Comparison and verification of two deep learning models for the detection of chest CT rib fractures

Abstract

Background A high false-positive rate remains a technical glitch hindering the broad spectrum of application of deep-learning-based diagnostic tools in routine radiological practice from assisting in diagnosing rib fractures. Purpose To examine the performance of two versions of deep-learning-based software tools in aiding radiologists in diagnosing rib fractures on chest computed tomography (CT) images. Material and Methods In total, 123 patients (708 rib fractures) were included in this retrospective study. Two groups of radiologists with different experience levels retrospectively reviewed images for rib fractures in the concurrent mode aided with RibFrac-High Sensitivity (HS) and RibFrac-High Precision (HP). We compared their diagnostic performance against the reference standard in terms of sensitivity and positive predictive value (PPV). Results On a per-patient basis, RibFrac-HS exhibited a higher sensitivity compared with RibFrac-HP (mean difference=0.051, 95% CI=0.012–0.090; P = 0.011), whereas the latter significantly outperformed the former in terms of the PPV (mean difference=0.273, 95% CI=0.238–0.308; P < 0.0001). The use of RibFrac-HP significantly improved the junior and the senior groups’ sensitivities respectively by 0.058 (95% CI=0.033–0.083; P < 0.0001) and 0.058 (95% CI=0.034–0.081; P < 0.0001), and decreased the diagnosis time by 206 s (95% CI=191–220; P < 0.0001) and 79 s (95% CI=67–92; P < 0.0001), respectively, when compared to no software assistance. Conclusion The sensitivity and efficiency of radiologists in identifying rib fractures can be improved by using RibFrac-HS and/or RibFrac-HP. With an added module for false-positive suppression, RibFrac-HP maintains the sensitivity and increases the PPV in fracture detection compared to Rib-Frac-HS.

FasterRib: A deep learning algorithm to automate identification and characterization of rib fractures on chest computed tomography scans

“…Sixth, this was an initial methodologic study, the first of many steps required to integrate a clinical decision support system. Previous studies have shown that fracture prediction models do improve radiologists' detection sensitivity and reading time 8,23–25 . FasterRib offers novel advances that were designed with clinician end users in mind, and we hope our open-source code can facilitate further model performance improvement and mitigate future implementation barriers.…”

“…8,[23][24][25] FasterRib offers novel advances that were designed with clinician end users in mind, and we hope our open-source code can facilitate further model performance improvement and mitigate future implementation barriers.…”

Feasibility of rib fracture detection in low-dose computed tomography images with a large, multicenter datasets-based model

“…Zhang et al [ 13 ] proposed a nnU-Net and DenseNet combination model, which achieved a sensitivity of 95 % for rib fractures and reduced the false-positive and false-negative rates for rib fracture detection to 5 %. Hong et al [ 14 ] used deep learning-based diagnostic tools to improve the sensitivity and efficiency in identifying rib fractures. Wu et al [ 15 ] developed a deep learning algorithm to detect rib fractures using multicenter CT datasets.…”