Deep-learning assistance closes the accuracy gap in fracture detection across clinician types

Imagen’s FDA-cleared AI software, FractureDetect, designed to detect fractures in musculoskeletal radiographs, improves clinicians’ accuracy at diagnosing fractures. Clinical testing showed that both clinicians with extensive training in musculoskeletal imaging and clinicians with limited training in musculoskeletal imaging were significantly more accurate when assisted by FractureDetect.

Deep neural network improves fracture detection by clinicians

Imagen developed an AI software designed to detect and localize fractures in wrist radiographs. A rigorous clinical study showed that the clinicians utilizing the AI software, referred to as a deep learning model, had a 47% relative reduction in misinterpretations. The significant improvement in diagnosing fractures shows that information provided by Imagen’s AI software can substantially enhance physician accuracy.

Assessment of a deep-learning system for fracture detection in musculoskeletal radiographs

Imagen developed an AI software designed to detect fractures across multiple anatomical areas throughout the musculoskeletal system. A multi-site study showed the AI software, referred to as a deep learning system, accurately identified fractures throughout the adult musculoskeletal system. The results were used in part to support FDA clearance of Imagen’s software (FractureDetect) which now assists clinicians in detecting fractures for the subset of the anatomic regions reported in this paper.

Reevaluation of missed lung cancer with artificial intelligence

Imagen’s FDA-cleared AI software, Chest-CAD, designed to detect suspicious regions of interest in chest radiographs, highlighted a region of a chest X-ray that was previously missed during unaided interpretation. This case shows that additional information provided by Imagen’s AI software was able to detect a life threatening finding.