Related Experiment Video
Updated: Aug 8, 2025
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Evaluation of grouped capsule network for intracranial hemorrhage segmentation in CT scans
Lingying Wang1,2, Menglin Tang1, Xiuying Hu3
1West China School of Nursing, Sichuan University/ West China Hospital Critical Care Medicine Department, Sichuan University, Chengdu, 610041, China.
Insights
This study introduces GroupCapsNet, a novel deep learning model for segmenting intracranial hemorrhage from CT scans. The method accelerates processing while maintaining high accuracy in identifying bleeding in the brain.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Neurology
Background:
- Intracranial hemorrhage (ICH) is a critical cerebrovascular disease with high mortality rates.
- Accurate and timely diagnosis of ICH is crucial for effective neurosurgical treatment planning and improving patient survival.
- Automated segmentation of ICH on Computed Tomography (CT) scans can significantly aid clinicians.
Purpose of the Study:
- To develop and evaluate a deep learning model for automated segmentation of intracranial hemorrhage regions in non-contrast CT scans.
- To improve the efficiency and accuracy of ICH detection and segmentation in clinical practice.
Main Methods:
- Design of a novel grouped capsule network, GroupCapsNet, for ICH segmentation.
- Implementation of constrained prediction capsules across different input groups to reduce intermediate capsules and accelerate network processing.
- Modification of the squashing function to enhance forward procedure speed without compromising performance.
Main Results:
- The GroupCapsNet model was evaluated on a dataset of 210 ICH CT scan slices.
- The proposed method demonstrated competitive performance in segmenting intracranial hemorrhage areas compared to existing techniques.
- The novel architecture achieved accelerated processing times due to reduced intermediate prediction capsules and modified squashing function.
Conclusions:
- GroupCapsNet offers an efficient and effective deep learning approach for the automated segmentation of intracranial hemorrhage in CT imaging.
- The developed method shows promise in assisting neurosurgeons with treatment planning, potentially improving patient outcomes for ICH.
- Further research can explore the integration of this model into clinical workflows for real-time ICH detection.
Abstract:
Intracranial hemorrhage is a cerebral vascular disease with high mortality. Automotive diagnosing and segmentation of intracranial hemorrhage in Computed Tomography (CT) could assist the neurosurgeon in making treatment plans, which improves the survival rate. In this paper, we design a grouped capsule network named GroupCapsNet to segment the hemorrhage region from a Non-contract CT scan. In grouped capsule network, we constrain the prediction capsules for output capsules produced from different groups of input capsules with various types in each layer. This method can reduce the number of intermediate prediction capsules and accelerate the capsule network. In addition, we modify the squashing function to further accelerate the forward procedure without sacrificing its performance. We evaluate our proposed method with a collected dataset containing 210 intracranial hemorrhage CT scan slices. In experiments, our proposed method achieves competitive results in intracranial hemorrhage area segmentation compared to the existing methods.
