Related Experiment Video
Updated: Oct 15, 2025
08:56
Nerve Ultrasound Protocol to Detect Dysimmune Neuropathies
Published on: October 7, 2021
2.9K
A nomogram-based diabetic sensorimotor polyneuropathy severity prediction using Michigan neuropathy screening
Fahmida Haque1, Mamun Bin Ibne Reaz1, Muhammad E H Chowdhury2
1Department of Electrical, Electronic and System Engineering, Universiti Kebangsaan Malaysia, Bangi, 43600, Selangor, Malaysia.
Computers in Biology and Medicine
|October 29, 2021
Summary
This study developed a machine learning model to grade diabetic sensorimotor polyneuropathy (DSPN) severity using Michigan neuropathy screening instrumentation (MNSI) data. The system accurately identifies DSPN severity, aiding clinicians in patient management.
Area of Science:
- Neurology
- Diabetology
- Medical Informatics
Background:
- Diabetic sensorimotor polyneuropathy (DSPN) is a common complication of diabetes.
- Michigan neuropathy screening instrumentation (MNSI) is widely used for DSPN screening but lacks a severity grading system.
Purpose of the Study:
- To develop and validate a machine learning-based severity grading system for DSPN using MNSI data.
- To identify key MNSI features predictive of DSPN severity.
Main Methods:
- Utilized 19 years of data from the EDIC clinical trials.
- Employed machine learning feature ranking (Multi-Tree Extreme Gradient Boost) to identify important MNSI features.
- Developed and validated a multivariable logistic regression-based nomogram for DSPN severity grading.
Main Results:
- Identified top-10 MNSI features, including foot appearance, ankle reflexes, and vibration perception, crucial for DSPN diagnosis.
- The nomogram achieved high accuracy (97.95% on test set, 98.84% on independent set).
- A DSPN severity score was created, stratifying patients into four levels: absent, mild, moderate, and severe.
Conclusions:
- A novel machine learning-based MNSI severity grading system for DSPN has been developed.
- This system can serve as a decision support tool for clinicians and researchers.
- It aids in identifying high-risk DSPN patients for better clinical management and trial stratification.
