Citations

Clear

4 citation(s) found for "Hughes, Kevin".

Current state of mammography-based artificial intelligence for future breast cancer risk prediction: a systematic review
Lowry, Kathryn P. , Jeong, Han Eol , Kim, Ki Hwan , Hughes, Kevin S. , Lee, Christoph I. , Kerlikowske, Karla, Vachon, CM, Kerlikowske, Karla , Kerlikowske, Karla M , Vachon, Celine M. , Hughes, Kevin , Yala, Adam

Mayo Clinic

JNCI: Journal of the National Cancer Institute | January 6, 2026 | Volume: 118 | Issue: 3 | Pages: 392-403

10.1093/jnci/djag002 | PMID 41501990 Get It

Abstract Background There is growing interest in artificial intelligence (AI) models for predicting future breast cancer. We performed a systematic review of studies of mammography-based AI models for future breast cancer risk prediction to summarize current evidence, identify knowledge gaps, and inform future research directions. Methods We searched 6 databases for studies from January 1, 2012, to February 28, 2025, that evaluated mammography-based AI models for future breast cancer risk prediction. We extracted study design, participants’ race and ethnicity, geographic origin, mammogram type, vendor, prediction time frame, breast cancer type predicted, external validation, and exclusion of cancers diagnosed on the index screening mammogram. Areas under the receiver operating curve (AUCs) were summarized overall and by study characteristics. Results A total of 41 studies met our inclusion criteria. All studies were retrospective, and most used 2D mammograms (n = 37 studies) acquired using Hologic equipment (n = 25) and performed in the United States (n = 17); White, non-Hispanic women were most represented. Nearly all (n = 40) studies assessed discrimination performance with a median AUC of 0.71 for no longer than 2-year risk prediction, 0.72 for 3-4 years, and 0.71 for 5 years or more prediction. Median AUC was 0.75 for studies including index cancers vs 0.68 when excluded. Six studies reported model calibration performance ranging from good to overestimation of risk. Conclusion Future studies should evaluate models using digital breast tomosynthesis, examine performance for aggressive or advanced breast cancer, include diverse populations, and evaluate both discrimination and model calibration. Prospective evaluations are needed to determine the clinical utility of mammography-based AI models for personalized risk-based breast cancer screening before implementation.
Exploiting Rules to Enhance Machine Learning in Extracting Information From Multi-Institutional Prostate Pathology Reports
Santus, Enrico , Schuster, Tal , Tahmasebi, Amir , Li, Clara, Lanahan, Conor R., Prinsen, Peter , Thompson, Scott F., Coons, Samuel, Mynderse, Lance A. , Barzilay, Regina , Tahmasebi, Amir M , Mynderse, Lance, Hughes, Kevin , Yala, Adam , Hughes, Kevin S.

Mayo Clinic

JCO Clinical Cancer Informatics | October 2, 2020 | Volume: No volume available | Issue: 4 | Pages: 865-874

10.1200/cci.20.00028 | PMID 33006906 Get It

PURPOSE Literature on clinical note mining has highlighted the superiority of machine learning (ML) over hand-crafted rules. Nevertheless, most studies assume the availability of large training sets, which is rarely the case. For this reason, in the clinical setting, rules are still common. We suggest 2 methods to leverage the knowledge encoded in pre-existing rules to inform ML decisions and obtain high performance, even with scarce annotations. METHODS We collected 501 prostate pathology reports from 6 American hospitals. Reports were split into 2,711 core segments, annotated with 20 attributes describing the histology, grade, extension, and location of tumors. The data set was split by institutions to generate a cross-institutional evaluation setting. We assessed 4 systems, namely a rule-based approach, an ML model, and 2 hybrid systems integrating the previous methods: a Rule as Feature model and a Classifier Confidence model. Several ML algorithms were tested, including logistic regression (LR), support vector machine (SVM), and eXtreme gradient boosting (XGB). RESULTS When training on data from a single institution, LR lags behind the rules by 3.5% (F1 score: 92.2% v 95.7%). Hybrid models, instead, obtain competitive results, with Classifier Confidence outperforming the rules by +0.5% (96.2%). When a larger amount of data from multiple institutions is used, LR improves by +1.5% over the rules (97.2%), whereas hybrid systems obtain +2.2% for Rule as Feature (97.7%) and +2.6% for Classifier Confidence (98.3%). Replacing LR with SVM or XGB yielded similar performance gains. CONCLUSION We developed methods to use pre-existing handcrafted rules to inform ML algorithms. These hybrid systems obtain better performance than either rules or ML models alone, even when training data are limited.
American Society of Clinical Oncology Expert Statement: Collection and Use of a Cancer Family History for Oncology Providers
Lu, Karen H. , Wood, Marie , Daniels, Molly S. , Burke, Cathy, Ford, James M. , Kauff, Noah D. , Kohlmann, Wendy , Lindor, Noralane M. , Mulvey, Therese M. , Robinson, Linda, Rubinstein, Wendy S. , Snyder, Carrie, Merrill, Janette K. , Wollins, Dana S., Hughes, Kevin S. , Stoffel, Elena M. , Syngal, Sapna , Hughes, Kevin

Mayo Clinic

Journal of Clinical Oncology | February 4, 2014 | Volume: 32 | Issue: 8 | Pages: 833-840

10.1200/jco.2013.50.9257 | PMID 24493721 Get It

Future Research Directions in Acute Lung Injury: Summary of a National Heart, Lung, and Blood Institute Working Group
Matthay, Michael A. , Esmon, Charles, Bhattacharya, Jahar , Floros, Joanna, Gimbrone, Michael A., Hubmayr, Rolf D., Matalon, Sadis , Munford, Robert S., Parsons, Polly E., Slutsky, Arthur S. , Tracey, Kevin J. , Ward, Peter A. , Gail, Dorothy B., Harabin, Andrea L., Zimmerman, Guy A. , Esmon, Charles T., Coller, Barry S. , Doerschuk, Claire M. , Hoffman, Eric P. , Leppert, Mark

Mayo Clinic

American Journal of Respiratory and Critical Care Medicine | March 27, 2003 | Volume: 167 | Issue: 7 | Pages: 1027-1035

10.1164/rccm.200208-966ws | PMID 12663342 Get It

Abstract Acute lung injury (ALI) and its more severe form, the acute respiratory distress syndrome (ARDS), are syndromes of acute respiratory failure that result from acute pulmonary edema and inflammation. The development of ALI/ARDS is associated with several clinical disorders including direct pulmonary injury from pneumonia and aspiration as well as indirect pulmonary injury from trauma, sepsis, and other disorders such as acute pancreatitis and drug overdose. Although mortality from ALI/ARDS has decreased in the last decade, it remains high. Despite two major advances in treatment, low VT ventilation for ALI/ARDS and activated protein C for severe sepsis (the leading cause of ALI/ARDS), additional research is needed to develop specific treatments and improve understanding of the pathogenesis of these syndromes. The NHLBI convened a working group to develop specific recommendations for future ALI/ARDS research. Improved understanding of disease heterogeneity through use of evolving biologic, genomic, and genetic approaches should provide major new insights into pathogenesis of ALI. Cellular and molecular methods combined with animal and clinical studies should lead to further progress in the detection and treatment of this complex disease.