Publication – AI-AGE

10.03.2025.11.03.2025.

Advancing Medical Education: Collaboration with DECODE on Global Digital Health Guidelines

A groundbreaking initiative led by a group of 211 international experts from 79 countries has resulted in the publication of new guidelines aimed at integrating digital health competencies into medical education worldwide. These guidelines, titled Digital Health Competencies in Medical Education (DECODE), were published in JAMA Network Open and provide a comprehensive framework to help medical institutions prepare future doctors for the digital transformation of healthcare. Among the contributors to this landmark study were experts from the University of Montenegro’s Faculty of Medicine, highlighting the country’s role in shaping the future of medical education. AI-AGE team members participated in this effort, with one of its team members co-authoring the study.

The DECODE guidelines focus on four key areas: professionalism in digital health, patient and population digital health, health information systems, and health data science. These competencies are already being adopted in various countries, where they have influenced new learning outcomes for medical graduates. To support their implementation, an online event will be held on March 14, 2025, offering insights into how institutions can integrate these competencies into their curricula. As the Faculty of Medicine aligns its education strategy with DECODE, this initiative represents a significant step toward equipping future healthcare professionals with the necessary skills to navigate the evolving landscape of digital healthcare.

Click on image to view the paper at the JAMA Network Open website

16.06.2024.14.09.2024.

Symorg 2024 Conference

A research paper prepared by our young researcher was published at the SymOrg 2024 conference, organized by the Faculty of Organizational Science, University of Belgrade, at Zlatibor, Serbia on June 12-14, 2024. The conference, traditionally envisioned as a platform for knowledge innovation and empirical research, bringing together representatives from the scientific and professional community, was themed: ”Unlocking The Hidden Potential Of Organization Through Merging Of Humans And Digitals”, aiming to address the newfound need for balance in the era of AI. This paper was done with the support of EUROCC2 and NCC Montenegro team.

The scientific paper “Detection of Scoliosis” by Elvis Taruh, Enisa Trubljanin, and Dejan Babić explores the application of a deep learning model integrated with a web application to detect scoliosis using x-ray images. Utilizing a dataset of 198 x-ray images from Roboflow, the initial model performance was unsatisfactory, prompting manual annotation of 245 images, which significantly improved the model’s accuracy. YOLOv8, a state-of-the-art object detection algorithm, was used to train two models, demonstrating improved performance with manual annotations. The web application, built with Flask, HTML, CSS, and JavaScript, provides a user-friendly interface for analyzing scoliosis detection results. The backend uses MySQL for data storage and management, facilitating efficient image processing, result display, and feedback from doctors. Evaluation metrics indicate that the second model, which underwent refined annotation and augmentation, performed better, avoiding overfitting and demonstrating higher precision. This approach enhances early scoliosis diagnosis and offers a scalable solution for other medical detection challenges, supporting healthcare providers with more accurate diagnostic tools and improving patient care.

30.05.2024.08.03.2025.

Participation at the EAsDEC 2024 Conference

Prof. Adzic-Zecevic and prof. Popofic at the EAsDEC conference

Researchers from the AI-AGE team participate in the he 34th EUROPEAN ASSOCIATION FOR DIABETIC EYE COMPLICATIONS (EAsDEC) Meeting, which was hosted in the city of Milan, Italy. Initial research results from the AI-AGE project were presented in a form of a poster by prof. N Popovic and prof. A. Adzic Zecevic. The conference took place from 30 May to 1 June. More information can be found here.

ABSTRACT – Introduction / study design: Type 2 diabetes mellitus (T2D) is associated with changes in retinal microvascular complexity measured by fractal dimension (Df). Expression of micro RNAs (miRs), -146a and -101 is also affected by T2D, but the studies investigating these miRs in context of microvascular changes in T2D are scarce. Since hypertension (HTN) and Alzheimer’s dementia (AD) frequently coexist in patients with T2D, in the present cross-sectional observational prospective study, participants were divided in two groups – healthy (H, n=8), and with chronic disease (D, n=20, suffering from T2D, HTN and/or AD). The purpose: Study explores association between changes of retinal microvascular Df and expression levels of circulating miR-146a and miR-101 in patients with T2D. The influence of HTN and AD on this association is also investigated. Methods: Retinal fundus images were captured by using a non-mydriatic, hand-held MIIS-HORUS scope DES200. The optic disc-centered images were manually segmented, binarized, cropped to 350-pixel radius, and Df was determined by using ImageJ 1.53q. MiRs were isolated from plasma, quantified by qRT-PCR and normalized to expression levels of miR-361-5p. SPSS Statistics 29.0.1.0., t-test and ANCOVA were used to compare the two groups. P<0.05 was considered significant. Results: Age was not different between the 2 groups (Hvs.D mean age±SE=63.6±2.9 vs.68.5±1.8, p=0.17). Df and miR-101 expression were decreased in the group D (Hvs.D mean: Df±SE=1.36±0.01 vs.1.32±0.01, p=0.016; miR-101±SE=1.68±0.32 vs.0.83±0.22, p=0.041). Eight participants in the group D had T2D (1-moderate, and 7-no diabetic retinopathy). All participants with T2D had HTN, and 5 of them also had AD. Next, we used HTN and AD as covariates to account for effects of these comorbidities, and to determine effects of T2D. This analysis showed: in addition to decreased Df and miR-101 expression, T2D was associated with increased expression of miR-146a (Hvs.D mean miR-146a±SE=0.58±0.29 vs.1.69±0.17, p=0.018). Conclusions: fa Changes in Df and in expression of miR-101 are non-specific, and can be caused by T2D and concurrent comorbidities. Increased expression of miR-146a might be a part of the unique expression pattern of the circulatory miRNAs associated with T2D.