Wearable Movement Exploration Device with Machine Learning Algorithm for Screening and Tracking Diabetic Neuropathy—A Cross-Sectional, Diagnostic, Comparative Study

Radunovic, Goran; Velickovic, Zoran; Pavlov-Dolijanovic, Slavica; Janjic, Sasa; Stojic, Biljana; Jeftovic Velkova, Irena; Suljagic, Nikola; Soldatovic, Ivan

doi:10.3390/bios14040166

Open AccessArticle

Wearable Movement Exploration Device with Machine Learning Algorithm for Screening and Tracking Diabetic Neuropathy—A Cross-Sectional, Diagnostic, Comparative Study

by

Goran Radunovic

^1,2,*,†,

Zoran Velickovic

^1,†

,

Slavica Pavlov-Dolijanovic

^1,2,

Sasa Janjic

¹,

Biljana Stojic

¹,

Irena Jeftovic Velkova

^3,4,

Nikola Suljagic

^3,5 and

Ivan Soldatovic

^2,3,*

¹

Institute of Rheumatology, 11000 Belgrade, Serbia

²

Faculty of Medicine, University of Belgrade, 11000 Belgrade, Serbia

³

DIVS Neuroinformatics DOO, 11000 Belgrade, Serbia

⁴

General Hospital Loznica, 15300 Loznica, Serbia

⁵

Faculty of Electrical Engineering, University of Belgrade, 11000 Belgrade, Serbia

^*

Authors to whom correspondence should be addressed.

^†

These authors contributed equally to this work.

Biosensors 2024, 14(4), 166; https://doi.org/10.3390/bios14040166

Submission received: 16 February 2024 / Revised: 23 March 2024 / Accepted: 27 March 2024 / Published: 29 March 2024

(This article belongs to the Special Issue Applications of AI and Wearable Biosensors in Precision, Personalized and Predictive Medicine)

Download

Browse Figures

Versions Notes

Abstract

Background: Diabetic neuropathy is one of the most common complications of diabetes mellitus. The aim of this study is to evaluate the Moveo device, a novel device that uses a machine learning (ML) algorithm to detect and track diabetic neuropathy. The Moveo device comprises 4 sensors positioned on the back of the hands and feet accompanied by a mobile application that gathers data and ML algorithms that are hosted on a cloud platform. The sensors measure movement signals, which are then transferred to the cloud through the mobile application. The cloud triggers a pipeline for feature extraction and subsequently feeds the ML model with these extracted features. Methods: The pilot study included 23 participants. Eleven patients with diabetes and suspected diabetic neuropathy were included in the experimental group. In the control group, 8 patients had suspected radiculopathy, and 4 participants were healthy. All participants underwent an electrodiagnostic examination (EDx) and a Moveo examination, which consists of sensors placed on the feet and back of the participant’s hands and use of the mobile application. The participant performs six tests that are part of a standard neurological examination, and a ML algorithm calculates the probability of diabetic neuropathy. A user experience questionnaire was used to compare participant experiences with regard to both methods. Results: The total accuracy of the algorithm is 82.1%, with 78% sensitivity and 87% specificity. A high linear correlation up to 0.722 was observed between Moveo and EDx features, which underpins the model’s adequacy. The user experience questionnaire revealed that the majority of patients preferred the less painful method. Conclusions: Moveo represents an accurate, easy-to-use device suitable for home environments, showing promising results and potential for future usage.

Keywords: wearable device; diabetic neuropathy; screening; tracking

Share and Cite

MDPI and ACS Style

Radunovic, G.; Velickovic, Z.; Pavlov-Dolijanovic, S.; Janjic, S.; Stojic, B.; Jeftovic Velkova, I.; Suljagic, N.; Soldatovic, I. Wearable Movement Exploration Device with Machine Learning Algorithm for Screening and Tracking Diabetic Neuropathy—A Cross-Sectional, Diagnostic, Comparative Study. Biosensors 2024, 14, 166. https://doi.org/10.3390/bios14040166

AMA Style

Radunovic G, Velickovic Z, Pavlov-Dolijanovic S, Janjic S, Stojic B, Jeftovic Velkova I, Suljagic N, Soldatovic I. Wearable Movement Exploration Device with Machine Learning Algorithm for Screening and Tracking Diabetic Neuropathy—A Cross-Sectional, Diagnostic, Comparative Study. Biosensors. 2024; 14(4):166. https://doi.org/10.3390/bios14040166

Chicago/Turabian Style

Radunovic, Goran, Zoran Velickovic, Slavica Pavlov-Dolijanovic, Sasa Janjic, Biljana Stojic, Irena Jeftovic Velkova, Nikola Suljagic, and Ivan Soldatovic. 2024. "Wearable Movement Exploration Device with Machine Learning Algorithm for Screening and Tracking Diabetic Neuropathy—A Cross-Sectional, Diagnostic, Comparative Study" Biosensors 14, no. 4: 166. https://doi.org/10.3390/bios14040166

APA Style

Radunovic, G., Velickovic, Z., Pavlov-Dolijanovic, S., Janjic, S., Stojic, B., Jeftovic Velkova, I., Suljagic, N., & Soldatovic, I. (2024). Wearable Movement Exploration Device with Machine Learning Algorithm for Screening and Tracking Diabetic Neuropathy—A Cross-Sectional, Diagnostic, Comparative Study. Biosensors, 14(4), 166. https://doi.org/10.3390/bios14040166

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Wearable Movement Exploration Device with Machine Learning Algorithm for Screening and Tracking Diabetic Neuropathy—A Cross-Sectional, Diagnostic, Comparative Study

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI