Video Disease Activity Index: A novel video measure to monitor rheumatoid arthritis in telehealth

Title Video Disease Activity Index: a video measure to monitor rheumatoid arthritis in telemedicine Project Summary/Abstract Rheumatoid arthritis (RA) is a degenerative autoimmune condition of the joints that is treated with medications that suppress the immune system. Medications must be regularly adjusted according to disease activity, based

on clinical examination. Telemedicine has gained a key role in rheumatology during the COVID-19 pandemic, with rheumatologists amending treatment based on patient reports of disease activity and signs of swelling observed on video. However, numerous studies have proven that telemedicine may miss critical information that

affects how disease activity is treated. The proposed project aims to improve the accuracy of assessing disease activity and functioning in RA during telemedicine visits by using laptops and smartphone standard cameras. This will also help reduce the costs associated with objective evaluation for RA. The overall goal of this

exploratory and developmental R21 project is to assess technical feasibility and patient usability of camera- based remote assessment system. The first aim is to develop a web-based system that leverages computer vision to quantify joint range of motion and joint thickness as an indication of joint swelling in RA to determine

disease activity. We also aim to modify and improve the current method of assessing functional impairment by incorporating isometric grip strength using an in-house squeezable ball. During this first phase, the vision-based system and the squeezable ball will be validated on young and older adults through comparison with gold-

standard techniques (e.g., motion capture). The second aim is to evaluate a new scoring system called the Video Disease Activity Index (VDAI) in a cross-sectional feasibility investigation with RA patients (n = 50). The VDAI scoring system will be produced by quantifying joint range of motion and joint thickness to determine the number

of swollen joints. This will provide a measure of disease activity that aligns with the clinically endorsed Clinical Disease Activity Index (CDAI), which will be measured by a rheumatologist in the clinic. Two tests of the VDAI will be conducted: one with a researcher present to evaluate the system's sensitivity and reliability against clinical

examination (CDAI), and another where RA patients will use the web-based application alone, while still in the clinic, with a vision-based feedback algorithm to perform the required activities. As the VDAI and CDAI are on the same scale (0–24), the study will use joint-level power calculations to permit standard limits of agreement

analysis between the two measures (e.g., Bland-Altman). The correlation between grip strength and functional questionnaire scores will also be evaluated using Pearson's tests. These analyses will provide a robust evaluation of the sensitivity and reliability of the VDAI system and its potential to improve the accuracy of

assessing disease activity and functioning in RA during telemedicine visits. By leveraging ubiquitous cameras, this investigation has the potential to significantly increase the quality of disease activity and functioning in RA during telemedicine visits.