Related Experiment Video
Updated: Jul 9, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.8K
Fully automatic segmentation of brain lacunas resulting from resective surgery using a 3D deep learning model
Raphael Fernandes Casseb1, Brunno Machado de Campos1, Wallace Souza Loos2
1Universidade Estadual de Campinas (UNICAMP), Neuroimaging Laboratory, Campinas, SP, Brazil.
Medrxiv : the Preprint Server for Health Sciences
|November 28, 2023
Summary
ResectVol DL accurately segments brain lacunas using deep learning, outperforming other methods. This tool aids in analyzing epilepsy patient images for better predictive models of surgical outcomes.
Area of Science:
- Medical imaging analysis
- Artificial intelligence in neuroscience
- Deep learning for medical applications
Background:
- Deep learning (DL) advances image segmentation, crucial for analyzing brain abnormalities like lesions and tumors.
- Accurate delineation of brain structures, such as resective lacunas, is vital for extracting volumetric and positional data.
- This information aids clinicians and fuels predictive models for patient outcomes.
Conclusions:
- ResectVol DL accurately segments brain lacunas in epilepsy patients.
- The tool's high performance indicates its potential to assist in developing predictive models for postoperative cognitive and seizure outcomes.
- Automated segmentation tools like ResectVol DL are valuable for advancing neuroimaging analysis and clinical decision-making.
