Colorimetric Evaluation of Cortisol Hormone and Stress Levels with Articial Intelligence Assisted Image Processing

Çapan E., Cingöz E., Öncel H. U., Arıcan E., Bulut F.

APPLIED BIOCHEMISTRY AND BIOTECHNOLOGY - PART B MOLECULAR BIOTECHNOLOGY, cilt.4, sa.4, ss.1-20, 2024 (SCI-Expanded)

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 4 Sayı: 4
  • Basım Tarihi: 2024
  • Doi Numarası: 10.21203/rs.3.rs-4269783/v1
  • Dergi Adı: APPLIED BIOCHEMISTRY AND BIOTECHNOLOGY - PART B MOLECULAR BIOTECHNOLOGY
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, BIOSIS, Biotechnology Research Abstracts, CAB Abstracts, Chemical Abstracts Core, Compendex, Food Science & Technology Abstracts, Index Islamicus, INSPEC, MEDLINE, Veterinary Science Database
  • Sayfa Sayıları: ss.1-20
  • İstanbul Üniversitesi Adresli: Evet

Özet

A novel approach utilizing Convolutional Neural Networks (CNN), Decision Trees, and Vector Regression methods, along with image processing techniques, has been developed to measure cortisol hormone levels through color change, one of the stress biomarkers. Visible color change of sweat samples obtained from users was achieved using the blue tetrazolium method, followed by analytical evaluation of this color change with the proposed system. Validated with ELISA references and a colorimeter device, the system reached a success rate of 98% in determining main reference values and 84.2% in determining users' cortisol values. Additionally, participants (n=20) underwent the Copenhagen Psychosocial Risk Assessment and Perceived Stress Scale tests, and the correlation between evaluated cortisol levels and stress was examined. Emphasizing the importance of cortisol in stress assessment, an effective tool for rapid cortisol and stress level assessment is proposed.