Smartphone App Will Be Able to Predict Diabetes

MedicalResearch.com Interview with:

Robert Avram MD MScDivision of CardiologyUniversity of California, San Francisco

Dr. Robert Avram

Robert Avram MD MSc
Division of Cardiology
University of California, San Francisco

MedicalResearch.com: What is the background for this study? Would you briefly describe what is meant by Photoplethysmography?

While analyzing the heart rate data as collected using smartphones apps in the Health-eHeart study, we noticed that diabetic patients had, on average, a higher ‘free-living’ heart rate than non-diabetic patients when adjusted from multiple factors. This pushed us to analyze the signal to see if there were other features that would help differentiate diabetes patients from non-diabetes patients. By identifying these features, we saw a huge opportunity to develop a screening tool for diabetes using deep learning and a smartphone camera and flash, in order to classify patients as having prevalent diabetes/no-diabetes.

Photoplethysmography is the technique of measuring the difference in light absorption by the skin in order to detect blood volume changes in the microvasculature. Most modern mobile devices, including smartphones and many fitness trackers (Apple Wathc, FitBit), have the ability to acquire PPG waveforms, providing a unique opportunity to detect diabetes-related vascular changes at population-scale. 

MedicalResearch.com: What are the main findings?

Response: Using smartphone-based PPG, we can detect prevalent diabetes in a large ambulatory sample of nearly 3 million recordings with reasonable discrimination.  After coupling the app-based screening with common risk factors for diabetes (age, gender, ethnicity and body mass index) our tool was comparable to many traditional diabetes risk scores that are used in clinics to predict diabetes and does not require a physician to administer.

MedicalResearch.com: What should readers take away from your report?

Response: After further validation, smartphone users will be able to screen for prevalent diabetes using the Instant Heart Rate app. This will still require diagnostic confirmation by a physician. 

MedicalResearch.com: What recommendations do you have for future research as a result of this work?

Response: Our research was done using self-reported diabetes in the Health-eHeart cohort. While self-reported diagnoses in the Health-eHeart study have previously been found to be accurate and match electronic medical records diagnoses, we would like to further validate our tool in an ‘in-clinic’ cohort using the diagnostic of diabetes as abstracted from the medical record. Additionally, we need to validate our tool in minorities, such as African Americans and Asians , since those groups were under represented in our cohort, yet have a higher prevalence of undiagnosed diabetes. 

MedicalResearch.com: Is there anything else you would like to add?

Response: This research needs further validation before being disseminated to the public. The potential to transition screening that’s normally done by physicians or nurses to the patient themselves through a smartphone app is a very novel concept and gives us a glimpse into how health care might work in the future, which has the potential to lead to greater independence for patients while also reducing costs.

Disclosures: My research is supported by the “Fonds de la recherche en santé du Québec  » (Grant 35261). I have no relevant financial disclosures to this research.

Citation: ACC 2019 abstract

Predicting Diabetes from Photoplethysmography Using Deep Learning

[last-modified]

 

 

 

The information on MedicalResearch.com is provided for educational purposes only, and is in no way intended to diagnose, cure, or treat any medical or other condition. Always seek the advice of your physician or other qualified health and ask your doctor any questions you may have regarding a medical condition. In addition to all other limitations and disclaimers in this agreement, service provider and its third party providers disclaim any liability or loss in connection with the content provided on this website.

 

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.