Related Experiment Video
Updated: Jun 2, 2025
11:03
High-resolution In Vivo Manual Segmentation Protocol for Human Hippocampal Subfields Using 3T Magnetic Resonance Imaging
Published on: November 10, 2015
9.3K
A semi-supervised deep neuro-fuzzy iterative learning system for automatic segmentation of hippocampus brain MRI
M Nisha1, T Kannan2, K Sivasankari3
1Department of Computer Science and Engineering, Akshaya College of Engineering and Technology, Coimbatore, Tamil Nadu, India.
Mathematical Biosciences and Engineering : MBE
|January 14, 2025
Summary
This study introduces a faster, more accurate method for segmenting the hippocampus in brain MR images using a novel Deep Neuro-Fuzzy technique. The new approach improves diagnostic accuracy for earlier disease detection.
Area of Science:
- Neuroimaging
- Medical Image Analysis
- Artificial Intelligence in Medicine
Background:
- The hippocampus, a seahorse-shaped structure in the medial temporal lobe, is vital for memory and spatial navigation.
- Accurate segmentation of the hippocampus from Magnetic Resonance (MR) images is crucial for diagnosing neurological conditions.
- Manual segmentation is time-consuming and prone to inter-observer variability.
Purpose of the Study:
- To develop and evaluate an automated hippocampus segmentation technique.
- To improve accuracy and reduce false positive/negative rates compared to existing methods.
- To enable faster and more reliable diagnosis of hippocampus-related disorders.
Main Methods:
- A novel Semi-Supervised Deep Neuro-Fuzzy Iterative Learning System (SS-DNFIL) algorithm was proposed.
- The algorithm learns image distribution rapidly through semi-supervised iterative learning.
- The technique was validated on a large dataset of 18,900 Kaggle MR images.
Main Results:
- The SS-DNFIL achieved a 0.97 Dice coefficient and 0.93 Jaccard coefficient.
- High sensitivity (0.95) and specificity (0.97) were recorded.
- Low false positive (0.09) and false negative (0.08) rates were observed, outperforming existing methods.
Conclusions:
- The proposed SS-DNFIL technique offers superior performance for automatic hippocampus segmentation.
- This automated method significantly enhances diagnostic accuracy and efficiency.
- Early and accurate diagnosis facilitated by this technique can improve patient outcomes and longevity.
