Publications by authors named "Alois De La Comble"
Article Synopsis
- This study analyzes the accuracy of four convolutional neural networks (CNNs) in evaluating canine thoracic radiographs compared to 13 veterinary radiologists, using a set of 50 radiographic studies as a reference.
- The research established a gold standard through evaluations by three board-certified veterinary radiologists, focusing on 15 specific thoracic labels, and found that the CNNs generally performed similarly, with some variations based on training methods.
- Overall, the veterinary radiologists outperformed the CNNs, showing lower error rates, particularly for five of the 15 labels, though two CNNs did excel in identifying esophageal dilation, prompting further exploration into AI's role in veterinary radiology
Heart disease is a leading cause of death among cats and dogs. Vertebral heart scale (VHS) is one tool to quantify radiographic cardiac enlargement and to predict the occurrence of congestive heart failure. The aim of this study was to evaluate the performance of artificial intelligence (AI) performing VHS measurements when compared with two board-certified specialists.
View Article and Find Full Text PDFTo date, deep learning technologies have provided powerful decision support systems to radiologists in human medicine. The aims of this retrospective, exploratory study were to develop and describe an artificial intelligence able to screen thoracic radiographs for primary thoracic lesions in feline and canine patients. Three deep learning networks using three different pretraining strategies to predict 15 types of primary thoracic lesions were created (including tracheal collapse, left atrial enlargement, alveolar pattern, pneumothorax, and pulmonary mass).
View Article and Find Full Text PDF