In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.
ESC Congress 2020 Register

Artificial Intelligence: a Prime Characterisation of the Digital Revolution in Medicine



As discussed in the introduction editorial of the first issue of the Journal of ESC Digital Health (JEDH), artificial intelligence (AI) has been chosen as the theme for this issue as our Editorial Board considers this topic to be a prime characterisation of the digital revolution in medicine.

Figure 1. Figure by Krittanawong et al. Relationship of deep learning to clinical and translational medicine. Venn diagrams show deep learning as one type of machine learning within the scope of artificial intelligence. Statistical methods are applied across clinical and translational science, and the form known as statistical learning theory overlaps with machine learning. Automated decision- making is often used in clinical practice. Deep learning may extend statistical approaches in some key areas by analysing large multivariate data sets that often show complex interactions in which simple hypotheses are difficult to formulate. Deep learning has been successful in medical image recognition (e.g. electrocardiogram, echocardiogram, and magnetic resonance imaging) and holds the promise of enhancing clinic decision-making. Reprinted with permission: Krittanawong C, Johnson KW, Rosenson RS, et al. Deep learning for cardiovascular medicine: a practical primer, European Heart Journal 2019; 40 (25): 2058–2073, doi:10.1093/eurheartj/ehz056.The first curated paper, published in the European Heart Journal, is ‘deep learning for cardiovascular medicine: a practical primer’ by the group Krittanawong et al (3). This article reviews the current state of deep learning (DL) for the cardiovascular clinical and scientific communities by addressing the technical conditions to grasp and understand both the affirmative potentials and the difficult challenges whilst considering the myriad opportunities in this new, appealing field. Figure 1 illustrates nicely the association of deep learning with clinical and translational medicine.

The second paper, by Rajkomar et al., discusses the convenience of deploying AI for general medicine by delineating how valuable AI could be in clinical decision-making to support capitalising on all available patient data (4). A paper by Johnson et al., recently published in the Journal of the American College of Cardiology, links three distinct areas in cardiovascular medicine: namely research and development, clinical practice, and population health (5).

Figure 2. Goldstein et al. One perspective on the intersection of statistical modelling (blue) and machine-learning (green) goals. The figure highlights that while the processes differ, the overarching goals are often the same. Reprinted with permission: Krittanawong C, Johnson KW, Rosenson RS, et al. Deep learning for cardiovascular medicine: a practical primer, European Heart Journal 2019; 40 (25): 2058–2073, doi:10.1093/eurheartj/ehz056.In concurrence with the aforementioned papers, AI relies on statistics in order to execute predictive analytics. The paper by Goldstein et al. (6) – moving beyond regression techniques in cardiovascular risk prediction: applying machine learning to address analytic challenges – delves deeper into the statistical modelling approach necessary to move forward in this AI era. Figure 2 provides a good illustration of this. Eric Boersma provides an interesting perspective.

Imaging modalities have become decisive diagnostic workhorses within cardiovascular treatment strategies. Our colleagues within radiology had the agile foresight to recognise the capabilities of AI within their specialty. Their experiences and learnings have traversed into cardiovascular imaging as deliberated in review by Al’Aref et al (7). This analysis is accompanied by an opinion piece from Filippo Cademartiri.

Figure 3. Brun et al. reproduced with permission. Shared view of heart hologram as used in surgical conference, similar to the evaluation study situation, with one person manipulating the model and the others looking at the process. The image is manipulated to show the heart model as it would be seen by a third spectator on the other side of the room. Reprinted with permission: Brun H, Bugge RAB, Suther LKR, et al. Mixed reality holograms for heart surgery planning: first user experience in congenital heart disease, European Heart Journal ¬ Cardiovascular Imaging 2019; 20(8): 883–888, doi:10.1093/ehjci/jey184Whilst acknowledging that the adult heart is a complex structure, congenital pathologies are inherently more complex. With this in mind, and taking a modest step outside the AI-themed issue, the board could not resist the temptation to select the following paper-based on holography recently published by Brun et al. (8) (figure 3). Although holograms are ‘trending’, are they an effective tool in clinical practice? Ad Bogers seeks to address this contemporary question. A nice link with congenital diseases, big data, and machine learning is the paper by Diller et al.. (9) which illuminates the benefits of these new technologies.

The previously highlighted papers focus on the interactions on a patient level. Parallel to these advances in patient care is the cautious concern whether automation technologies will result in physician job losses. This has led to an ongoing yet lively debate within the radiology arena. Shameer et al. positions these discussions in a broader perspective with the paper titled, Machine Learning in Cardiovascular Medicine: Are We There Yet? (10)  Quer et al. complements this debate with the paper: Augmenting Diagnostic Vision with AI (11).

Somewhat contrasting but interesting, nonetheless, is the paper: Pharma Blockchains AI for Drug Development. This report by the group Burki et al. examines the evolution in performing pharmaceutical trials for new drug developments from the classic formats using animals and humans to in-silico configurations (in-silico defined as solely computer-based) (12). Last but not least are the ethical concerns when employing big data, machine- and deep-learning algorithms, a topic addressed by Prabhu S.P. (13).

While working on the portal and this editorial, we stumbled by chance on a thesis written in 1980 with the title: Computer Analysis of Cardiac Catheterization Data by G.T. Meester, Rotterdam, The Netherlands. He described such automated systems as useful, improving the quality of cardiac catheterisation and providing the clinician with relevant data that would not otherwise be available. He concluded his thesis with an agreeable quote from John Benjamin Murphy (1857–1916), a quote which today is especially pertinent for digital health technologies: ‘The patient is the center of our medical universe, around which all our work revolves, and towards which all our efforts tend.’

References


  1. Bruining N, Barendse R, Cummins P. The future of computers in cardiology: 'the connected patient'? Eur Heart J 2017;38(23):1781-1794.
  2. Cowie MR. Exploring digital technology's potential for cardiology. Eur Heart J 2019;40(28):2283-2284.
  3. Krittanawong C, Johnson KW, Rosenson RS, Wang Z, Aydar M, Baber U, Min JK, Tang WHW, ` Halperin JL, Narayan SM. Deep learning for cardiovascular medicine: a practical primer. Eur Heart J 2019;40(25):2058-2073.
  4. Rajkomar A, Dean J, Kohane I. Machine Learning in Medicine. N Engl J Med 2019;380(14):1347-1358.
  5. Johnson KW, Torres Soto J, Glicksberg BS, Shameer K, Miotto R, Ali M, Ashley E, Dudley JT. Artificial Intelligence in Cardiology. J Am Coll Cardiol 2018;71(23):2668-2679.
  6. Goldstein BA, Navar AM, Carter RE. Moving beyond regression techniques in cardiovascular risk prediction: applying machine learning to address analytic challenges. Eur Heart J 2017;38(23):1805-1814.
  7. Al'Aref SJ, Anchouche K, Singh G, Slomka PJ, Kolli KK, Kumar A, Pandey M, Maliakal G, van Rosendael AR, Beecy AN, Berman DS, Leipsic J, Nieman K, Andreini D, Pontone G, Schoepf UJ, Shaw LJ, Chang HJ, Narula J, Bax JJ, Guan Y, Min JK. Clinical applications of machine learning in cardiovascular disease and its relevance to cardiac imaging. Eur Heart J 2019;40(24):1975-1986.
  8. Brun H, Bugge RAB, Suther LKR, Birkeland S, Kumar R, Pelanis E, Elle OJ. Mixed reality holograms for heart surgery planning: first user experience in congenital heart disease. Eur Heart J Cardiovasc Imaging 2019;20(8):883-888.
  9. Diller GP, Kempny A, Babu-Narayan SV, Henrichs M, Brida M, Uebing A, Lammers AE, Baumgartner H, Li W, Wort SJ, Dimopoulos K, Gatzoulis MA. Machine learning algorithms estimating prognosis and guiding therapy in adult congenital heart disease: data from a single tertiary centre including 10 019 patients. Eur Heart J 2019;40(13):1069-1077.
  10. Shameer K, Johnson KW, Glicksberg BS, Dudley JT, Sengupta PP. Machine learning in cardiovascular medicine: are we there yet? Heart 2018;104(14):1156-1164.
  11. Quer G, Muse ED, Nikzad N, Topol EJ, Steinhubl SR. Augmenting diagnostic vision with AI. Lancet 2017;390(10091):221.
  12. Burki T. Pharma blockchains AI for drug development. Lancet 2019;393(10189):2382.
  13. Prabhu SP. Ethical challenges of machine learning and deep learning algorithms. Lancet Oncol 2019;20(5):621-622.

 

Notes to editor


Editorial on Krittanawong C, Johnson KW, Rosenson RS, Wang Z, Aydar M, Baber U, Min JK, Tang WHW, Halperin JL, Narayan SM. Deep learning for cardiovascular medicine: a practical primer. Eur Heart J 2019;40(25):2058-2073.

Declaration of Interests: The author(s) have declared no conflicts of interest.

The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.