Basics of Deep Learning: A Radiologist's Guide to Understanding Published Radiology Articles on Deep Learning

Abstract

Artificial intelligence has been applied to many industries, including medicine. Among the various techniques in artificial intelligence, deep learning has attained the highest popularity in medical imaging in recent years. Many articles on deep learning have been published in radiologic journals. However, radiologists may have difficulty in understanding and interpreting these studies because the study methods of deep learning differ from those of traditional radiology. This review article aims to explain the concepts and terms that are frequently used in deep learning radiology articles, facilitating general radiologists' understanding.

Effect of Missing Data Imputation on Deep Learning Prediction Performance for Vesicoureteral Reflux and Recurrent Urinary Tract Infection Clinical Study

“…The number of iterations is equivalent to the number of batches needed to complete one epoch. When a dataset includes 500 cases split into minibatches of 50 cases, it will take 10 iterations to complete a single epoch [ 41 ].…”

A bird’s-eye view of deep learning in bioimage analysis

“…The impact of deep learning has been reviewed more specifically in a wide range of medical imaging areas, including abdominal imaging [103] , atherosclerosis imaging [104] , structural and functional brain imaging [105] , [106] , in-vivo cancer imaging [107] , dermatological imaging [108] , endoscopy [109] , mammography [110] , musculoskeletal imaging [111] , nuclear imaging [112] , ophthalmology [113] , pulmonary imaging [114] , thoracic imaging [115] , as well as in radiotherapy [116] , interventional radiology [117] , and radiology in general [118] , [119] , [120] . The massive body of papers on deep learning in virtually all areas of medical imaging has inspired many to write primers [121] , [122] , [123] , guides [124] , [125] , [126] , white papers or roadmaps [127] , [128] , [129] , and other commentaries [130] , [131] , [132] . There is now growing evidence that deep learning methods can perform on par with, if not better than, radiologists in specific tasks [133] , though the latter will continue to play a critical role in integrating such methods in clinical workflows [127] .…”

Inconsistency in the use of the term “validation” in studies reporting the performance of deep learning algorithms in providing diagnosis from medical imaging

“…Despite the notable difference between the tuning and testing steps, existing literature on DL show inconsistency in the use of the terminology “validation”, with some using it for the tuning step and others for the testing step [ 6 , 12 , 17 – 19 ]. Such inconsistency in terminology usage or inaccurate use of “validation” to refer to testing are likely due to the fact that the term is typically used in general communication as well as in medicine to refer to the testing of the accuracy of a completed algorithm [ 6 , 20 ], while the field of machine learning uses it as a very specific term that refers to the tuning step [ 4 – 6 , 12 , 17 , 19 , 21 ]. Also, the tuning step sometimes uses “cross-validation” procedures, which may create further confusion regarding the terminology for researchers who are less familiar with the methods and terms.…”