A rib fracture is a crack or break in the bones enclosing the chest (Figure 1). Rib fractures usually result from strong blunt force, such as a motor vehicle crash. CT is used as the main imaging modality for evaluation thoracic trauma, including rib fractures and associated complications. However, it takes time to all ribs and it is sometimes difficult to evaluate the locations and types of rib fractures at the emergency because a doctor must read hundreds of thin-slice CT images. In this paper, authors develop and validate an algorithm for the detection of acute rib fractures on thoracic CT images and to investigate the effect on radiologists’ performance. A three–dimensional faster region-based CNN was trained for rib fractures detection CNN (Figure 2). The observer performance study involved eight radiologists who first evaluated CT images without CNN output and then evaluated CT images with CNN output. The results have shown that CNN can improve the radiologists’ diagnostic performance regardless of the type of fractures and reader’s experience (Figure 3). Although further studies are needed to clarify the usefulness of CNN in actual clinical practice, it is expected that CNN will assist doctors and lead to more accurate treatment in emergencies.
 Azuma, M., Nakada, H., Takei, M. et al. Detection of acute rib fractures on CT images with convolutional neural networks: effect of location and type of fracture and reader’s experience. Emerg Radiol 29, 317–328 (2022).
CAUTION：This is Fujifilm Global Website. Fujifilm makes no representation that products on this website are commercially available in all countries. Approved uses of products vary by country and region. Specifications and appearance of products are subject to change without notice.