Fig. 3From: The value of deep learning-based computer aided diagnostic system in improving diagnostic performance of rib fractures in acute blunt traumaa A male patient (age between 40 and 45Â years old) injured in a traffic accident. The interns and attending radiologists all accurately diagnosed rib fracture in two rounds of reading. The diagnostic confidence scores were all 5 points. With the assistance of DL-CAD, the reading time was shortened from 95 and 59Â s to 45Â s and 20Â s, respectively. b A female patient (age between 30 and 35Â years old) with blunt chest trauma. The interns and attending radiologists missed the fracture in independent reading, and the diagnostic confidence score was 1 point. With the assistance of DL-CAD, the fracture was correctly diagnosed and the diagnostic confidence increased to 4 points. c A 30Â years old male patient with blunt on the left chest trauma, which is a false negative rib fracture image example. DL-CAD, the interns and the attending radiologists all diagnosed that the ribs were normal, but the left second external bone cortex was partially folded and warped (red arrow), and the senior radiologists diagnosed that the rib was a ORFBack to article page