Automated detection of cerebral microbleeds in patients with traumatic brain injury

Abstract

In this paper a Computer Aided Detection (CAD) system is presented to automatically detect Cerebral Microbleeds (CMBs) in patients with Traumatic Brain Injury (TBI). It is believed that the presence of CMBs has clinical prognostic value in TBI patients. To study the contribution of CMBs in patient outcome, accurate detection of CMBs is required. Manual detection of CMBs in TBI patients is a time consuming task that is prone to errors, because CMBs are easily overlooked and are difficult to distinguish from blood vessels.This study included 33 TBI patients. Because of the laborious nature of manually annotating CMBs, only one trained expert manually annotated the CMBs in all 33 patients. A subset of ten TBI patients was annotated by six experts. Our CAD system makes use of both Susceptibility Weighted Imaging (SWI) and T1 weighted magnetic resonance images to detect CMBs. After pre-processing these images, a two-step approach was used for automated detection of CMBs. In the first step, each voxel was characterized by twelve features based on the dark and spherical nature of CMBs and a random forest classifier was used to identify CMB candidate locations. In the second step, segmentations were made from each identified candidate location. Subsequently an object-based classifier was used to remove false positive detections of the voxel classifier, by considering seven object-based features that discriminate between spherical objects (CMBs) and elongated objects (blood vessels). A guided user interface was designed for fast evaluation of the CAD system result. During this process, an expert checked each CMB detected by the CAD system.A Fleiss' kappa value of only 0.24 showed that the inter-observer variability for the TBI patients in this study was very large. An expert using the guided user interface reached an average sensitivity of 93%, which was significantly higher (p = 0.03) than the average sensitivity of 77% (sd 12.4%) that the six experts manually detected. Furthermore, with the use of this CAD system the reading time was substantially reduced from one hour to 13 minutes per patient, because the CAD system only detects on average 25.9 false positives per TBI patient, resulting in 0.29 false positives per definite CMB finding.

Risk factors of radiotherapy‐induced cerebral microbleeds and serial analysis of their size compared with white matter changes: A 7T MRI study in 113 adult patients with brain tumors

“…14 van den Heuvel et al reported an average sensitivity of 77% based on manual detection by six experts, which increased to 93% with the aid of a guided user interface. 17 The implementation of our semiautomated, user-guided CMB detection algorithm in this study, 12 likely enhanced our ability to detect CMBs earlier on.…”

Automated algorithm for counting microbleeds in patients with familial cerebral cavernous malformations

“…Several automated algorithms have been proposed to detect microbleeds on SWI images in various populations, including Alzheimer’s disease [10], atherosclerotic disease [11], stroke [12, 13], traumatic brain injury [14], and brain tumors [15]. These algorithms can be divided into two categories: hypothesis-driven approaches that are based on pre-defined geographic features (such as shape and size) and data-driven approaches that are based on extracted high-level features.…”

“…These algorithms can be divided into two categories: hypothesis-driven approaches that are based on pre-defined geographic features (such as shape and size) and data-driven approaches that are based on extracted high-level features. In general, hypothesis-driven approaches [11, 12, 14, 15] are relatively simple to implement, but rely on limited known physical features of microbleeds. On the other hand, data-driven approaches [10, 13]can extract hundreds and thousands of high-level features but place higher demands on the computational resources required and may come with the risk of over-fitting.…”

Cerebral micro‐bleeding identification based on a nine‐layer convolutional neural network with stochastic pooling

“…12 Considering that the manual detection of CMB is time consuming and prone to errors because of the complex morphological nature of CMB. van den Heuvel et al 19 proposed a two-step method. Researchers made some progress in the past few years.…”

“…Then, within each bounding box, a set of robust 3-dimension Radon-and Hessian-based shape descriptors were extracted to train a cascade of binary random forests (RF). van den Heuvel et al 19 proposed a two-step method. Based on the dark and spherical nature of CMBs, each voxel was characterized by twelve features, and CMB candidates' locations were identified via random forest classifier.…”