Affiliation:
1. Saraswati College of Engineering
Abstract
Abstract
Background
Peripheral Pulse Analyser (PPA) is developed by Bhabha Atomic Research Centre (BARC) Mumbai. The recording of peripheral pulses is a highly important and necessary non-invasive technology used by clinicians to diagnose numerous disorders. Nonexperts may struggle to estimate waveforms accurately; motion artefacts may emerge during tonometry measurements if the skin-sensor contact pressure is insufficient. The aim of waveform analysis was to study differentiating parameters of PPA waveforms for early disease detection.
Methods
We used 70 subjects in two groups as healthy and unhealthy subjects. The pulse morphology is seen to be different in healthy and unhealthy subject. Each PPA signal repeats two or three pulse morphologies. Study was performed at Saraswati College of Engineering (SCOE), Kharghar in collaboration with Board of Research in Nuclear Sciences (BRNS).
Findings: With peaks and valleys different parameters were introduced for eight pulse morphologies (P1 to P8). Polarity, amplitude and intervals are key parameters for waveform analysis of P1 to P8.
Novelty: Automatically analysed blood flow variation of datasets can be used to classify the patterns into various classes to detect diseases. Each PPA signal repeats two or three pulse morphology for the duration of 300 seconds. These pulse morphologies differ in healthy and unhealthy subject. The percentage of pulse morphologies in each PPA signal decides diseases like hypertension, diabetes, and coronary artery diseases (CAD) etc. The waveforms are analysed and tested and are used for disease detection. we have obtained satisfactory results with 96% accuracy.
Publisher
Research Square Platform LLC
Reference18 articles.
1. Patil, Nishant B1,; Bhat, Sushma N1; Christopher, Janisha J1; Abhane, Rameshwari V1; Bagal, Uttam R1; Nagare, Gajanan D1. Role of machine learning techniques in classification of peripheral pulse morphology. MGM Journal of Medical Sciences 10(1):p 126–134, January-March 2023. | DOI: 10.4103/mgmj.mgmj_13_23.
2. https://doi.org/10.4103/mgmj.mgmj_13_23.
3. Kouz K, Scheeren TWL, de Backer D, Saugel B. Pulse Wave Analysis to Estimate Cardiac Output. Anesthesiology. 2021;134(1):119–126. doi: 10.1097/ALN.0000000000003553. PMID: 32914174.
4. https://pubmed.ncbi.nlm.nih.gov/32914174/
5. Azmal Huda Chowdhury, Borzooye Jafarizadeh, Amin Rabiei Baboukani, Nezih Pala, Chunlei Wang, Monitoring and analysis of cardiovascular pulse waveforms using flexible capacitive and piezoresistive pressure sensors and machine learning perspective, Biosensors and Bioelectronics, Volume 237, 2023, 115449, ISSN 0956–5663, https://doi.org/10.1016/j.bios.2023.115449.