Training of a Artificial Intelligence Model to Detect Venous Diseases Using PPG Technology

The Whiteley Clinic20 enrolled

Overview

This clinical research aims to evaluate the effectiveness of using Photoplethysmography (PPG) signals combined with Artificial Intelligence (AI) algorithms, for the precise classification and diagnosis of Venous Diseases of the lower limb. This study invites a group of participants who currently undergoing investigations for venous disease at The Whiteley Clinic (hereinafter referred to as TWC). The Participants will be classified into control (healthy individuals with no significant venous disease) and chronic venous disease (CVD) (diagnosed with proven venous disease) groups. Prospective participants who express an interest in being included in the study will be given a patient information sheet and will undergo a briefing of the pilot study. If they consent and sign the relevant consent forms, the participants will perform a series of standardized exercises under the supervision of a consultant vascular surgeon. Throughout the exercises, a data acquisition device attached to the ankle records the PPG signals, capturing the changes in blood volume due to the reflected PPG signals from the red blood cells during the movement. Thus, once the data is collected and recorded, this allows for the analysis of the data of the control group and CVD group against each other. During the analysis of the two groups' PPG signals, the objective lies within the capability to detect subtle nuances in the patterns of the PPG signals during the performed movements using AI algorithms. The AI algorithms will distinguish patterns or features indicating the presence or absence of venous disease. This study seeks to contribute valuable insights into enhancing the diagnosis of venous disease using PPG and AI algorithms, paving novel approaches to Venous healthcare.

Study Type

OBSERVATIONAL

Enrollment

Conditions

Venous Disease

Interventions

The study investigates venous competence through three distinct exercises using photoplethysmography (PPG) technology to record blood flow in the leg veins of 20 subjects, split into two groups: those with chronic venous disease (CVD) and those without. The null hypothesis is that there will be no significant difference in venous filling times (VFT) and PPG trace variations between subjects with CVD and those without under different physical conditions. The alternative hypothesis suggests that individuals with CVD will show distinct PPG patterns, particularly shorter VFT and varied pressure changes, indicative of venous reflux or obstruction. This hypothesis is chosen based on prior evidence suggesting observable differences in venous function between affected and non-affected individuals.

Eligibility

Sex: ALLMin age: 18 YearsMax age: 80 YearsHealthy volunteers:

Medical Language ↔ Plain English

Inclusion Criteria: * Patients are attending for investigation of suspected venous disease. Patients must be able to walk and mobile normally and have good skin integrity of the lower leg, where the PPG is attached. All patients attending TWC are 18 years or older. Exclusion Criteria: * Subjects with known arterial occlusive disease or physical disability affecting gait or ankle movement will be excluded. Patients unable to have a PPG attached to the lower leg (ie: active ulceration) will be excluded. Patients unable to give consent. Pregnant female.

Outcomes

Primary Outcomes

Diagnostic Accuracy of an AI Model for Venous Disease Detection Using PPG Signals

The primary outcome measure of this study is to evaluate the diagnostic accuracy of an AI model in detecting venous disease through the using PPG signals. This will be quantified by assessing the sensitivity and specificity of the AI model when analysing PPG signals from healthy participants without venous diease, and non-healthy participants with venous disease, without the need for direct intervention of a vascular consultant. These results will help evaluate the AI model in terms of how accurately it can identify Venous disease.

Time frame: June 2024 - September 2024