Medical Diagnosis of Cerebral Palsy Rehabilitation Using Eye Images in Machine Learning Techniques

Area of Science:

• Biomedical Engineering
• Computer Science
• Ophthalmology

Background:

• Cerebral Palsy (CP) is a non-progressive neurological disorder often accompanied by developmental disabilities like strabismus.
• Early diagnosis and regular assessment of CP patients, particularly children, are crucial for effective management.
• Abnormal eye movements, such as strabismus and nystagmus, are common in children with CP.

Purpose of the Study:

• To develop and evaluate a computational methodology for the automatic diagnosis and performance evaluation of CP children.
• To quantify the improvement in CP children's condition through non-invasive eye movement analysis.
• To assess the potential of machine learning in diagnosing eye abnormalities in CP patients.

Main Methods:

• Eye movement data from 40 children with mild CP (aged 3-11 years) were captured using a camera.
• Biomedical image processing and machine learning classification algorithms were employed for analysis.
• A Neural Network Classifier was utilized to classify abnormal eye conditions and evaluate performance improvement.

Main Results:

• The proposed computational method achieved a high classification accuracy of 94.17%.
• The system demonstrated a Specificity Rate of 0.9800 and a Sensitivity Rate of 0.9165.
• The machine learning approach effectively measured and quantified performance improvements in CP children.

Conclusions:

• The developed non-invasive method offers an accurate and automatic approach for detecting abnormalities in CP children.
• This technique provides a reliable tool for evaluating treatment efficacy and monitoring the progress of CP patients.
• Machine learning classification shows significant potential in aiding the diagnosis and management of CP-related eye conditions.