A comparison of performance between a deep learning model with residents for localization and classification of intracranial hemorrhage

Abstract

Intracranial hemorrhage (ICH) from traumatic brain injury (TBI) requires prompt radiological investigation and recognition by physicians. Computed tomography (CT) scanning is the investigation of choice for TBI and has become increasingly utilized under the shortage of trained radiology personnel. It is anticipated that deep learning models will be a promising solution for the generation of timely and accurate radiology reports. Our study examines the diagnostic performance of a deep learning model and compares the performance of that with detection, localization and classification of traumatic ICHs involving radiology, emergency medicine, and neurosurgery residents. Our results demonstrate that the high level of accuracy achieved by the deep learning model, (0.89), outperforms the residents with regard to sensitivity (0.82) but still lacks behind in specificity (0.90). Overall, our study suggests that the deep learning model may serve as a potential screening tool aiding the interpretation of head CT scans among traumatic brain injury patients.

Radiomics and Deep Features: Robust Classification of Brain Hemorrhages and Reproducibility Analysis Using a 3D Autoencoder Neural Network

“…While Lee et al's [33] work showcases an algorithmic approach, our study complements this by demonstrating the broader spectrum of analysis possible in the field of AI and medicine. In the study conducted by Angkurawaranon et al [34], the focus lies on the critical realm of ICH stemming from traumatic brain injury (TBI). Their investigation underscores the urgency of timely radiological assessment and physician recognition in such scenarios.…”

“…The advent of CT scanning, chosen as the investigation method of preference for TBI cases, gains prominence due to the scarcity of trained radiology personnel. Against this backdrop, Angkurawaranon et al [34] delved into the potential of deep learning models as a promising avenue to expedite the generation of accurate radiology reports. Their study not only evaluates the diagnostic proficiency of a deep learning model but also contrasts its performance with the abilities of radiology, emergency medicine, and neurosurgery residents in detecting, localizing, and classifying traumatic ICHs.…”

“…Furthermore, ICC's adaptability and versatility make it a valuable asset in radiomics research [25]. It transcends the constraints of measurement scales and data types, enabling its application to a wide range of radiomic features and datasets [26,28,[30][31][32][33][34]. This adaptability is invaluable in handling the inherent diversity of radiomics data.…”