Developing Trustworthy Artificial Intelligence (AI)-Driven Tools to Predict Vascular Disease Risk and Progression: the Prospective VASCUL-AID Study

The VASCUL-AID AI tools aim to predict AAA (rupture, mortality, cerebrovascular accident) and PAD (mortality, myocardial infarction, major amputation) progression, MACE and MALE.

The internal validation is done by comparing the predicted disease progression, MACE, MALE and mortality rates with actual disease progression.

Accuracy of developed VASCUL-AID AI-driven algorithms that classify AAA patients according to high or low risk disease progression and MACE. In addition, models will be developed on the other predefined outcomes from the core outcome set, these include: clinical success after an operation [yes/no], graft infection after an operation [yes/no], reintervention [yes/no], the AAA can be treated through endovascular treatment [yes/no], adherence to prescribed drugs [yes/no], secondary prevention, treatment options and complications.

Accuracy of developed VASCUL-AID AI-driven algorithms that classify PAD patients according to high or low risk disease progression, MALE and MACE. In addition, models will be developed on the other predefined outcomes from the core outcome set, these include: acute limb ischemia [yes/no], thromboembolic complications [yes/no], disease progression [yes/no], pain-free walking distance [meters], physical activity, smoking cessation, secondary prevention, treatment options and complications.

Good communication by healthcare worker about diagnosis [Questionnaires].

Numeric (Pain) Rating Scale (0-10, no pain to most intense pain).

PROMIS Scale v1.2 - Global Health measure. Answer rank on a scale of 0-5, with the higher score being more preferred.

EQ-5D-5L Euroqol questionnaire.

The descriptive system comprises five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has 5 levels: no problems, slight problems, moderate problems, severe problems and extreme problems. The patient is asked to indicate his/her health state by ticking the box next to the most appropriate statement in each of the five dimensions. This decision results in a 1-digit number that expresses the level selected for that dimension. The digits for the five dimensions can be combined into a 5-digit number that describes the patient's health state.

Vascular Quality of Life Questionnaire (VascuQol) consists of 25 questions regarding the QoL in patients with vascular diseases, regarding activity level, symptoms, pain, emotion and social consequences. It uses a 7-point Likert scale, where a low score equels a severe sense of limitations.

AneurysmDQoL - Abdominal Aortic Aneurysm Quality of Life

The AneurysmDQoL is a 24-item measure including two overview items designed for audit purposes which measure generic 'present QoL' and AAA-specific 'impact of AAA on QoL'. A further 22 items measure the impact of having an abdominal aortic aneurysm on specific aspects of life and the importance of these aspects of life for QoL. In addition the questionnaire includes a free-text item. This allows patients to comment on any other QoL domains not covered in the questionnaire.

A custom list of questions developed to gather all changes in medical history of the patients.

The 6-minute walking test is a test in which patients are asked to walk for a duration of 6 minutes and then report the distance they have walked in that time and whether they experienced any complaints.

Asks the patient to write down all active medication.

A custom list of questions asking patients about smoking cessation and how they are managing quitting.

This questionnaire scores how well the patient experienced contact with their health care provider. Each question is answered on a scale between strongly agree - strongly disagree.

The project will include data collection of costs and Health-Related Quality of Life (HRQoL) prospectively. The economic analysis will be performed in accordance with the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement. An estimation of costs and HRQoL progression until MACE will be performed and descriptive statistics will be used to describe their respective evolution (up to 4 years).

To describe the evolution and further estimate threshold where the implementation of VASCULAID within hospitals could be more efficient, the number of patients with at least one MACE during the follow-up period will be used to estimate the costs and HRQoL per event among the dedicated subgroup of population where HRQoL is collected. Only direct costs will be considered in the analysis.

The result will be expressed in euros per patient (then distinguishing MACE vs free of MACE) over the follow-up. QALYs will be expressed per patients.