Disk hernia and spondylolisthesis diagnosis using biomechanical features and neural network

Area of Science:

• Medical Informatics
• Biomedical Engineering
• Artificial Intelligence in Medicine

Background:

• Artificial neural networks (ANNs) are increasingly used in medical diagnosis due to their ability to identify complex patterns in data.
• Accurate diagnosis of disk hernia and spondylolisthesis is challenging due to overlapping symptoms, leading to potential misclassification.
• Decision support systems leveraging ANNs can aid clinicians in complex diagnostic scenarios.

Purpose of the Study:

• To propose and evaluate a fast, automatic diagnostic system for disk hernia and spondylolisthesis.
• To utilize biomechanical features and neural network models for patient classification.
• To address the inter-misclassification errors common in diagnosing these spinal conditions.

Main Methods:

• Development of a diagnostic system employing artificial neural networks.
• Training of feedforward neural networks and radial basis function networks.
• Utilizing biomechanical features extracted from a public database for model training.

Main Results:

• The proposed system demonstrated the capability for fast decision-making in diagnosing disk hernia and spondylolisthesis.
• Reasonable accuracies were achieved in classifying patients into categories: disk hernia, spondylolisthesis, or normal.
• The study showed promising results for the application of neural networks in this diagnostic context.

Conclusions:

• Neural networks show significant potential as efficient and effective expert systems for diagnosing disk hernia and spondylolisthesis.
• The developed system offers a viable approach to improve diagnostic speed and accuracy for these conditions.
• Further application of ANNs in medical decision support systems for complex spinal disorders is warranted.