Monitoring of Cardiorespiratory Signals Using Thermal Imaging: A Pilot Study on Healthy Human Subjects

• Published in: Sensors (Basel, Switzerland) Vol. 18; no. 5; p. 1541
• Main Authors: Barbosa Pereira, Carina, Czaplik, Michael, Blazek, Vladimir, Leonhardt, Steffen, Teichmann, Daniel
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 13.05.2018; MDPI
• Subjects: Adolescent; Adult; Algorithms; Female; Heart Rate - physiology; Heat detection; Humans; Infrared Rays; Male; Measurement techniques; Middle Aged; Monitoring, Physiologic - methods; Pilot Projects; Principal Component Analysis; Respiratory Rate - physiology; Signal Processing, Computer-Assisted; Thermography - methods; Young Adult; infrared thermography; heart rate; infrared imaging; respiratory rate; thermal imaging; contactless measurement
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s18051541

Heart rate (HR) and respiratory rate (RR) are important parameters for patient assessment. However, current measurement techniques require attachment of sensors to the patient’s body, often leading to discomfort, stress and even pain. A new algorithm is presented for monitoring both HR and RR using thermal imaging. The cyclical ejection of blood flow from the heart to the head (through carotid arteries and thoracic aorta) leads to periodic movements of the head; these vertical movements are used to assess HR. Respiratory rate is estimated by using temperature fluctuations under the nose during the respiratory cycle. To test the viability and feasibility of this approach, a pilot study was conducted with 20 healthy subjects (aged 18–36 and 1 aged 50 years). The study consisted of two phases: phase A (frontal view acquisitions) and phase B (side view acquisitions). To validate the results, photoplethysmography and thoracic effort (piezoplethysmography) were simultaneously recorded. High agreement between infrared thermography and ground truth/gold standard was achieved. For HR, the root-mean-square errors (RMSE) for phases A and B were 3.53 ± 1.53 and 3.43 ± 1.61 beats per minute, respectively. For RR, the RMSE between thermal imaging and piezoplethysmography stayed around 0.71 ± 0.30 breaths per minute (phase A). This study demonstrates that infrared thermography may be a promising, clinically relevant alternative for the assessment of HR and RR.