Publications by authors named "Sebastiaan Broos"
Bone Joint J
November 2024
Article Synopsis
- - This study aimed to create a convolutional neural network (CNN) to detect and classify fractures, focusing on specific characteristics like greater tuberosity displacement and neck-shaft angle, using plain X-rays.
- - The CNN was trained with over 1,700 X-rays from Australia and validated with data from the Netherlands, comparing results with CT scans evaluated by experts.
- - The CNN demonstrated a high detection accuracy of 94% for fractures, but less effectiveness in identifying specific fracture characteristics, particularly showing lower performance for greater tuberosity displacement and neck-shaft angles.
