Clinical Applications of AI in Cardiovascular Medicine: Key Points
- Authors:
- Lüscher TF, Wenzl FA, D'Ascenzo F, Friedman PA, Antoniades C.
- Citation:
- Artificial Intelligence in Cardiovascular Medicine: Clinical Applications. Eur Heart J 2024;Aug 19:[Epub ahead of print].
The following are key points to remember from a state-of-the-art review on the clinical applications of artificial intelligence (AI) in cardiovascular medicine:
- The practice of clinical medicine requires the integration of various forms of patient data including demographics, symptom characteristics, electrocardiogram (ECG) findings, laboratory values, biomarker levels, and imaging studies.
- Decision-making for optimal management should be based on a high probability that the envisaged treatment is appropriate, provides benefit, and bears no or little potential harm. To that end, personalized risk-benefit considerations should guide the management of individual patients to achieve optimal results.
- These basic clinical tasks have become more and more challenging with the massively growing data now available; AI and machine learning (ML) may provide assistance for clinicians by obtaining and comprehensively preparing the history of patients, analyzing face and voice and other clinical features, by integrating laboratory results, biomarkers, and imaging.
- AI enables machines, in particular computer systems, to mimic human cognitive function. It integrates tasks like learning, reasoning, problem solving, perception, and understanding language, allowing computers to derive insights from data, make informed decisions, and solve complex problems.
- Furthermore, AI/ML can provide a comprehensive risk assessment as a basis of optimal acute and chronic care. However, the clinical usefulness of AI/ML algorithms should be carefully assessed, validated with confirmation datasets before clinical use, and repeatedly re-evaluated as patient phenotypes change.
- In the catheterization laboratory, AI and ML may be helpful in: a) preparing the procedure (risk assessment, noninvasive imaging); b) early treatment planning (coronary anatomy, lesion characteristics, regional ischemia); c) lesion detection, characterization, and functional flow assessment during the procedure; and d) eventually guiding robotic procedures.
- It is very likely that AI/ML will massively change the practice of medicine. It will make medicine more precise and faster. AI/ML algorithms provide information not accessible for the clinician, particularly in imaging and ECG analysis. For example, risk prediction is more precise, as documented by the AI/ML-enabled GRACE 3.0 score, among others.
- AI/ ML-enabled information is much faster. As a consequence, physicians will have better information and more time to discuss management options with their patients. Indeed, AI/ML-provided information on diagnostics and guideline-based therapeutic options are provided comprehensively and timely. However, in contrast to humans, AI/ML cannot yet provide the same degree of empathy, personal interaction, and trust as good physicians.
- Authorities around the globe do require registration of AI/ML tools whether as stand-alone or embedded into a medical device prior to medical use. The proposed EU Artificial Intelligence Act that will come into force later in 2024 considers AI/ML a device that has to undergo a conformity assessment procedure prior to being approved through notified bodies.
- Finally, AI and ML algorithms must comply with the highest quality standards as they are increasingly used for the clinical management of patients. Such standards are related to: a) clear intended use, b) data quality, c) adequate sample size, d) representativeness for the patient population (e.g., males and females, different ethnicities, and age groups), e) appropriate external validation of developed algorithms, f) openness of data and software, and g) ongoing adaptations as patient phenotypes and diagnoses change over time.
Clinical Topics: Noninvasive Imaging, Cardiovascular Care Team, Prevention
Keywords: Artificial Intelligence, Diagnostic Imaging, Machine Learning
< Back to Listings