Automated Quantification of Brain Lesion Volume From Post-trauma MR Diffusion-Weighted Images.

Thomas Mistral,Alan Tucholka,Irène Troprès,Stéphane Kremer,Adrian Kastler,Emmanuel L Barbier,Damien Galanaud,Christophe Maggia,Alexandre Krainik,Jean-François Payen,Michel Dojat,Senan Doyle,Florence Forbes,Pauline Roca,Emmanuelle Schmitt

doi:10.3389/fneur.2021.740603

Abstract

ObjectivesDetermining the volume of brain lesions after trauma is challenging. Manual delineation is observer-dependent and time-consuming and cannot therefore be used in routine practice. The study aimed to evaluate the feasibility of an automated atlas-based quantification procedure (AQP) based on the detection of abnormal mean diffusivity (MD) values computed from diffusion-weighted MR images.MethodsThe performance of AQP was measured against manual delineation consensus by independent raters in two series of experiments based on: (i) realistic trauma phantoms (n = 5) where low and high MD values were assigned to healthy brain images according to the intensity, form and location of lesion observed in real TBI cases; (ii) severe TBI patients (n = 12 patients) who underwent MR imaging within 10 days after injury.ResultsIn realistic TBI phantoms, no statistical differences in Dice similarity coefficient, precision and brain lesion volumes were found between AQP, the rater consensus and the ground truth lesion delineations. Similar findings were obtained when comparing AQP and manual annotations for TBI patients. The intra-class correlation coefficient between AQP and manual delineation was 0.70 in realistic phantoms and 0.92 in TBI patients. The volume of brain lesions detected in TBI patients was 59 ml (19–84 ml) (median; 25–75th centiles).ConclusionsOur results support the feasibility of using an automated quantification procedure to determine, with similar accuracy to manual delineation, the volume of low and high MD brain lesions after trauma, and thus allow the determination of the type and volume of edematous brain lesions. This approach had comparable performance with manual delineation by a panel of experts. It will be tested in a large cohort of patients enrolled in the multicenter OxyTC trial (NCT02754063).

Highlights

Traumatic brain injury (TBI) remains a leading cause of death and disability among individuals
automatic quantification procedure (AQP) could detect additional lesions undetected by manual delineation and present in ground truth (GT), and could exclude image artifacts
Dice and precision showed no significant difference between manual delineation and AQP (Dice: 0.75 and 0.72; and Precision: 0.66 and 0.70, respectively) (Table 1)

Summary

Introduction

Traumatic brain injury (TBI) remains a leading cause of death and disability among individuals. Predicting neurological outcome after severe TBI is challenging due to the complexity of the traumatic lesion, its evolution over time, and the number of external factors that may affect the outcome. Data are very limited concerning the use of automated methods to quantify brain injury post-trauma [3, 4]. Intracranial blood in the brain tissue, the presence of cerebrospinal fluid (CSF) and the heterogeneity of brain tissue injury make the segmentation of traumatic brain lesions challenging. Automated approaches using non-contrast CT imaging were developed for cranial cavity segmentation [5], cistern segmentation or detection of intracranial hematomas [6]. More intracranial lesions (e.g., brain swelling or intracranial hemorrhage) can be detected by MRI, due to its higher sensitivity [2]

Objectives

Methods

Results

Discussion

Conclusion