Towards Complex Regional Pain Syndrome (CRPS) Monitoring and Management Using an AI Powered Mobile Application

Complex Regional Pain Syndrome (CRPS), also known as the suicide disease according to the Reflex Sympathetic Dystrophy Syndrome Association (RSDSA) affects approximately 200,000 patients each year. This disease causes the patient to experience pain that is out of proportion with the severity of the...

Full description

Saved in:

Bibliographic Details
Published in:	2023 11th International Symposium on Digital Forensics and Security (ISDFS) pp. 1 - 6
Main Authors:	Loncala, Julia, Irvin, Madeline, Globis, Alyson, Harvat, Katie, Seitz, Mason, Goodman, Garrett
Format:	Conference Proceeding
Language:	English
Published:	IEEE 11-05-2023
Subjects:	Artificial Intelligence Complex Regional Pain Syndrome Fuzzy Inference System Heart rate Mobile App Mobile applications Pain Pulse oximeter Synthetic Data Temperature measurement Temperature sensors Training
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Complex Regional Pain Syndrome (CRPS), also known as the suicide disease according to the Reflex Sympathetic Dystrophy Syndrome Association (RSDSA) affects approximately 200,000 patients each year. This disease causes the patient to experience pain that is out of proportion with the severity of the actual injury. Furthermore, flare-ups, which are outbursts of pain, muscle weakness, and sensitivity, are unpredictable. From this, a significant portion of CRPS patients consider suicide or euthanasia. Therefore, we find it important to provide a system to assist in alleviating the burden of CRPS. To accomplish this, we have created a prototype mobile application powered by a Fuzzy Inference System (FIS), an Artificial Intelligence (AI) algorithm, to predict an appropriate exercise for a patient to do which should alleviate some pain. We first curated a set of severe pain, moderate pain, and mild pain exercises, provided by a physical fitness expert. Then, using expected environmental data and biometric data to be gathered by a wearable heart rate and pulse oximeter monitor, and self-reported pain levels, we generated a set of synthetic data to train our FIS. After splitting the training and test set in the standard 80/20 training and test split, we obtained a 74.86\% training and 75.56\% test accuracy with a 0.94 F1 score for our severe pain exercise category.
DOI:	10.1109/ISDFS58141.2023.10131798