Comprehensive Summary

Deligiannis et al. analyzed the impact of artificial intelligence, specifically machine learning and deep learning, and how it affected exercise based cardiovascular health interventions. In order to do this, the researchers analyzed previous studies from 2015 to the present that utilized AI in their research on cardiovascular health. It is known that cardiovascular disease is a leading cause of death globally; however, exercise and cardiac rehabilitation can reduce these risk factors, like hypertension, and will increase VO2 max, increasing lifespan. Artificial intelligence has the potential to personalize exercise plans based on accurate heart data (heart rate, general fitness, VO2 max) and uses reinforcement learning to help with patient motivation with digital coaching programs. Numerous databases like PubMed, Scopus, Embase, and IEEE Xplore were utilized to find articles; many of these articles included adaptive goal setting, reinforcement learning, digital coaching, exercise prescription engines, wearable monitoring, diagnostic support, and population-level strategies. Most of these topics were utilized in their respective studies and the outcomes were increased physical activity, increased daily steps, increased motivation, increased VO2 max, and lower blood pressure in most of the participants in each of the nine studies analyzed. The benefits of each study revealed that there was greater adherence and activity in terms of long term participation in exercise plans. AI generated adaptive step goals and the reinforcement learning improved this among the participants. The types of wearables used also tracked oxygen consumption and heart rate more specifically, facilitating the cardiovascular exercise program.

Outcomes and Implications

Implementation of artificial intelligence in cardiovascular exercise programs has proven to be quite effective among the many studies that Deligiannis et al. analyzed. However, there are still significant limitations that will impact implementation. Many of the models are “black boxes”, which make it difficult for clinicians to completely trust AI decisions. Also, the reliance on wearables and internet access has the potential to exclude “low resource” populations. Additionally, to support the usage of the AI tools, health systems need specific workflows and reimbursement models. However, if the artificial intelligence were to be implemented in clinical settings, accessibility for patients who are unable to attend rehabilitation in-person would be greatly improved. In addition, the hybrid models at home combined with supervised care will help reduce clinician workload in the long term. Finally, the artificial intelligence can provide the patients with continuous safety monitoring with advanced data tracking of biometrics; this can help clinicians determine next steps with treatment or exercise plans for patients.

Our mission is to

Connect medicine with AI innovation.

No spam. Only the latest AI breakthroughs, simplified and relevant to your field.

Our mission is to

Connect medicine with AI innovation.

No spam. Only the latest AI breakthroughs, simplified and relevant to your field.

Our mission is to

Connect medicine with AI innovation.

No spam. Only the latest AI breakthroughs, simplified and relevant to your field.

AIIM Research

Articles

© 2025 AIIM. Created by AIIM IT Team

AIIM Research

Articles

© 2025 AIIM. Created by AIIM IT Team

AIIM Research

Articles

© 2025 AIIM. Created by AIIM IT Team