Related Experiment Video
Updated: Sep 26, 2025
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Automatic hemorrhage segmentation on head CT scan for traumatic brain injury using 3D deep learning model
Papangkorn Inkeaw1, Salita Angkurawaranon2, Piyapong Khumrin3
1Data Science Research Center, Department of Computer Science, Faculty of Science, Chiang Mai University, Chiang Mai, 50200, Thailand.
Insights
A new deep learning method automatically segments traumatic brain injury subtypes like subdural, epidural, and intraparenchymal hemorrhage on CT scans. This approach improves diagnostic accuracy for critical surgical decisions.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Neurosurgery
Background:
- Traumatic brain injury (TBI) is a leading cause of death and disability in young adults.
- Surgical intervention decisions for TBI, such as craniotomy, depend on injury type and neurological examination.
- Intracranial hemorrhage subtypes (subdural, epidural, intraparenchymal) often require surgical management.
Purpose of the Study:
- To develop and evaluate a novel automatic method for segmenting traumatic brain injury hemorrhage subtypes on CT scans.
- To integrate CT scans with bone window settings as input for a deep learning model.
- To improve the accuracy and efficiency of hemorrhage segmentation for surgical planning.
Main Methods:
- A deep learning model, a 3D convolutional neural network with four parallel pathways, was utilized for segmentation.
- Brain CT scans from adult patients were preprocessed using subdural and bone window settings.
- The model was trained on a dataset and segmentation results were refined using post-processing techniques, including a region-growing algorithm.
Main Results:
- The proposed method achieved median Dice similarity coefficients greater than 0.37 for all hemorrhage subtypes.
- The deep learning model demonstrated improved segmentation performance compared to existing literature.
- The integration of bone window CT scans enhanced the segmentation capabilities.
Conclusions:
- The novel automatic deep learning method effectively segments traumatic brain injury hemorrhage subtypes.
- The approach shows promise for improving diagnostic accuracy and aiding surgical decision-making in TBI cases.
- This method offers a significant advancement over previous techniques for CT-based hemorrhage segmentation.
Abstract:
The most common cause of long-term disability and death in young adults is a traumatic brain injury. The decision for surgical intervention for craniotomy is dependent on the injury type and the patient's neurologic exam. The potential subtypes of intracranial hemorrhage that may necessitate surgical intervention include subdural hemorrhage, epidural hemorrhage, and intraparenchymal hemorrhage. We proposed a novel automatic method for segmenting the hemorrhage subtypes on a CT scan by integrated CT scan with bone window as input of a deep learning model. Brain CT scans were collected from adult patients and annotated regions of subdural hemorrhage, epidural hemorrhage, and intraparenchymal hemorrhage by neuroradiologists. Their raw DICOM images were preprocessed by two different window settings i.e., subdural and bone windows. The collected CT scans were divided into two datasets namely training and test datasets. A deep-learning model was modified to segment regions of each hemorrhage subtype. The model is a three-dimensional convolutional neural network including four parallel pathways that process the input at different resolutions. It was trained by a training dataset. After the segmentation result was produced by the deep-learning model, it was then improved in the post-processing step. The size of the segmented lesion was considered, and a region-growing algorithm was applied. We evaluated the performance of the proposed method on the test dataset. The method reached the median Dice similarity coefficients higher than 0.37 for each hemorrhage subtype. The proposed method demonstrates higher Dice similarity coefficients and improved segmentation performance compared to previously published literature.
