Related Experiment Video
Updated: Sep 5, 2025
09:03
Transpupillary Two-Photon In Vivo Imaging of the Mouse Retina
Published on: February 13, 2021
4.5K
Classification of Transgenic Mice by Retinal Imaging Using SVMS
Farrukh Sayeed1, K Rafeeq Ahmed2, M S Vinmathi3
1Department of Electrical and Electronics Engineering, ACE College of Engineering, Trivandrum 695027, Kerala, India.
Computational Intelligence and Neuroscience
|July 11, 2022
Summary
This study uses retinal imaging and optical coherence tomography (OCT) to detect Alzheimer's disease (AD) in mice. Support vector machines achieved 92% accuracy in classifying transgenic mice, offering a new diagnostic approach.
Area of Science:
- Neuroscience
- Ophthalmology
- Biomedical Engineering
Background:
- Alzheimer's disease (AD) diagnosis is challenging due to complex brain pathologies.
- Amyloid-β (Aβ) plaques are a hallmark of AD.
- Current diagnostic methods for AD can be invasive and complex.
Purpose of the Study:
- To investigate the utility of retinal imaging for Alzheimer's disease detection.
- To classify wild-type (WT) and transgenic mice models (TMM) of AD using optical coherence tomography (OCT) images.
- To develop an accurate and non-invasive method for AD detection.
Main Methods:
- Retinal images from WT and TMM mice were analyzed using optical coherence tomography (OCT).
- Support vector machines (SVM) with genetic algorithm-optimized kernel selection were employed for classification.
- Texture features from retinal images were extracted and selected for enhanced SVM classification.
Main Results:
- The classification model achieved an overall accuracy of 92%.
- A precision of 91% was obtained for the classification of transgenic mice.
- The radial basis kernel function demonstrated superior performance within the SVM framework.
Conclusions:
- Retinal imaging, particularly OCT, shows promise as a non-invasive tool for detecting Alzheimer's disease.
- SVM with optimized kernel selection and texture feature extraction provides an effective classification strategy.
- This approach could lead to earlier and more accessible AD diagnosis.
