Related Experiment Video
Updated: Jul 2, 2025
06:40
Screening of Axonal Degeneration in Carpal Tunnel Syndrome Using Ultrasonography and Nerve Conduction Studies
Published on: January 11, 2019
11.6K
Carpal Tunnel Syndrome Automated Diagnosis: A Motor vs. Sensory Nerve Conduction-Based Approach
Dimitrios Bakalis1, Prokopis Kontogiannis1, Evangelos Ntais2
1Department of Computer Science and Engineering, School of Engineering, University of Ioannina, 45110 Ioannina, Greece.
Bioengineering (Basel, Switzerland)
|February 23, 2024
Summary
Machine learning accurately diagnoses carpal tunnel syndrome (CTS) using nerve conduction studies (NCS). Advanced algorithms analyzing motor and sensory nerve signals achieved high accuracy, aiding clinical diagnosis.
Area of Science:
- Neurology
- Biomedical Engineering
- Artificial Intelligence
Background:
- Carpal tunnel syndrome (CTS) diagnosis relies on electrophysiological tests.
- Objective interpretation of nerve conduction studies (NCS) can be challenging.
Purpose of the Study:
- To assess machine learning (ML) effectiveness for automated CTS diagnosis using NCS data.
- To compare two ML methodologies for classifying CTS severity.
Main Methods:
- Motor NCS signals were converted to time-frequency spectrograms for deep 2D convolutional neural network (CONV2D) analysis.
- Sensory NCS signals underwent multilevel wavelet decomposition (MWD) for feature extraction and classifier training.
- Classifiers distinguished between controls, mild CTS, and moderate-to-severe CTS.
Main Results:
- The CONV2D model using motor signals achieved 94% accuracy.
- The MWD-based model using sensory signals achieved 97.1% accuracy.
- Both ML approaches outperformed previous automated CTS diagnostic methods.
Conclusions:
- ML classification techniques significantly enhance the accuracy of NCS for CTS diagnosis.
- These methods offer a valuable tool for clinicians in diagnosing and managing neuropathies like CTS.
- Integrating AI into NCS analysis improves diagnostic efficiency and precision.
