Contribution of CT-Scan Analysis by Artificial Intelligence to the Clinical Care of TBI Patients

Abstract

The gold standard to diagnose intracerebral lesions after traumatic brain injury (TBI) is computed tomography (CT) scan, and due to its accessibility and improved quality of images, the global burden of CT scan for TBI patients is increasing. The recent developments of automated determination of traumatic brain lesions and medical-decision process using artificial intelligence (AI) represent opportunities to help clinicians in screening more patients, identifying the nature and volume of lesions and estimating the patient outcome. This short review will summarize what is ongoing with the use of AI and CT scan for patients with TBI.

Keywords: artificial intelligence; classification; computed tomography; review; segmentation; traumatic brain injury.

Figure 1

(A) Contribution of computed tomography (CT) scan analysis by artificial intelligence to the clinical care of traumatic brain injury (TBI) patients. References and terms are defined in Table 1. (B) Example of the use of artificial intelligence (AI) algorithms on clinical routine. CT scans of two patients (P1 and P2) at D0 were quantified with state of the art algorithms. On the right, CT scans of the same two patients acquired at D1 are shown. P1 and P2 had different clinical care. P1 underwent a decompressive craniectomy and not P2. Biggest extra axial hemorrhage (EAH) lesion was segmented with Brain Lesion Analysis and Segmentation Tool for Computed Tomography (BLAST-CT) (16) and radiomic metrics on this region of interest (ROI) were extracted as in (17). At first sight, the two lesions have the same profile, with equivalent volumes and means, but the variance of P1 is higher than twice the one of P2. That could for instance be a biomarker evaluated in further studies to predict the need for craniectomy. ICH, intracranial hemorrhage; GCS, Glasgow Coma Score; MLR, multivariate logical regression; RF, random forest; ANN, artificial neural network; GOS, Glasgow Outcome Score; CT scan, computed tomography image; Ref, References; HU, Hounsfield Units; ROI, region of interest; EAH, extra axial hemorrhage; D, day.