Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 Diabetes
BackgroundDiabetes affects millions of people worldwide and is steadily increasing. A serious condition associated with diabetes is low glucose levels (hypoglycemia). Monitoring blood glucose is usually performed by invasive methods or intrusive devices, and these devices are...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
JMIR Publications
2023-04-01
|
| Series: | JMIR Diabetes |
| Online Access: | https://diabetes.jmir.org/2023/1/e40990 |
| _version_ | 1827857396144275456 |
|---|---|
| author | Reza Jahromi Karim Zahed Farzan Sasangohar Madhav Erraguntla Ranjana Mehta Khalid Qaraqe |
| author_facet | Reza Jahromi Karim Zahed Farzan Sasangohar Madhav Erraguntla Ranjana Mehta Khalid Qaraqe |
| author_sort | Reza Jahromi |
| collection | DOAJ |
| description |
BackgroundDiabetes affects millions of people worldwide and is steadily increasing. A serious condition associated with diabetes is low glucose levels (hypoglycemia). Monitoring blood glucose is usually performed by invasive methods or intrusive devices, and these devices are currently not available to all patients with diabetes. Hand tremor is a significant symptom of hypoglycemia, as nerves and muscles are powered by blood sugar. However, to our knowledge, no validated tools or algorithms exist to monitor and detect hypoglycemic events via hand tremors.
ObjectiveIn this paper, we propose a noninvasive method to detect hypoglycemic events based on hand tremors using accelerometer data.
MethodsWe analyzed triaxial accelerometer data from a smart watch recorded from 33 patients with type 1 diabetes for 1 month. Time and frequency domain features were extracted from acceleration signals to explore different machine learning models to classify and differentiate between hypoglycemic and nonhypoglycemic states.
ResultsThe mean duration of the hypoglycemic state was 27.31 (SD 5.15) minutes per day for each patient. On average, patients had 1.06 (SD 0.77) hypoglycemic events per day. The ensemble learning model based on random forest, support vector machines, and k-nearest neighbors had the best performance, with a precision of 81.5% and a recall of 78.6%. The results were validated using continuous glucose monitor readings as ground truth.
ConclusionsOur results indicate that the proposed approach can be a potential tool to detect hypoglycemia and can serve as a proactive, nonintrusive alert mechanism for hypoglycemic events. |
| first_indexed | 2024-03-12T12:40:56Z |
| format | Article |
| id | doaj.art-fad0bda694994566b50cf797c1472dc9 |
| institution | Directory Open Access Journal |
| issn | 2371-4379 |
| language | English |
| last_indexed | 2024-03-12T12:40:56Z |
| publishDate | 2023-04-01 |
| publisher | JMIR Publications |
| record_format | Article |
| series | JMIR Diabetes |
| spelling | doaj.art-fad0bda694994566b50cf797c1472dc92023-08-28T23:55:01ZengJMIR PublicationsJMIR Diabetes2371-43792023-04-018e4099010.2196/40990Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 DiabetesReza Jahromihttps://orcid.org/0000-0002-9175-1234Karim Zahedhttps://orcid.org/0000-0002-5087-764XFarzan Sasangoharhttps://orcid.org/0000-0001-9962-5470Madhav Erraguntlahttps://orcid.org/0000-0003-0017-5866Ranjana Mehtahttps://orcid.org/0000-0002-8254-8365Khalid Qaraqehttps://orcid.org/0000-0002-0766-9212 BackgroundDiabetes affects millions of people worldwide and is steadily increasing. A serious condition associated with diabetes is low glucose levels (hypoglycemia). Monitoring blood glucose is usually performed by invasive methods or intrusive devices, and these devices are currently not available to all patients with diabetes. Hand tremor is a significant symptom of hypoglycemia, as nerves and muscles are powered by blood sugar. However, to our knowledge, no validated tools or algorithms exist to monitor and detect hypoglycemic events via hand tremors. ObjectiveIn this paper, we propose a noninvasive method to detect hypoglycemic events based on hand tremors using accelerometer data. MethodsWe analyzed triaxial accelerometer data from a smart watch recorded from 33 patients with type 1 diabetes for 1 month. Time and frequency domain features were extracted from acceleration signals to explore different machine learning models to classify and differentiate between hypoglycemic and nonhypoglycemic states. ResultsThe mean duration of the hypoglycemic state was 27.31 (SD 5.15) minutes per day for each patient. On average, patients had 1.06 (SD 0.77) hypoglycemic events per day. The ensemble learning model based on random forest, support vector machines, and k-nearest neighbors had the best performance, with a precision of 81.5% and a recall of 78.6%. The results were validated using continuous glucose monitor readings as ground truth. ConclusionsOur results indicate that the proposed approach can be a potential tool to detect hypoglycemia and can serve as a proactive, nonintrusive alert mechanism for hypoglycemic events.https://diabetes.jmir.org/2023/1/e40990 |
| spellingShingle | Reza Jahromi Karim Zahed Farzan Sasangohar Madhav Erraguntla Ranjana Mehta Khalid Qaraqe Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 Diabetes JMIR Diabetes |
| title | Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 Diabetes |
| title_full | Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 Diabetes |
| title_fullStr | Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 Diabetes |
| title_full_unstemmed | Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 Diabetes |
| title_short | Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 Diabetes |
| title_sort | hypoglycemia detection using hand tremors home study of patients with type 1 diabetes |
| url | https://diabetes.jmir.org/2023/1/e40990 |
| work_keys_str_mv | AT rezajahromi hypoglycemiadetectionusinghandtremorshomestudyofpatientswithtype1diabetes AT karimzahed hypoglycemiadetectionusinghandtremorshomestudyofpatientswithtype1diabetes AT farzansasangohar hypoglycemiadetectionusinghandtremorshomestudyofpatientswithtype1diabetes AT madhaverraguntla hypoglycemiadetectionusinghandtremorshomestudyofpatientswithtype1diabetes AT ranjanamehta hypoglycemiadetectionusinghandtremorshomestudyofpatientswithtype1diabetes AT khalidqaraqe hypoglycemiadetectionusinghandtremorshomestudyofpatientswithtype1diabetes |