Hypertension Classification Using PPG Signals


Hekim Tanç Y., Öztürk M.

Medical Technologies Congress, TIPTEKNO’22, Antalya, Türkiye, 31 Ekim - 02 Kasım 2022, ss.1-4, (Tam Metin Bildiri)

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Antalya
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.1-4
  • İstanbul Üniversitesi-Cerrahpaşa Adresli: Evet

Özet

A growing number of people throughout the world are suffering from cardiovascular disorders like hypertension. Photoplethysmography (PPG) signals are a noninvasive measurement technique that can reveal the condition of the cardiovascular system in real-time. These aspects have made the PPG a key tool for noninvasive cardiovascular health screening and detection, in addition to its affordability and ease. Based on a Synchrosqueezing Transform and a pre-trained convolutional neural network (GoogLeNet), this paper is offering an automatic classification and early diagnosis technique for hypertension utilizing PPG signals. The F1-Score of 96.83% was achieved when discriminating normotensive patients from prehypertensive and hypertensive patients. Our method outperforms all statistical metrics compared to the method using the classical time-frequency analyzing method.