Introduction
The first article by Al’Aref et al. (2019) focuses on discussing ML in the sphere of cardiovascular diseases. The paper provides a brief examination of ML methodologies in cardiology. This paper is a clinical review, a type of research study that discusses a specific topic and makes rational judgments in case of a lack of evidence founded on a reviewer’s expertise. The researchers pose no research questions, and the method is the collection, classification, and summary of contemporary data on the topic. Al’Aref et al. (2019) discuss several applications of ML, including electrocardiography and echocardiography.
Discussion
Moreover, the paper discusses recently designed non-invasive imaging techniques such as coronary computed tomography angiography and coronary artery calcium scoring. In conclusion, the authors consider the limitations linked to the current employment of ML algorithms in the field of cardiovascular diseases (Al’Aref et al., 2019). The significant advantage of the study is the discussion of the drawbacks of machine learning related to remuneration, management, and assessment. Moreover, the study includes advanced analysis from statistical, practical, and theoretical perspectives.
Another resource written by Dudchenko et al. (2020) focuses on reviewing the studies that allocate the machine learning algorithms previously applied within the scope of experimental studies. The population is not identified due to the analysis of the algorithm. The research method is based on the PRISMA statement helping to evaluate the randomized trials found through the PubMed search engine (Dudchenko et al., 2020). The rationale of this work is to find studies where ML algorithms are employed in the field of cardiology and present an overview to help researchers in the development of ML-based systems in the future. The reliability and validity of this work are underpinned by the method and the relevance of the resources utilized. Limitations of this paper include the fact of a limited time period being resorted to when selecting articles, as well as the fact of a single literary base utilized for article searching.
The significant findings of the work show that there are no appropriate ML-based models for the application in the domain of cardiology yet. This is vital for the research paper to estimate the future perspectives of ML in cardiology. The primary strength of the article is that the systematic review could serve as a valuable tool for specialists developing ML-based systems in the medical field, especially in cardiology.
The paper by Cuocolo et al. (2019) focuses on analyzing the recent application of ML in cardiology. The mentioned methods include diagnostic, echocardiography, and quantification of heart function. The body of the article is presented in the form of a literature review study, which estimates the progress of ML implementation in the sphere of cardiology. The study has a record of research analysis having no particular chosen population. The sample included recent articles regarding the application of ML in cardiology (Cuocolo et al., 2019). Findings showed that by implementing ML in cardiology, clinics follow the particular path that the authors of the article schematically depicted. The results show that ML in cardiology contributes to the more efficient treatment and management of patients, improving general health outcomes. The limitations of the offered scheme are in the practical difficulties of working with low-quality data flow.
The primary strength of the article is its significant validity due to allocating a considerable number of experimental research. The major weakness is the lack of the described framework, which makes it challenging to differentiate the authors’ ideas from the opinions of researchers whose articles were used as a sample. The article is relevant to the current research as it provides valuable insights about ML in practical cardiology implementations.
The article by Juarez-Orozco et al. (2019) focuses on applying ML in evaluating myocardial ischemia. The study describes the research in the sphere of the subcategory of ML called deep learning. The research showed that the mechanisms of the mentioned technologies represent remarkable results in diagnosis and prognosis in nuclear cardiology. The authors only review the results without providing any limitations to the study. The findings show that the most efficient techniques of ML in cardiology are SPECT and PET imaging which can improve the quality of myocardial ischemia diagnosis and predict the adverse consequences of illness, minimizing the fatal outcomes (Juarez-Orozco et al., 2019). The validity of the research is conditioned by the fact that this study was conducted by the specialists who participated in the experimental studies addressed in the paper.
The significant advantage of the article is the clear description of the detailed quantitative studies funding in cardiology using the mentioned above techniques. The weakness is the lack of clearly defined limitations and statistical details due to the descriptive type of the research. Despite the drawbacks, the paper is still relevant for the discussed topic as it describes the practical data techniques helping to improve the nuclear cardiology diagnosis.
The authors acknowledge that, despite the potential, the real impact of ML on the work of clinical cardiologists and their patient outcomes has been relatively sparse. No research question is posed, and the method is collecting, systematizing, and summarizing relevant data on the topic. As per Quer et al. (2021), the rationale of their study is to familiarize clinicians who are not experts in data science with ML’s central concepts, which will allow them to better understand the field. The current data on ML in cardiology are then aggregated using select case studies and a bibliometric survey. Finally, a number of ways for clinicians to be involved in this area are proposed. Quer et al.’s (2021) findings show that ML can help test and monitor patients remotely, manage the data acquirement, and send it to a hospital or clinic.
The study has no observable limitations, and its reliability and validity are corroborated by the method and the resources used to form it. The general worth of this study is in its demonstration that the comprehension of the notion of ML and initiatives linked to it is helpful for any cardiologist or specialist in the field. The primary strength of the research is the discussion of cardiologists in machine learning. The weakness of the study is that researchers provide no data regarding the limitations and further research.
Conclusion
The works selected for this literature review do not contradict one another, and one central theme emerges. There are no appropriate ML-based models for the application in the domain of cardiology yet. This is despite the fact that there are numerous potential benefits of such a technology that can transform the field. However, researchers see trends that point to it changing soon and warn that specialists must be prepared to operate under new circumstances.
References
Al’Aref, S. J., Anchouche, K., Singh, G., Slomka, P. J., Kolli, K. K., Kumar, A., Pandey, M., Maliakal, G., van Rosendael, A. R., Beecy, A. N., Berman, D. S., Leipsic, J., Nieman, K., Andreini, D., Pontone, G., Schoepf, J., Shaw, L. J., Chang, H-J., Narula, J.,… & Min, J. K. (2019). Clinical applications of machine learning in cardiovascular disease and its relevance to cardiac imaging. European Heart Journal, 40(24), 1975-1986. Web.
Cuocolo, R., Perillo, T., Petretta, M., Rosa, E., Ugga, L. (2019). Current applications of big data and machine learning in cardiology. Journal of geriatric cardiology, 16(8), 601–607. Web.
Dudchenko, A., Ganzinger, M., & Kopanitsa, G. (2020). Machine learning algorithms in cardiology domain: A systematic review. The Open Bioinformatics Journal, 13(1), 25-40. Web.
Juarez-Orozco, L.E., Martinez-Manzanera, O., Knuuti, J., & Storti, A.E. (2019). Machine learning in the evaluation of myocardial ischemia through nuclear cardiology. Current Cardiovascular Imaging Reports, 12, 1–8. Web.
Quer, G., Arnaout, R., Henne, M., & Arnaout, R. (2021). Machine learning and the future of cardiovascular care: JACC state-of-the-art review. Journal of the American College of Cardiology, 77(3), 300-313. Web.