We searched the Web of Science Core Collection database on November 22, 2021 (Textbox 1).

We excluded digitalin*, digitalis*, supplemental digital and digital ulcer* because these words and their derivatives did not refer to digital technology, but instead referred to the drug digitalin, to the plant genus Digitalis, to supplemental digital content, and to the medical condition digital ulcer, respectively. No additional filters were applied to restrict the search results. The search resulted in 12,529 papers. The Analyze Results and Citation Report functions of the Web of Science platform were utilized for basic frequency counts and the number of citations per publication (mean citations per item within a subset) of the most productive authors, institutions, countries, journals, and journal categories. We also defined a subset—literature that included the terms smartphone*, app, or wearable* in the title, abstract, or keywords—which contained 632 papers.

The full record and cited references were then exported into VOSviewer as tab delimited files to synthesize a term map. For clarity, only terms that appeared in at least 0.5% of the literature set (>63) were included in the map. A list of top 5000 common words from the Corpus of Contemporary American English was entered to remove generic (and therefore, less meaningful) words from the term map [85]. VOSviewer was also used to identify the top 20 recurring author keywords.

As the latest digital technology uses often involve smartphone apps and wearable devices, This analysis described above, except for the term map, was similarly conducted on a subset of the concerned.

The 12,529 papers were published from 1965 to November 22, 2021. The earliest publication was a report on the development and demonstration of an analog-digital analyzing unit to screen heart sounds in children [86]. The literature growth seemed to be accelerating in the 2000s and especially into the 2010s (Figure 1). Approximately three-quarters of the papers (9271/12,529, 74.0%) were original articles, and review papers accounted for 6.3% (789/12,529). Proceedings papers and meeting abstracts accounted for 14.2% (1779/12,529) and 6.0% (752/12,529), respectively.

The most productive author was Professor David J Sahn from Oregon Health and Science University, whose highly cited papers were focused on real-time 3D echocardiography [87-89]. Of the 5 most productive institutions, 4 were based in the United States of America, with Harvard University having the highest number of citations per publication. The most productive journals were from the area of cardiology or cardiovascular system, with Circulation having the highest citations per publication among the top 5 (Table 1). Cardiac and cardiovascular systems was the most productive journal category, accounting for nearly one-third of the papers.

The variety of digital technology uses in cardiology can be observed (Figure 2), with uses related to blood pressure (n=727, citations per publication: 20.1), hypertension (n=642, citations per publication: 21.1), arterial stiffness (n=128, citations per publication: 19.4), and stenosis (n=500, citations per publication: 23.7). Terms that appeared in more recent papers included wearable device (n=79, citations per publication: 10.1), smartphone (n=143, citations per publication: 12.0), and COVID (n=111, citations per publication: 3.2) (Figure 2), as well as pandemic (n=74, citations per publication: 3.0), machine learning (n=97, citations per publication: 11.1), artificial intelligence (n=112, citations per publication: 8.8), and app (n=149, citations per publication: 10.3) (not in Figure 2).

The terms telemedicine, digital health, and mHealth (mobile health) were among the top 20 author keywords (Table 2), suggesting that digital technology uses have been a major research focus. Such uses are also of increasing interest in the context of the COVID-19 pandemic.

The literature subset contained 632 papers on smartphone apps and wearable devices. The first paper of this subset was published in 2005, and it introduced a wearable multiparameter (including heart rate and blood pressure) ambulatory physiological monitoring system that could digitally record and continuously stream data to a base station [90]. Approximately 62.2% (393/632) of papers were original articles. Review papers accounted for 15.2% (96/632). The most productive author within this subset was Dr. Mohamed Elgendi affiliated with University of British Columbia and Simon Fraser University. His research interest focused on using data from photoplethysmography (PPG) to detect hypertension, potentially with the aid of machine learning [91,92]. Of the 5 most productive institutions, 4 were based in the United States of America. Furthermore, in the top 5 countries, the first 4 places remained unchanged from the those of the full data set (United States, United Kingdom, Germany, and China). Fifth place was taken by Australia, while Italy (ranked as the fifth most productive country in the full data set) moved to tenth place. JMIR Publications had the top 3 journals, which collectively accounted for more than 10% of the 632 papers (Table 3).

The top 20 author keywords (Table 4), notably, included atrial fibrillation, a cardiac condition that causes rapid and irregular heart rate.

Cardiovascular diseases that were frequently indicated as author keywords in the 12,529 papers included hypertension, atrial fibrillation, atherosclerosis, heart failure, and arterial stiffness. A recent meta-analysis reported that using smartphone app–based interventions could significantly lower blood pressure and improve medication adherence in patients with hypertension [93]. It was found that both wearable, ambulatory, and home monitoring devices recorded blood pressure with comparable values [94]. Smartphone and smartwatch apps could already readily distinguish atrial fibrillation from sinus rhythm and detect them with high sensitivity and specificity comparable to 12-lead electrocardiography (ECG) [95,96]. Authoritative bodies such as the European Society of Cardiology have also developed smartphone apps for patient education on atrial fibrillation [97]. The use of smartphone apps could help general physicians and trainee cardiologists decide whether a patient with heart failure should receive an implantable cardioverter defibrillator or cardiac resynchronization therapy [98]. Researchers found that these apps could potentially reduce hospital staff and facility costs by enabling patients to self-perform simple diagnostic tests, such as the 6-minute walk test, a functional exercise test used to assess patients with cardiopulmonary problems [99]. Similarly, improved access and participation in cardiac rehabilitation in terms of physical activity counselling and exercise training could be achieved by using digital health interventions that were not facility-based [100]. Apart from patient and physicians, digital technology could also target people outside of health care. For instance, massively multiplayer virtual worlds could be modified for use as a serious game to efficiently and reliably teach high school students how to perform cardiopulmonary resuscitation, an act that can be life-saving [101]. Virtual reality, a research hotspot in recent years [16], could also be utilized to teach cardiopulmonary resuscitation for medical students [102].

Meanwhile, recurring investigative modalities highlighted by the current analysis included ECG, angiography, echocardiography, digital subtraction angiography, and PPG. PPG is one of the most heavily researched diagnostic tools, and it is noninvasive, inexpensive, and convenient [103]. It could also be performed with a smartphone to detect heart rate with an average error rate as low as 1 to 1.5% [104]. Applying deep learning to PPG data could also stratify patients’ risk of hypertension [92]. Moreover, artificial intelligence could interpret ECGs rapidly with human-like performance and even detect signals and patterns largely unrecognizable by humans [105]. Overall, use cases, in which physiological parameters from wearable sensing devices are extracted and artificial intelligence is applied to draw insights, are a focal point in the literature; there is a large cluster of prominent terms such as parameter, monitoring, detection, and algorithm (Figure 2). Machine learning methods such as deep learning are frequently used to represent data structures and to make predictions or classifications, with the overall intention of supporting clinicians in data-based decision-making [106]. The expectation is that this will contribute to increasing the efficiency and effectiveness of care delivery, in particular with respect to precision health and personalized care [17]. In fact, digital technology could be very useful, with predictive models and interventions in the personalized management of cardiovascular disease patients for predicting sudden cardiac death, ventricular tachycardia, and ventricular fibrillation [107,108].

During the COVID-19 pandemic, the value of digital technology use under extreme measures for infection control has become evident. For instance, electronic stethoscopes could be utilized for contactless auscultation with real-time playback, digital storage of data, and subsequent data transmission for further assessment [109]. With the reduction of in-person hospital visits, digital technology could facilitate telemonitoring programs to serve as alternative to support patient access to care [110]. Indeed, a recent bibliometric analysis on digital health papers listed telemedicine and telehealth as two of the most frequently used keywords, indicating their relevance beyond cardiology [111].

In the subset of smartphone app and wearable device literature, we found that mHealth, physical activity, and eHealth were among the top author keywords, and most papers had been published in JMIR mHealth and uHealth, JMIR Research Protocols, and Journal of Medical Internet Research. These findings were highly consistent with a recent bibliometric analysis on digital health behavior change technology [59], but where United States, United Kingdom, and the Netherlands had been the most productive countries, in our findings, the Netherlands was replaced by Germany and China. This suggests that there are some geographical differences in research interest between cardiology-specific and general research on health behavior change. Meanwhile, another recent bibliometric analysis on mobile health apps also identified the 3 abovementioned journals as the most productive [112].

In principle, smartphone apps could offer an ideal modality for delivering digital interventions to empower patients’ self-management, by providing health literacy support and coaching content (eg, a smartphone coaching app for blood pressure control [113]). However, in line with findings from this bibliometric analysis, recent reviews [114,115] have highlighted that there is a relative paucity of health literacy interventions and, more specifically, a paucity of digital health literacy interventions for cardiovascular patient groups [116]. Moreover, apps designed to empower patients often include a narrow range of features and lack explicit linkage with theories of empowerment [117,118]. This is an area for further research—the development of content and features for such apps should be based on relevant theoretical underpinnings.

Interestingly, this bibliometric analysis did not identify top-listed terms related to primary prevention or secondary prevention/cardiac rehabilitation of cardiovascular disease. This may seem surprising, since there has been a rapid growth in the development of health apps and other digital technology interventions, and primary and secondary prevention of cardiovascular disease and cardiac rehabilitation are important areas of application [119-121]. A number of recent reviews demonstrate that a sizeable body of literature is available, for example, a systematic review and meta-analysis [122], which included 51 primary studies of digital health interventions for the primary and secondary prevention of cardiovascular disease; a systematic review and meta-analysis [123], which reported on 25 original studies of digital technology interventions for cardiovascular risk factor modification; a scoping review [124], which summarized 13 trials of mobile technology interventions for improving exercise capacity in cardiac rehabilitation; and a systematic review [100], which reported on 31 primary studies of digital health interventions for physical activity and exercise adherence in cardiac rehabilitation. In the context of this bibliometric analysis, this indicates that the literature on digital technology cardiology uses appears more accessible through disease- and condition-specific key terms (hypertension, atrial fibrillation, atherosclerosis, cardiovascular disease, and heart failure), as opposed to more service- and patient pathway–oriented terms (primary prevention, secondary prevention, and cardiac rehabilitation), which may be a relevant consideration in designing literature search strategies for researchers targeting the latter [107,108].

Another aspect of digital technology use in cardiology that was not featured prominently among the findings of this bibliometric analysis is the use of digital technology to increase the efficiency and quality of research in cardiology [125]. This refers to new possibilities afforded by mobile apps, smart devices, and implantable or wearable technologies for the design and management of research studies. Digital processes for data collection, monitoring, communication, documentation and approvals in research hold potential cost and time savings, and functionalities of digital devices open new avenues in the collection and quality control of real-time continuous data acquisition [125] (eg, real-time capture of self-reported measures and symptoms in web-based forms, and the verification of subjective data through concurrent objective measurement, for example, by supplementing subjective reports of physical activity with continuously worn activity tracking devices). The use of digital technologies in the design and management of research studies in cardiology is an emerging focus in the literature, with opportunities for robust evaluations of the advantages of digital research designs over traditional nondigital approaches.

We observed that cardiology journals were predominant in the entire literature set. The top 5 journals were also among those that had published the 100 most cited cardiovascular papers in a previous study [126], with Circulation and European Heart Journal together accounting for 64% of the top 100. However, it should be noted that digital technology use does not only involve cardiology but is an intersection between medical informatics, engineering and health sciences and services in general. With this in mind, when the literature subset on smartphone app and wearable devices was examined, it could be seen that the traditional cardiology journals have given way to newer journals that focus on digital technology and medical informatics. The Journal of Medical Internet Research and JMIR-portfolio journals were found to be the major publishing venues for these papers. Therefore, readers should focus not only on traditional cardiology journals when seeking the latest advancements of digital technology use in the cardiology field.

There are several limitations. First, not all journals (and hence papers) are indexed by Web of Science. Alternative databases are available, each with their own shortcomings. For example, Scopus may contain erroneous data [127], Google Scholar does not allow automated extraction of title and abstract information, and PubMed does not contain citation data. Second, publication and citation counts do not necessarily equate to scientific quality. Within the diverse cardiology research field, the baseline research productivity in particular areas could be inhomogeneous; therefore, the ranking of clinicians or researchers is given for readers’ general reference only. Notwithstanding, this study should allow readers to gain a better understanding of the literature on digital technology uses in cardiology.

Cardiovascular diseases that were frequently investigated in the literature included hypertension, atrial fibrillation, atherosclerosis, heart failure, and arterial stiffness. Recurring investigative modalities included ECG, angiography, echocardiography, digital subtraction angiography, and PPG. Readers searching for relevant information and authors searching for suitable publication venues for their work may consider that, while cardiology or cardiovascular system–focused journals were predominant in the overall literature set, the major publishing venues for the literature subset on smartphone apps and wearable devices were Journal of Medical Internet Research and JMIR-portfolio journals. Digital uses targeted physicians and patients as well as the general public, and their functions included assisting diagnosis, recording cardiovascular parameters, patient education, and teaching laypersons about cardiopulmonary resuscitation. The scientific body of literature on digital technology use in cardiology is rapidly growing, and its impact on health care is also expected to greatly increase in the near future.