Next Article in Journal
Potential Trends of Point-of-Care Diagnostics—The Next Generation of the Laboratory Diagnosis
Next Article in Special Issue
CoSinGAN: Learning COVID-19 Infection Segmentation from a Single Radiological Image
Previous Article in Journal
Relationship between Semi-Quantitative Parameters of Thallium-201 Myocardial Perfusion Imaging and Coronary Artery Disease
Previous Article in Special Issue
Analyzing Malaria Disease Using Effective Deep Learning Approach
Open AccessArticle

Automated Segmentation and Severity Analysis of Subdural Hematoma for Patients with Traumatic Brain Injuries

1
Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI 48109, USA
2
Michigan Center for Integrative Research in Critical Care (MCIRCC), University of Michigan, Ann Arbor, MI 48109, USA
3
Department of Neurological Surgery and Neurology, University of Michigan, Ann Arbor, MI 48109, USA
4
Department of Neurology, University of Michigan, Ann Arbor, MI 48109, USA
5
Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI 48109, USA
6
Department of Radiology, University of Michigan, Ann Arbor, MI 48109, USA
7
Michigan Institute for Data Science (MIDAS), University of Michigan, Ann Arbor, MI 48109, USA
8
Department of Emergency Medicine, University of Michigan, Ann Arbor, MI 48109, USA
*
Author to whom correspondence should be addressed.
Diagnostics 2020, 10(10), 773; https://doi.org/10.3390/diagnostics10100773
Received: 5 September 2020 / Revised: 18 September 2020 / Accepted: 21 September 2020 / Published: 30 September 2020
(This article belongs to the Special Issue Deep Learning for Computer-Aided Diagnosis in Biomedical Imaging)
Detection and severity assessment of subdural hematoma is a major step in the evaluation of traumatic brain injuries. This is a retrospective study of 110 computed tomography (CT) scans from patients admitted to the Michigan Medicine Neurological Intensive Care Unit or Emergency Department. A machine learning pipeline was developed to segment and assess the severity of subdural hematoma. First, the probability of each point belonging to the hematoma region was determined using a combination of hand-crafted and deep features. This probability provided the initial state of the segmentation. Next, a 3D post-processing model was applied to evolve the initial state and delineate the hematoma. The recall, precision, and Dice similarity coefficient of the proposed segmentation method were 78.61%, 76.12%, and 75.35%, respectively, for the entire population. The Dice similarity coefficient was 79.97% for clinically significant hematomas, which compared favorably to an inter-rater Dice similarity coefficient. In volume-based severity analysis, the proposed model yielded an F1, recall, and specificity of 98.22%, 98.81%, and 92.31%, respectively, in detecting moderate and severe subdural hematomas based on hematoma volume. These results show that the combination of classical image processing and deep learning can outperform deep learning only methods to achieve greater average performance and robustness. Such a system can aid critical care physicians in reducing time to intervention and thereby improve long-term patient outcomes. View Full-Text
Keywords: clinical decision support system; subdural hematoma; medical image processing; machine learning; deep learning clinical decision support system; subdural hematoma; medical image processing; machine learning; deep learning
Show Figures

Graphical abstract

MDPI and ACS Style

Farzaneh, N.; Williamson, C.A.; Jiang, C.; Srinivasan, A.; Bapuraj, J.R.; Gryak, J.; Najarian, K.; Soroushmehr, S.M.R. Automated Segmentation and Severity Analysis of Subdural Hematoma for Patients with Traumatic Brain Injuries. Diagnostics 2020, 10, 773.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Search more from Scilit
 
Search
Back to TopTop