclusion This work demonstrates the potential of a semiautonomous breast cancer screening system to reduce false positives, unnecessary procedures, patient anxiety, and medical expenses. <b>Keywords:</b> Artificial Intelligence, Semiautonomous Deep Learning, Breast Cancer, Screening Mammography <i>Supplemental material is available for this article.</i> Published under a CC BY 4.0 license.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e230033"},"PeriodicalIF":9.8,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140506/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140872447","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":"Erratum for: Performance of the Winning Algorithms of the RSNA 2022 Cervical Spine Fracture Detection Challenge.","authors":"Ghee Rye Lee, Adam E Flanders, Tyler Richards, Felipe Kitamura, Errol Colak, Hui Ming Lin, Robyn L Ball, Jason Talbott, Luciano M Prevedello","doi":"10.1148/ryai.249002","DOIUrl":"10.1148/ryai.249002","url":null,"abstract":"","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"6 3","pages":"e249002"},"PeriodicalIF":9.8,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140496/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140865979","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":"Artificial Intelligence for Breast Cancer Screening: Trade-offs between Sensitivity and Specificity.","authors":"Manisha Bahl, Synho Do","doi":"10.1148/ryai.240184","DOIUrl":"10.1148/ryai.240184","url":null,"abstract":"","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"6 3","pages":"e240184"},"PeriodicalIF":8.1,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140497/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140877471","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":"Noninvasive Molecular Subtyping of Pediatric Low-Grade Glioma with Self-Supervised Transfer Learning.","authors":"Divyanshu Tak, Zezhong Ye, Anna Zapaischykova, Yining Zha, Aidan Boyd, Sridhar Vajapeyam, Rishi Chopra, Hasaan Hayat, Sanjay P Prabhu, Kevin X Liu, Hesham Elhalawani, Ali Nabavizadeh, Ariana Familiar, Adam C Resnick, Sabine Mueller, Hugo J W L Aerts, Pratiti Bandopadhayay, Keith L Ligon, Daphne A Haas-Kogan, Tina Y Poussaint, Benjamin H Kann","doi":"10.1148/ryai.230333","DOIUrl":"10.1148/ryai.230333","url":null,"abstract":"<p><p>Purpose To develop and externally test a scan-to-prediction deep learning pipeline for noninvasive, MRI-based <i>BRAF</i> mutational status classification for pediatric low-grade glioma. Materials and Methods This retrospective study included two pediatric low-grade glioma datasets with linked genomic and diagnostic T2-weighted MRI data of patients: Dana-Farber/Boston Children's Hospital (development dataset, <i>n</i> = 214 [113 (52.8%) male; 104 (48.6%) <i>BRAF</i> wild type, 60 (28.0%) <i>BRAF</i> fusion, and 50 (23.4%) <i>BRAF</i> V600E]) and the Children's Brain Tumor Network (external testing, <i>n</i> = 112 [55 (49.1%) male; 35 (31.2%) <i>BRAF</i> wild type, 60 (53.6%) <i>BRAF</i> fusion, and 17 (15.2%) <i>BRAF</i> V600E]). A deep learning pipeline was developed to classify <i>BRAF</i> mutational status (<i>BRAF</i> wild type vs <i>BRAF</i> fusion vs <i>BRAF</i> V600E) via a two-stage process: <i>(a)</i> three-dimensional tumor segmentation and extraction of axial tumor images and <i>(b)</i> section-wise, deep learning-based classification of mutational status. Knowledge-transfer and self-supervised approaches were investigated to prevent model overfitting, with a primary end point of the area under the receiver operating characteristic curve (AUC). To enhance model interpretability, a novel metric, center of mass distance, was developed to quantify the model attention around the tumor. Results A combination of transfer learning from a pretrained medical imaging-specific network and self-supervised label cross-training (TransferX) coupled with consensus logic yielded the highest classification performance with an AUC of 0.82 (95% CI: 0.72, 0.91), 0.87 (95% CI: 0.61, 0.97), and 0.85 (95% CI: 0.66, 0.95) for <i>BRAF</i> wild type, <i>BRAF</i> fusion, and <i>BRAF</i> V600E, respectively, on internal testing. On external testing, the pipeline yielded an AUC of 0.72 (95% CI: 0.64, 0.86), 0.78 (95% CI: 0.61, 0.89), and 0.72 (95% CI: 0.64, 0.88) for <i>BRAF</i> wild type, <i>BRAF</i> fusion, and <i>BRAF</i> V600E, respectively. Conclusion Transfer learning and self-supervised cross-training improved classification performance and generalizability for noninvasive pediatric low-grade glioma mutational status prediction in a limited data scenario. <b>Keywords:</b> Pediatrics, MRI, CNS, Brain/Brain Stem, Oncology, Feature Detection, Diagnosis, Supervised Learning, Transfer Learning, Convolutional Neural Network (CNN) <i>Supplemental material is available for this article.</i> © RSNA, 2024.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e230333"},"PeriodicalIF":8.1,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140508/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140040504","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":"Impact of Deep Learning Image Reconstruction Methods on MRI Throughput.","authors":"Anthony Yang, Mark Finkelstein, Clara Koo, Amish H Doshi","doi":"10.1148/ryai.230181","DOIUrl":"10.1148/ryai.230181","url":null,"abstract":"<p><p>Purpose To evaluate the effect of implementing two distinct commercially available deep learning reconstruction (DLR) algorithms on the efficiency of MRI examinations conducted in real clinical practice within an outpatient setting at a large, multicenter institution. Materials and Methods This retrospective study included 7346 examinations from 10 clinical MRI scanners analyzed during the pre- and postimplementation periods of DLR methods. Two different types of DLR methods, namely Digital Imaging and Communications in Medicine (DICOM)-based and k-space-based methods, were implemented in half of the scanners (three DICOM-based and two k-space-based), while the remaining five scanners had no DLR method implemented. Scan and room times of each examination type during the pre- and postimplementation periods were compared among the different DLR methods using the Wilcoxon test. Results The application of deep learning methods resulted in significant reductions in scan and room times for certain examination types. The DICOM-based method demonstrated up to a 53% reduction in scan times and a 41% reduction in room times for various study types. The k-space-based method demonstrated up to a 27% reduction in scan times but did not significantly reduce room times. Conclusion DLR methods were associated with reductions in scan and room times in a clinical setting, though the effects were heterogeneous depending on examination type. Thus, potential adopters should carefully evaluate their case mix to determine the impact of integrating these tools. <b>Keywords:</b> Deep Learning MRI Reconstruction, Reconstruction Algorithms, DICOM-based Reconstruction, k-Space-based Reconstruction © RSNA, 2024 See also the commentary by GharehMohammadi in this issue.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e230181"},"PeriodicalIF":9.8,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140511/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140176775","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":"Bone Age Prediction under Stress.","authors":"Shahriar Faghani, Bradley J Erickson","doi":"10.1148/ryai.240137","DOIUrl":"10.1148/ryai.240137","url":null,"abstract":"","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"6 3","pages":"e240137"},"PeriodicalIF":9.8,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140503/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140862083","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":"Lessons Learned in Building Expertly Annotated Multi-Institution Datasets and Hosting the RSNA AI Challenges.","authors":"Felipe C Kitamura, Luciano M Prevedello, Errol Colak, Safwan S Halabi, Matthew P Lungren, Robyn L Ball, Jayashree Kalpathy-Cramer, Charles E Kahn, Tyler Richards, Jason F Talbott, George Shih, Hui Ming Lin, Katherine P Andriole, Maryam Vazirabad, Bradley J Erickson, Adam E Flanders, John Mongan","doi":"10.1148/ryai.230227","DOIUrl":"10.1148/ryai.230227","url":null,"abstract":"<p><p>The Radiological Society of North America (RSNA) has held artificial intelligence competitions to tackle real-world medical imaging problems at least annually since 2017. This article examines the challenges and processes involved in organizing these competitions, with a specific emphasis on the creation and curation of high-quality datasets. The collection of diverse and representative medical imaging data involves dealing with issues of patient privacy and data security. Furthermore, ensuring quality and consistency in data, which includes expert labeling and accounting for various patient and imaging characteristics, necessitates substantial planning and resources. Overcoming these obstacles requires meticulous project management and adherence to strict timelines. The article also highlights the potential of crowdsourced annotation to progress medical imaging research. Through the RSNA competitions, an effective global engagement has been realized, resulting in innovative solutions to complex medical imaging problems, thus potentially transforming health care by enhancing diagnostic accuracy and patient outcomes. <b>Keywords:</b> Use of AI in Education, Artificial Intelligence © RSNA, 2024.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e230227"},"PeriodicalIF":9.8,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140499/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140111550","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":"Impact of AI for Digital Breast Tomosynthesis on Breast Cancer Detection and Interpretation Time.","authors":"Eun Kyung Park, SooYoung Kwak, Weonsuk Lee, Joon Suk Choi, Thijs Kooi, Eun-Kyung Kim","doi":"10.1148/ryai.230318","DOIUrl":"10.1148/ryai.230318","url":null,"abstract":"<p><p>Purpose To develop an artificial intelligence (AI) model for the diagnosis of breast cancer on digital breast tomosynthesis (DBT) images and to investigate whether it could improve diagnostic accuracy and reduce radiologist reading time. Materials and Methods A deep learning AI algorithm was developed and validated for DBT with retrospectively collected examinations (January 2010 to December 2021) from 14 institutions in the United States and South Korea. A multicenter reader study was performed to compare the performance of 15 radiologists (seven breast specialists, eight general radiologists) in interpreting DBT examinations in 258 women (mean age, 56 years ± 13.41 [SD]), including 65 cancer cases, with and without the use of AI. Area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and reading time were evaluated. Results The AUC for stand-alone AI performance was 0.93 (95% CI: 0.92, 0.94). With AI, radiologists' AUC improved from 0.90 (95% CI: 0.86, 0.93) to 0.92 (95% CI: 0.88, 0.96) (<i>P</i> = .003) in the reader study. AI showed higher specificity (89.64% [95% CI: 85.34%, 93.94%]) than radiologists (77.34% [95% CI: 75.82%, 78.87%]) (<i>P</i> < .001). When reading with AI, radiologists' sensitivity increased from 85.44% (95% CI: 83.22%, 87.65%) to 87.69% (95% CI: 85.63%, 89.75%) (<i>P</i> = .04), with no evidence of a difference in specificity. Reading time decreased from 54.41 seconds (95% CI: 52.56, 56.27) without AI to 48.52 seconds (95% CI: 46.79, 50.25) with AI (<i>P</i> < .001). Interreader agreement measured by Fleiss κ increased from 0.59 to 0.62. Conclusion The AI model showed better diagnostic accuracy than radiologists in breast cancer detection, as well as reduced reading times. The concurrent use of AI in DBT interpretation could improve both accuracy and efficiency. <b>Keywords:</b> Breast, Computer-Aided Diagnosis (CAD), Tomosynthesis, Artificial Intelligence, Digital Breast Tomosynthesis, Breast Cancer, Computer-Aided Detection, Screening <i>Supplemental material is available for this article.</i> © RSNA, 2024 See also the commentary by Bae in this issue.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e230318"},"PeriodicalIF":9.8,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140510/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140858350","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}