Key result
A deep neural network algorithm for photoplethysmography analysis accurately distinguished between older patients with diabetes and hypertension and young healthy subjects, achieving an AUC of 0.998.
Why the study?
Automated PPG methods have been limited by susceptibility to motion artifacts and individual waveform variation, creating a need for deep neural network algorithms to efficiently analyze telemedicine data.
Does a deep neural network algorithm accurately classify photoplethysmography waveforms between healthy individuals and patients with hypertension and diabetes?
Cross-Sectional (n=368)
No
Does a deep neural network algorithm accurately classify photoplethysmography waveforms between healthy individuals and patients with hypertension and diabetes?
Effect estimate: AUC 0.998
A deep neural network-based algorithm can highly accurately distinguish between healthy individuals and older patients with comorbidities using photoplethysmography waveforms, facilitating automated telemedicine monitoring.
May aid remote comorbidity screening; hypothesis-generating and requires prospective validation before clinical use.
Photoplethysmography (PPG) is easy to perform and provides a variety of measurements, including details of heart rate and arrhythmia. However, automated PPG methods have not been developed because of their susceptibility to motion artifacts and differences in waveform characteristics among individuals. With increasing use of telemedicine, there is growing interest in application of deep neural network (DNN) technology for efficient analysis of vast amounts of PPG data. This study proposes an automatic algorithm incorporating DNNs for individual and patient-group identification; this is achieved by selecting normally measured waveforms, deleting error regions, and normalizing the pulse wave to obtain 10 “section values” that can be easily compared to other waveforms. The proposed algorithm was able to distinguish between patients aged 60–75 years with diabetes and hypertension and healthy subjects aged 25–35 years (AUC = 0.998). On the other hand, errors were frequently observed in identification of individuals (AUC = 0.819).
No takes yet. Share an insight, caveat, or question.
Kim et al. (2021) conducted a cross-sectional in Diabetes and hypertension (n=368). Deep neural network (DNN) algorithm for PPG analysis vs. Healthy controls was evaluated on Patient group identification (distinguishing patients with diabetes and hypertension from healthy subjects) (AUC 0.998). A deep neural network algorithm for photoplethysmography analysis accurately distinguished between older patients with diabetes and hypertension and young healthy subjects, achieving an AUC of 0.998.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: