Understanding artificial intelligence based radiology studies: What is overfitting?

• Published in: Clinical imaging Vol. 65; pp. 96 - 99
• Main Authors: Mutasa, Simukayi, Sun, Shawn, Ha, Richard
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.09.2020; Elsevier Limited
• Subjects: Algorithms; Animal training; Artificial Intelligence; Cancer; Datasets; Deep learning; Human performance; Humans; Machine learning; Medical imaging; Medical research; Medical screening; Neural networks; Overfitting; Population; Radiologists; Radiology; Radiology - methods; Researchers; Artificial intelligence; Machine learning; Overfitting
• ISSN: 0899-7071; 1873-4499; 1873-4499
• DOI: 10.1016/j.clinimag.2020.04.025

Artificial intelligence (AI) is a broad umbrella term used to encompass a wide variety of subfields dedicated to creating algorithms to perform tasks that mimic human intelligence. As AI development grows closer to clinical integration, radiologists will need to become familiar with the principles of artificial intelligence to properly evaluate and use this powerful tool. This series aims to explain certain basic concepts of artificial intelligence, and their applications in medical imaging starting with a concept of overfitting. •This series aims to explain basic concepts of artificial intelligence (AI), and its applications in medical imaging.•Overfitting means that an AI model has learned in a manner that is mainly applicable to the training data.•Overfitting is a major obstacle for AI technology hindering its generalizability to the overall population.•Overfitting can be minimized by a large training dataset, data augmentation, or techniques such as regularization and dropout.•Before AI algorithms can be incorporated clinically, external validation will be necessary to ensure generalizability.