Related Experiment Video
Updated: Sep 14, 2025
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
42.8K
EXPEDITION: an Exploratory deep learning method to quantitatively predict hematoma progression after intracerebral
Siqi Chen1, Zixiao Li1, Yinsheng Li2
1Department of Neurology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China.
Neurological Research
|July 24, 2025
Summary
A new deep learning method, EXPEDITION, accurately predicts hematoma expansion after intracerebral hemorrhage (ICH). This tool aids in managing ICH by forecasting disease progression in basal ganglia or thalamus.
Area of Science:
- Neurology
- Radiology
- Artificial Intelligence
Background:
- Intracerebral hemorrhage (ICH) is a serious condition with significant mortality and morbidity.
- Predicting hematoma progression is crucial for effective patient management and treatment strategies.
Purpose of the Study:
- To develop and validate an Exploratory deep learning method (EXPEDITION) for quantitative prediction of hematoma progression.
- To assess the accuracy of EXPEDITION in predicting hematoma volume changes after ICH.
Main Methods:
- Retrospective enrollment of patients with primary ICH in the basal ganglia or thalamus.
- Collection of baseline non-contrast CT (NCCT), CT perfusion (CTP) images, and serial NCCT scans.
- Extraction of cerebral venous hemodynamic features from CTP and training of the EXPEDITION deep learning model.
Main Results:
- The study included 73 patients (58 in training, 15 in testing).
- For the test set, EXPEDITION showed a mean difference of -0.96 [-9.64, +7.71] mL in hematoma volume prediction compared to reference.
- The model demonstrated good consistency between true and predicted hematoma volumes, indicating accurate quantitative prediction.
Conclusions:
- EXPEDITION, an Exploratory deep learning method, is proposed for quantitative prediction of hematoma progression.
- The method shows promise for aiding in the management of patients with ICH in specific brain regions.
