Cardiovascular disease is the primary cause of morbidity and mortality worldwide. Cardiologists face challenges in clinical decision-making because of the demands for better treatment and the translation of the most recent scientific discoveries into executable strategies. On the contrary, the development of artificial intelligence (AI) and machine learning (ML) in recent decades has allowed healthcare professionals to make more effective and data-driven decisions.¹ Therefore, cardiology is a specialty that requires the intervention of AI-based systems to provide precise management, particularly in chronic conditions such as coronary artery disease (CAD) and atrial fibrillation (AF).

In the last decade, AI has shown its effectiveness in CAD and AF management. AI-based solutions are easily applicable to risk assessment, diagnosis and choice of therapy, procedural guidance, and remote monitoring in CAD patients. AI algorithms examine vast volumes of patient data, including medical history, laboratory findings, imaging investigations, and genetic information to detect CAD risk factors and predict the possibility of the disease. For instance, a recent study employed deep learning to predict CAD risk utilizing the data from the electronic health records of over 25,000 patients.² AI algorithms can help healthcare professionals diagnose CAD by examining imaging data obtained from computed tomography (CT) and magnetic resonance imaging to determine the presence and degree of plaque buildup in the coronary arteries. According to a study, a deep learning system can precisely identify and evaluate coronary artery stenosis on CT images.³ By examining patient data, AI algorithms can assist healthcare professionals to determine the most suitable treatment choice for CAD. Reports suggest that an ML system using patient and clinical data can predict stent restenosis.⁴ Medical personnel can benefit from the real-time feedback and guidance provided by AI algorithms while performing CAD procedures such as angioplasty and stent placement.

Virtual coronary intervention has demonstrated the ability to precisely predict the physiological reaction to stenting.⁵ AI systems can also estimate the results of interventional treatment for patients with CAD. In a multiethnic population study, individuals with ST-elevation myocardial infarction were more accurately identified using ML than thrombolysis in myocardial infarction scoring.⁶ Similarly, AI algorithms can analyze electrocardiogram (ECG) data and help in the diagnosis and management of AF. It has been reported that patients with AF may be identified using point-of-care AI-enabled ECG recorded during a normal sinus rhythm.⁷ AI systems can be utilized to examine large volumes of patient data to detect AF risk factors and assess the potential for developing the disease.

An AI-enabled ECG algorithm was used to effectively predict the recurrence of paroxysmal AF following catheter ablation.⁸ By evaluating patient data and determining the best strategy based on unique patient characteristics, AI algorithms can assist healthcare professionals in selecting the most effective AF treatment. Based on patient features and ECG data, an AI-based approach has been used to determine if electrical cardioversion will be effective for AF.⁹ AI algorithms can help healthcare professionals by giving real-time feedback and assistance during AF operations such as catheter ablation. Regardless of the AF type, low voltage-based ablation techniques are effective for substrate alteration and circumferential pulmonary vein separation.¹⁰ AI-based methods have been used to determine low-voltage regions and AI algorithms can predict the long-term effects of ablation for patients with AF. AFA-Recur, an ML-based probability score, effectively predicts the 1-year likelihood of recurrent atrial arrhythmia following AF ablation.¹¹

The potential and diversity of future AI applications in CAD and AF are encouraging and can completely transform CAD and AF diagnosis, risk classification, and therapy optimization. AI algorithms may combine data from several sources, including wearable devices, imaging data, and electronic health records, to offer a detailed overview of a patient’s health to enhance risk categorization and diagnostic accuracy. Furthermore, healthcare professionals may personalize treatment plans by identifying high-risk individuals. Researchers are investigating more sophisticated AI algorithms, such as deep learning and reinforcement learning, to increase the precision and therapeutic value of CAD and AF control. These algorithms have already demonstrated the potential to predict the risk of unfavorable occurrences and consequently modify the treatment plans.¹² Clinical processes can utilize AI algorithms to provide healthcare professionals real-time assistance in knowledgeable treatment decision-making and enhance patient outcomes. AI-based systems may be beneficial in precision medicine in CAD and AF management to improve outcomes and save healthcare costs by identifying the patient subgroups that can develop a reaction to certain treatments. Moreover, AI algorithms can remotely monitor patients with CAD and AF, improving results and reducing the need for hospitalization.

Although AI has shown a lot of promise in the administration of CAD and AF, its usage has certain limitations. Large quantities of high-quality data is required to develop AI systems. The capacity of AI to precisely predict outcomes or make informed treatment decisions may, in some circumstances, be constrained by the paucity of data for particular patient subgroups or unusual clinical presentations. The generalization of AI models to different populations or healthcare systems may be influenced by the data usage from a particular population or healthcare system during their development. This can be challenging when using AI models on various patient groups. Certain AI models are sophisticated and challenging to evaluate, making the understanding of the predictions generated by the model difficult. This may prevent healthcare professionals from adopting it as they require strong validation for decision-making. The use of AI in healthcare raises ethical issues, such as the possibility of biases in the data used to create the algorithms or concerns regarding the confidentiality of patient data.

In conclusion, as the health industry becomes increasingly digitalized, it is critical to adopt AI rather than resist it. AI will provide physicians a thorough understanding of patient-level data, particularly in chronic diseases such as CAD or AF. Physicians should prepare for the AI era by obtaining the necessary skills required to use AI models and appropriately interpret the findings.

REFERENCES

1. Koçak B, Cuocolo R, dos Santos DP, Stanzione A, Ugga L. Must-have Qualities of Clinical Research on Artificial Intelligence and Machine Learning. Balkan Med J. 2023;40:3-12.
2. Ding YD, Zhang Y, He LQ, et al. [A deep-learning model for the assessment of coronary heart disease and related risk factors via the evaluation of retinal fundus photographs]. Zhonghua Xin Xue Guan Bing Za Zhi.  2022;50:1201-1206.
3. Kang D, Dey D, Slomka PJ, et al. Structured learning algorithm for detection of nonobstructive and obstructive coronary plaque lesions from computed tomography angiography. J Med Imaging (Bellingham). 2015;2:014003.
4. Sampedro-Gomez J, Dorado-Diaz PI, Vicente-Palacios V, et al. Machine Learning to Predict Stent Restenosis Based on Daily Demographic, Clinical, and Angiographic Characteristics. Can J Cardiol. 2020;36:1624-1632.
5. Gosling RC, Morris PD, Silva Soto DA, Lawford PV, Hose DR, Gunn JP. Virtual Coronary Intervention: A Treatment Planning Tool Based Upon the Angiogram. JACC Cardiovasc Imaging. 2019;12:865-872.
6. Aziz F, Malek S, Ibrahim KS, et al. Short- and long-term mortality prediction after an acute ST-elevation myocardial infarction (STEMI) in Asians: A machine learning approach. PLoS One. 2021;16:e0254894.
7. Attia ZI, Noseworthy PA, Lopez-Jimenez F, et al. An artificial intelligence-enabled ECG algorithm for the identification of patients with atrial fibrillation during sinus rhythm: a retrospective analysis of outcome prediction. Lancet. 2019;394:861-867.
8. Jiang J, Deng H, Liao H, et  al. An Artificial Intelligence-Enabled ECG Algorithm for Predicting the Risk of Recurrence in Patients with Paroxysmal Atrial Fibrillation after Catheter Ablation. J Clin Med. 2023;12.
9. Vinter N, Frederiksen AS, Albertsen AE, et al. Role for machine learning in sex-specific prediction of successful electrical cardioversion in atrial fibrillation? Open Heart. 2020;7.
10. Rolf S, Kircher S, Arya A, et al. Tailored atrial substrate modification based on low-voltage areas in catheter ablation of atrial fibrillation. Circ Arrhythm Electrophysiol. 2014;7:825-833.
11. Saglietto A, Gaita F, Blomstrom-Lundqvist C, et al. AFA-Recur: an ESC EORP AFA-LT registry machine-learning web calculator predicting atrial fibrillation recurrence after ablation. Europace. 2023;25:92-100.
12. Krittanawong C, Johnson KW, Rosenson RS, et al. Deep learning for cardiovascular medicine: a practical primer. Eur Heart J. 2019;40:2058-2073.