Dense attentive GAN-based one-class model for detection of autism and ADHD

Diagnostic Performance of Artificial Intelligence-Based Computer-Aided Detection and Diagnosis in Pediatric Radiology: A Systematic Review

“…Twenty-three articles met the selection criteria and were included in this review [ 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 ]. Table 1 shows their AI-based CAD application areas in the pediatric radiology and the diagnostic performances.…”

“…Table 1 shows their AI-based CAD application areas in the pediatric radiology and the diagnostic performances. These studies covered brain ( n = 9) [ 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 ], respiratory ( n = 9) [ 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 ], musculoskeletal ( n = 2) [ 40 , 41 ], urologic ( n = 2) [ 51 , 52 ] and cardiac imaging ( n = 1) [ 39 ]. The commonest AI-based CAD application area (30.4%, 7/23) was pediatric pneumonia [ 43 , 45 , 46 , 47 , 48 , 49 , 50 ].…”

“…The commonest AI-based CAD application area (30.4%, 7/23) was pediatric pneumonia [ 43 , 45 , 46 , 47 , 48 , 49 , 50 ]. No study reported all seven diagnostic accuracy measures [ 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 ]. Most commonly, the papers (30.4%, 7/23) reported four metrics [ 30 , 32 , 35 , 42 , 44 , 45 , 52 ].…”

“…Most commonly, the papers (30.4%, 7/23) reported four metrics [ 30 , 32 , 35 , 42 , 44 , 45 , 52 ]. Accuracy ( n = 19) and sensitivity ( n = 18) were the two most frequently used evaluation metrics [ 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 ]. One study only used one measure, AUC [ 40 ].…”

“…Most of the articles (93.3%, 14/15; 77.8%, 14/18; 73.3%, 11/15; 80.0%, 8/10; 66.6%, 2/3; 84.2%, 16/19; 80.0%, 8/10) reported AI-based CAD model performances of at least 0.83 (AUC), 0.84 (sensitivity), 0.80 (specificity), 0.89 (PPV), 0.63 (NPV), 0.87 (accuracy), and 0.82 (F1 score), respectively. The ranges of the reported performance values were 0.698–0.999 (AUC), 0.420–0.987 (sensitivity), 0.585–1.000 (specificity), 0.600–1.000 (PPV), 0.260–0.971 (NPV), 0.643–0.986 (accuracy), and 0.626–0.983 (F1 score) [ 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 ]. For the seven studies about AI-based CAD for pneumonia, their model performances were at least 0.850 (AUC), 0.760 (sensitivity), 0.800 (specificity), 0.891 (PPV), 0.905 (accuracy) and 0.903 (F1 score).…”