Emotion Detection from ECG Signals with Continuous Wavelet Transform Sürekli Dalgacık Dönüşümü ile EKG Sinyallerinden Duygu Tespiti

Artuğ T., Subasi A.

2024 Medical Technologies Congress, TIPTEKNO 2024, Muğla, Türkiye, 10 - 12 Ekim 2024, (Tam Metin Bildiri) identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/tiptekno63488.2024.10755271
  • Basıldığı Şehir: Muğla
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: continuous wavelet transform, convolutional neural networks, electrocardiogram, emotion detection, signal processing, transfer learning
  • Yıldız Teknik Üniversitesi Adresli: Hayır

Özet

Knowing the individual's current emotional state through biomedical signals may give an opportunity for early intervention. ECG signal is one of the most well-known bio-signals. In this study, ECG signals were used which were recorded as a result of 3 sets of 7 different videos watched by 12 individuals. Two dimension images were obtained from ECG signals by using continuous wavelet transform. The resulting dataset includes 7 different emotions. This dataset was classified on pre-trained convolutional neural networks via transfer learning method. It is aimed to increase the classification performance and the highest value was achieved as 81.63% with the ResNet50 network.