From Diagnosis to Management: Unveiling the Challenges of Artificial Intelligence Solutions in Cardiovascular Healthcare

From Diagnosis to Management: Unveiling the Challenges of Artificial Intelligence Solutions in Cardiovascular Healthcare | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article From Diagnosis to Management: Unveiling the Challenges of Artificial Intelligence Solutions in Cardiovascular Healthcare Valentine Idakwo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4370656/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Cardiovascular diseases (CVDs) are the leading cause of mortality in the world. Artificial Intelligence (AI) offers an opportunity to improve the quality of care provided to cardiovascular patients due to its ability to handle large and complex data. Despite promising results obtained in several studies, widespread adoption of AI in cardiovascular care is lacking due to the existence of some gaps. The goal of this study is to analyze the existing challenges faced by AI solutions in cardiovascular care. This study adopted a mixed-methods research approach, combining semi-structured interviews with responses from a self-administered online survey. A total of 5 interviews were conducted and 91 valid survey responses were obtained. Survey respondents included doctors, nurses, medical researchers, health I specialists, hospital administrators, and other clinically affiliated participants working with cardiovascular patients. Participants identified 8 major challenges: data-related challenges, regulatory challenges, infrastructural challenges, gaps in knowledge, transparency challenges, ethical challenges, issues with change management, and acceptance challenges. These gaps hinder the adoption of AI in cardiovascular care and taking proactive measures to address these challenges is key to fostering AI adoption. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Cardiovascular diseases (CVDs) are the leading cause of mortality globally, accounting for nearly 19 million deaths in 2019 ( 1 ). This number is projected to increase due to the ageing demographics and the lack of healthcare workers ( 2 ). Artificial Intelligence (AI) offers a path in helping clinicians and healthcare institutions manage the burden of CVDs through risk stratification, early diagnosis, clinical decision support, patient monitoring and personalized medicine. Some studies have shown promising results in several applications of AI in cardiovascular care ( 3 – 7 ). Examples of these studies include the use of a machine learning (ML) model to predict the occurrence of in-hospital cardiac arrests using the variability in heart rates and the use of an ML model for the risk stratification of occlusion myocardial infarction ( 8 , 9 ). Despite these promising results, the adoption of AI in clinical practice is significantly low due to the novelty of the field and the existence of several challenges hindering its adoption ( 5 ). Although cardiovascular diseases (CVDs) have a significant impact on healthcare, there is a scarcity of experimental studies investigating the challenges associated with adopting AI in clinical cardiovascular care. A mixed-methods research, by Schepart et al. ( 10 ), identified the existence of 5 major challenges of AI in cardiovascular care: insufficient knowledge, limited useability, limited funding, inadequate compatibility with electronic health records, and trust issues. It is important to note that the scope of Schepart et al.'s study was confined to cardiologists and health IT specialists, which may not reflect the perspectives of the broader healthcare community. The goal of this study is to better understand the challenges hindering the adoption of AI in cardiovascular care through a mixed-methods study of healthcare workers and stakeholders involved in cardiovascular care. A larger group including doctors, nurses, medical researchers, hospital administrators, health IT specialists, cardiovascular assistants, and other stakeholders in cardiovascular care was adopted to offer a more comprehensive view of the challenges faced. Involving a diverse cohort of healthcare professionals provides a multifaceted understanding of the challenges, ensuring that solutions are robust and widely applicable. By understanding these challenges, targeted interventions could be developed to enhance the adoption of AI, ultimately improving patient outcomes in cardiovascular care. Methodology Study Design In this study, a sequential mixed methods research design, with initial semi-structured interviews of several stakeholders in cardiovascular care followed by a self-administered online survey of a wider group, was used. Results from the interviews were used to develop the quantitative survey. This approach allowed the capturing of in-depth perspectives from key stakeholders in cardiovascular care and the corroboration of these findings with quantitative evidence from a wider community of healthcare professionals. Qualitative Study Participant selection and recruitment Medical Doctors, nurses, health IT specialists, clinical scientists, medical researchers, hospital administrators, and cardiovascular AI specialists were eligible to participate in the interviews. The primary inclusion factors were: Knowledge of AI in cardiovascular care Over 3 years of experience in cardiovascular care or AI solutions in cardiovascular care. Participant recruitment was conducted using purposive sampling. This approach was chosen to ensure that the participants could provide in-depth and informed perspectives directly relevant to the research objectives. Eligible participants were recruited and identified by the researchers and invited to participate in the research via email and LinkedIn Data collection A semi-structured interview guide was developed following an extensive literature review of the challenges of AI in medical care. The interview guide was tested on a demographically similar subset of the target population. Feedback from this group led to refinements in question wording and order, optimizing clarity and response efficacy. All interviews lasted 30-40 minutes and were conducted online, via video conferencing, with one interviewer in November 2023. The interviewer asked follow-up questions for clarification and to allow participants to elaborate on the responses provided. The interviews were recorded, transcribed verbatim, and inductively analyzed to identify major themes and additional sub-themes. Data analysis The transcribed text was subjected to inductive thematic analysis. This entailed a systematic process of coding the data in multiple rounds, allowing themes to emerge directly from the participants' narratives without the constraint of pre-existing theoretical frameworks. Initial codes were generated by reading and re-reading the transcripts, which were then grouped into potential themes reflecting the core meanings evident in the data. These themes were reviewed and refined iteratively, ensuring they accurately represented the views of participants. Quantitative Study Participant selection and recruitment Medical Doctors, nurses, health IT specialists, clinical scientists, medical researchers, hospital administrators, and cardiovascular AI specialists were eligible to participate in the surveys. A mix of purposive sampling and snowball sampling was used to ensure that participants met the eligible professional criteria and to reach a larger portion of the target population. Participants were invited to participate in the survey via email and LinkedIn. Data collection A 10-minute online survey was created using Soscisurvey (Version 3.5.00) developed by SoSci Survey GmbH based on the key themes identified in the qualitative analysis. The questionnaire was developed in English and pretested with 7 individuals who were demographically similar to the target population. Feedback from the pretest indicated that certain questions in the survey were ambiguous and lacked clarity. Consequently, the highlighted questions were refined to be more precise and understandable. A screener question was employed to identify and filter out inattentive responses from participants. The survey link was shared with the participants and was online for 6 weeks from November to December 2023. The survey results were subsequently downloaded in Comma-Separated Values (CSV) format and subsequently analyzed. Data analysis The data was initially cleaned to remove incomplete responses and responses that incorrectly answered the screener. As responses were numerically coded automatically be the survey platform, additional coding was not necessary. Descriptive statistics were employed to gain insights into the characteristics of respondents and for univariate analysis. The non-parametric tests, Kruskal-Wallis and Mann-Whitney with Bonferroni correction, were used for pairwise comparisons. The analysis was done using Python libraries: Pandas (version 1.4.2), SciPy (version 1.8.1), and NumPy (version 1.23.3) due to their versatility, efficiency, and extensive capabilities for handling and analyzing large datasets. Statistical significance was set at p- value ≤ 0.05. Qualitative Results Description of participants Five experts representing diverse sectors within the cardiovascular care domain were interviewed: PT1, a medical doctor and AI researcher; PT2, the CEO of an AI-focused CVD company; PT3, a clinical scientist specializing in CVD care; PT4, a digital health hospital consultant; and PT5, an interventional cardiologist with experience in AI-assisted cardiovascular procedures. Identified Challenges Eight major challenges were identified by participants during the interviews. The findings are summarized in Table 1 and described below. Data-Related Challenges : Participants identified several issues related to data that existed and needed to be addressed. According to them, data is vital for training AI algorithms in cardiovascular care and issues with data affect the quality of the algorithms developed. Challenges related to data integration and data access were highlighted by participants. Data for training and validating AI models is scarce. However, this data scarcity is artificial as participants acknowledged that data exist in different cardiovascular institutions. Issues like data fragmentation, data silos, differences in regulation, and variations in annotation make accessing this data difficult. Participants also highlighted that cardiovascular care requires multimodal data for the diagnosis of CVDs. However, existing AI solutions do not allow for the integration of multimodal data in the course of treatment, posing a challenge to their use in cardiovascular care. Regulatory Challenges: Participants also highlighted unclear regulations and the inflexibility of existing regulations as the primary regulatory challenges to AI in cardiovascular care. Existing AI regulations are vague and do not have specific medical or cardiovascular applications. This lack of clarity results in a long and arduous process for obtaining regulatory approval for AI solutions in cardiovascular care. Additionally, participants indicated that existing regulations require developers to freeze their algorithms to obtain regulatory approval and developers have to apply for a new approval once changes are made to the algorithms. Participants believed that this poses a challenge as AI algorithms require constant training with real-world data to be safer and more efficient. Requiring regulatory approval each time an algorithm is trained results in an overly burdensome process. Infrastructural Challenges: Participants also acknowledged the dearth of human and technological infrastructure for the integration of AI in cardiovascular care. Having individuals with the right knowledge to develop and maintain AI infrastructure is key to AI use in clinical practice. However, participants indicated that most healthcare institutions lack the necessary IT department, equipped with an understanding of how AI systems work. In addition, healthcare facilities largely use legacy systems that are incompatible with the latest AI technology and participants believe that this hampers the use of AI in cardiovascular care. Participants also highlighted the existence of a rural-urban divide regarding infrastructure. While healthcare institutions in large urban areas are actively researching and investing in improving their AI infrastructure, healthcare institutions in rural areas are largely not focused on AI infrastructure. Knowledge Challenges: Participants highlighted a bidirectional knowledge gap between healthcare professionals and developers. They noted that healthcare professionals lack sufficient understanding of AI to effectively communicate its mechanisms to patients, while developers often lack the medical expertise necessary for creating clinically relevant applications. Participants also acknowledged the lack of AI training in medical and medically-affiliated curricula as one of the primary reasons for this gap. However, they also acknowledged the existence of advanced professional courses for individuals willing to improve their knowledge of AI. Transparency Challenges: A lack of explainability of existing AI solutions was identified as a primary challenge of AI in cardiovascular care by participants. They indicated that the transparency of AI models is key for legal and liability protection and existing AI models in cardiovascular care do not offer sufficient explanation of their decision-making process and functionality. Explainability is also key for regulatory compliance and entry into healthcare. Hence, the existence of black-box models does not spark confidence among clinicians and regulators, which results in slower adoption in clinical practice. Ethical Challenges: Participants highlighted fairness in data collection, lack of accountability, and vagueness of responsibility as the primary ethical challenges faced by AI in cardiovascular care. They indicated that training datasets used in the development of AI solutions are not representative of the target populations. Hence biases are replicated in the results of these models. They also believe that questions on responsibility pose a challenge to the integration of AI in clinical care. While traditional medical care places responsibility primarily on clinicians when negative outcomes occur, this is not clearly outlined when an AI algorithm is primarily responsible for the decisions made. Hence, reluctance exists amongst clinicians on the use of AI in cardiovascular care. Participants also believe that AI developers are not transparent enough in reporting the developmental process and the steps leading to the creation of the final product. They believe that this lack of accountability diminishes the trust of clinicians and their willingness to adopt AI solutions in cardiovascular care. Change Management Challenges : Participants acknowledged the lack of quality change management plans and the lack of medical and economic impact analysis for AI solutions. They believe that the integration of AI in cardiovascular care is significantly slowed as healthcare institutions lack a comprehensive and coherent plan on how to integrate AI in cardiovascular care. One of the key aspects of change management plans is the impact analysis. Participants indicated the lack of in-depth impact analysis that offers genuine insights into the effect of AI solutions on clinical care and administration. This lack of coherent change management plans creates hurdles for medical institutions and governments in the integration of AI solutions in cardiovascular care. Acceptance Challenges : Participants acknowledged that acceptance by healthcare professionals and patients was a key factor in the integration of AI in cardiovascular care. They believed that some skepticism exists among healthcare professionals on the use of AI in cardiovascular care due to the reluctance to change established ways of working, concerns about job security, and lack of trust in AI solutions. It is important to note that participants do not acknowledge this as being the majority opinion but they acknowledged that the minority opinion still impacts the integration of AI in cardiovascular care. [Table 1] Quantitative Results Description of participants Of the 134 individuals who initiated the survey, 94 (70.1%) completed the survey in its entirety. However, three participants (3.2% of completed surveys) provided incorrect responses to the screener and were subsequently excluded from the analysis. The decision to exclude three participants who provided incorrect responses to the screener was made to maintain data integrity and ensure the reliability of the study findings. Thus, a total of 91 valid responses were obtained for analysis. Although the majority of participants (n= 55, 60.4%) were doctors, the responses also included nurses, medical researchers, health IT specialists, hospital administrators, medical assistants, and cardiovascular technologists. The distribution of participants across different job descriptions is visualized in Figure 1. The majority of participants were from Europe (n = 56, 61.5% of valid responses), followed by Africa (n = 24, 26.4%), Asia (n = 7, 7.7%), and North America (n = 6, 6.6%). Notably, no responses were received from Australia or South America. Participants were also asked about their frequency of using AI for work. The responses indicated that 42.6% (n = 38) reported never using AI, 23.6% (n=21) reported monthly usage, 16.9% (n=15) reported weekly usage and another 16.9% (n=15) reported daily usage. Data-related Challenges : In terms of data access, the majority of respondents (51.5%) reported encountering difficulty (45.3% found it difficult, while 6.2% found it very difficult) in accessing cardiovascular health data for AI analysis and interpretation. 1.6% found data access to be very easy and 12.5% found it to be easy. 34.4% of respondents maintained a neutral position on data access. A pairwise comparison with Kruskal Wallis test indicated the presence of a statistically significant correlation [H(4) = 9.10, p = 0.028] between the frequency of use and the opinion regarding data access. Further exploration with post-hoc Mann-Whitney test with Bonferroni correction indicated the presence of a statistical difference ( p= 038) in data access views between daily AI users and monthly AI users. Figure 2 shows a box plot visualization in which daily AI users consistently rated data access as more challenging, compared to the more varied responses of monthly AI users. On the compatibility of different data sources for CVD analysis, 36.5% indicated that data sources for CVD analysis were incompatible (4.8% found them highly incompatible and 31.7% found them incompatible). 36.5% maintained a neutral position while 23.8% and 3.2% found data sources for CVD care to be compatible and highly compatible respectively. Regulatory Challenges : To measure the clarity of existing regulations for AI solutions in cardiovascular care, participants were asked about the transparency and comprehensibility of existing regulations. The majority of respondents (55%) indicated that existing regulations were not transparent or comprehensible (41.7% expressed disagreement and 13.3% expressed strong disagreement). 35% maintained a neutral position and 10% agreed that existing regulations were transparent and comprehensible. Notably, no participant expressed strong agreement. Participants were also asked about the adequacy of the current regulatory framework in facilitating the safe and efficient implementation of AI solutions in cardiovascular care. 40.7% and 5.1% expressed disagreement and strong disagreement with the notion that existing AI regulations were adequate. 37.3% remained neutral and 16.9% expressed agreement. Notably, there were no participants who strongly agreed. Infrastructural challenges : Regarding organizational readiness for dealing with the introduction of AI in cardiovascular care, 24.7% of respondents indicated that their organizations were not at all equipped. 39.3% and 30.2% of respondents considered their organizations to be slightly equipped and moderately equipped respectively. Only 5.6% indicated that their organizations were very equipped. Notably, none of the participants regarded their organizations as extremely equipped. The Kruskal-Wallis test showed a significant difference [H(3)= 7.893, p = 0.048] in organizational readiness based on geographical location. Post-hoc analysis using Mann-Whitney test with Bonferroni correction indicated a marginal difference ( p = 0.063) between the respondents in Europe and Africa. The boxplot visualization in Figure 3 shows that respondents from Africa consistently indicated lower levels of organizational readiness compared to the varied responses of respondents in Europe. Knowledge Challenges: The question asking participants to rate their knowledge of AI showed less than optimal levels in the majority (68.2%) of participants ( 3.3% had very poor knowledge, 18.7% had below-average knowledge, and 46.2% had average knowledge). The frequency and distribution of participants’ knowledge levels are shown in Figure 4. Additionally, participants were asked if they would be willing to take courses to improve their knowledge of AI. 78% of respondents were willing to take courses to improve their knowledge of AI, 7.7% were not willing to take additional courses and 14.3% were unsure. Transparency Challenges: On the importance of transparency, 82.4% of respondents indicated that AI systems needed to offer some form of explainability in their decision-making process (50.5% expressed agreement and 31.9% expressed strong agreement). 12.1% maintained a neutral position while 4.4% and 1.1% expressed disagreement and strong disagreement respectively. Additionally, on the impact of transparency on user trust, the majority of respondents (62%) indicated that they needed to understand the clinical decision-making process of AI systems to trust their recommendations (36.3% agreed and 26.4% strongly agreed). Conversely, 20.9% of respondents did not perceive understanding the decision-making process of AI systems as impactful on their trust in their recommendations. Specifically, 15.4% disagreed and 5.5% agreed. 16.5% of respondents were neutral. Ethical Challenges: On fairness, participants were asked the importance of AI solutions considering patient diversity in cardiovascular care. The majority of respondents (52.2%) considered it extremely important for AI solutions to consider patient diversity. 18.9% considered it to be important and 18.9% maintained a neutral position. 7% and 3.3% of respondents considered it to be slightly unimportant and not important at all respectively. Subsequently, participants were asked if existing AI solutions adequately addressed the diversity of patient population. 10.2% of participants responded in the affirmative, 38.6% maintained a neutral position and 51.2% responded in the negative. Change Management Challenges: 3% and 34.2% of respondents considered having an organizational plan for the integration of AI in cardiovascular care to be important and extremely important respectively. 19.7% of respondents took a neutral position. 10.5% and 5.3% indicated that having a plan was slightly unimportant and not important at all respectively. When asked about the quality of their organizational plans for the integration of AI in cardiovascular care, none of the respondents indicated that their organizational plans were fully optimized and comprehensive. 6.6% and 19.7% indicated that their organizational plans were well-developed and moderately developed respectively. 34.2% indicated that their organizational plans were poorly developed while 39.5% indicated that their organizational plans were non-existent. Acceptance Challenges : Participants were asked several questions to gauge their perspective of AI. Figure 5 contains a comprehensive overview of the distribution of responses. Discussion This mixed-methods study was conducted to identify the gaps in AI solutions for the diagnosis and management of CVDs. The interviews and surveys included diverse health professionals involved in the care of cardiovascular patients. This diverse group was selected to offer a comprehensive view of the challenges. To my knowledge, this is the first study to assess such a diverse group of health professionals in cardiovascular care. In the interviews, several gaps were identified that aided the development of the ensuing quantitative research. Specifically, participants identified data-related, regulatory, infrastructural, knowledge, transparency, ethical, change management, and acceptance challenges. Identifying these gaps underscores the complexity of integrating AI solutions into the diagnosis and management of CVDs. On data, participants highlighted the importance of vast amounts of data in the introduction of AI in cardiovascular. However, challenges related to accessing this data and integrating data from different sources persist. The position on data access was corroborated by the subsequent survey of the larger population as a majority of participants indicated the existence of some difficulty in data access. The survey also showed that increasing the use of AI exposes users to more of the challenges involved in accessing data needed for AI development. On the compatibility of different data sources, the responses of participants were more varied. A substantial portion of respondents indicated the presence of compatibility issues and a little under a third of respondents found data sources to be compatible, indicating a diversity of experiences on this issue. Additionally, a significant portion of respondents chose to maintain a neutral stance, indicating a sizable segment of the surveyed population withholding strong opinions or possibly lacking sufficient knowledge or experience to form definitive judgments on data compatibility. Unclear regulations and static regulations were identified by interviewees as the primary regulatory challenges being faced by AI in cardiovascular care. These regulatory challenges hinder the seamless integration of AI in cardiovascular care. The lack of clarity of regulations was corroborated by the survey as the majority of respondents believed that existing regulations were not transparent and comprehensible. Over a third of respondents maintained a neutral position indicating the absence of strong opinions or the lack of sufficient knowledge on the topic. Additionally, a significant portion of respondents also indicated that existing regulations are not adequate to facilitate the safe and efficient implementation of AI solutions in cardiovascular care. Further research is required to understand the specific reasons behind this view on adequacy. The successful introduction of new technology in cardiovascular care is dependent on the existence of the right infrastructure to aid in its integration. Interviewees indicated that existing healthcare institutions lacked the necessary human and technological infrastructure for the successful integration of AI in cardiovascular care. The existence of a rural-urban divide regarding infrastructure was also highlighted in the interview and would require additional research to understand the reasons behind this divide. The position on the lack of infrastructure was also highlighted in the survey as over 90% of respondents indicated that their organizations were either not equipped at all or insufficiently equipped. The survey also showed that participants in Africa consistently reported lower levels of organizational readiness than their counterparts in Europe, indicating that different geographical locations would need to adopt different methods to address this challenge. Interviewees also indicated a prevalent knowledge gap between healthcare professionals and developers. They indicated that medically-affiliated curricula lack relevant AI-related training. Hence, medical professionals lack the necessary understanding of AI to effectively work with it or communicate its mechanisms to patients. Exploring the reasons behind the absence of AI-related training in medical curricula, such as competing educational priorities or limited faculty expertise in AI, would provide valuable insights into the systemic challenges contributing to the knowledge gap. The interview findings were similar to the findings of the survey, in which most participants reported average or less than average knowledge of AI. However, the survey also revealed that the majority of respondents were also willing to take additional courses to improve their knowledge of AI. Interviewers highlighted the existence of transparency gaps with the prevalence of unexplainable models impacting the level of confidence among clinicians and regulators. Although survey respondents overwhelmingly agreed on the importance of explainability, there was a little less consensus on whether the explainability of AI impacts their level of trust in the recommendations offered by these solutions. Furthermore, while the survey results do not explicitly confirm the existence of transparency challenges, they reveal that consensus has not been achieved on the impact of transparency or the level of transparency required for fostering trust in AI recommendations. Further elaboration on the lack of consensus regarding the impact of explainability on trust in AI recommendations could include exploring factors contributing to this discrepancy, such as varying levels of familiarity with AI technology among healthcare professionals or differing perspectives on the importance of interpretability versus predictive accuracy. Fairness in data collection, lack of accountability and lack of clarity regarding responsibility were highlighted in the interviews as the primary ethical challenges. All three issues impact the level of trust that healthcare professionals have in AI solutions. Bias in data collection also results in faulty models as these biases are replicated in the model output. The survey results show that healthcare professionals understand the importance of fairness for AI solutions. However, the majority of respondents also indicated that existing solutions do not adequately address fairness in the patient population. Future research might explore potential strategies for enhancing accountability and clarity regarding responsibility in AI-driven healthcare systems, as well as developing methodologies to mitigate bias in data collection and ensure fairness in patient population representation. Interviewees also indicated the lack of quality change management plans and the lack of medical and economic impact analysis as key change management challenges for AI in cardiovascular care. The lack of a comprehensive change management plan for healthcare institutions results in a chaotic adoption process and diminishes the enthusiasm for adoption. Additionally, the lack of impact analysis diminishes the trust of stakeholders in introducing AI solutions into their established workflow. The lack of plans was also highlighted in the surveys as the majority of respondents indicated that their organizational plans for AI introduction were either non-existent or insufficiently developed despite overwhelmingly agreeing on the importance of having an organizational plan. Future research needs to explore the unique requirements and challenges of AI change management plans and impact analysis in healthcare settings, particularly in comparison to the protocols and considerations involved in the introduction of new medications. Interviewees also indicated the existence of some skepticism amongst end users for AI solutions in cardiovascular care. They highlighted that although this skepticism was not the majority position, it was impactful as overwhelming acceptance is required for the introduction of AI in cardiovascular care. This position was also confirmed in the in survey with the majority of respondents holding positive views on the impact of AI on the diagnosis and treatment of cardiovascular diseases, treatment errors, worker shortage, and patient benefits. On the trustworthiness of AI solutions, most participants maintained a neutral position underscoring the lack of positive or negative leanings on the trustworthiness of AI. Despite most participants indicating that AI should be used in the diagnosis and treatment of CVDs, about one-sixth of respondents indicated that AI should not be used in cardiovascular care, highlighting a divergence of opinions within the surveyed population regarding the appropriateness or efficacy of AI solutions in this context. The summary of findings is contained in Fig. 6 . The findings of this study are similar to other studies exploring the challenges of AI in cardiovascular care and AI in general. The survey by Schepart et al.( 10 ) also indicated the existence of minority skepticism toward AI from cardiologists and health IT professionals. Additionally, the 80% of respondents who highlighted the importance of transparency in trusting AI solutions from that study indicate a larger consensus on transparency than that expressed in this study. The lack of knowledge in AI by medical professionals was also highlighted in several studies of AI in cardiovascular care and medicine as a whole indicating that these problems are not confined to cardiovascular care ( 10 – 12 ). Our study extends the findings of challenges of AI in cardiovascular care by sampling a more diverse group of healthcare professionals involved in cardiovascular care. Limitations The sample size and sample distribution are limitations to this study. The relatively small sample size for both the interviews and the survey limits the generalizability of the conclusions generated. Despite this limitation, the focused nature of the sample allows for in-depth exploration of perspectives and insights from a diverse range of healthcare professionals involved in cardiovascular care, providing rich qualitative data and nuanced perspectives on the challenges and opportunities associated with the integration of AI solutions in this context. Additionally, our survey sample composition presents a skew, with the majority of respondents being doctors. This overrepresentation of one group might introduce bias and limit the perspectives of other healthcare professionals. While the over-representation of doctors might introduce bias, it positions the study to capture in-depth insights from some of the primary users of cardiovascular AI technology. The geographical skew of participants toward Europe potentially limits the global applicability of the research conclusions. These limitations highlight the need for a cautious interpretation of the study’s results and suggest the opportunity for extensive research to expand upon the findings. Conclusion In conclusion, this mixed-methods study provides valuable insights into the challenges of AI solutions in cardiovascular care. Through interviews and surveys with diverse healthcare professionals, key data-related, regulatory, transparency, acceptance, change management, knowledge, and infrastructural gaps were identified. These findings emphasize the complexity and multifaceted nature of implementing AI in cardiovascular settings. Future research should focus on developing tailored approaches to address the identified challenges. Overall, this study contributes to the growing body of literature on AI in healthcare by providing nuanced insights into the complexities of integrating AI solutions in cardiovascular care. By addressing these challenges and leveraging the opportunities presented by AI, we can strive towards more effective and personalized approaches to the diagnosis, management, and treatment of cardiovascular diseases, ultimately improving patient outcomes and advancing the field of cardiovascular medicine. Declarations Supplementary information The supplementary documents for this article can be found online at: Acknowledgments I thank all the experts who took part in the data collection and research development for this study. I also specifically thank Mrs. Stefanie Werkman, Professor Dr. Stefan Kääb, Professor Dr. Solveig Vieluf, and Dr. Christian Gölz for their role in pre-testing and data collection. Additionally, I would like to thank Professor Dr. Georgi Chaltikyan and Professor Mouzhi Ge for their support on this research. Authors’ contributions VI: conceptualization, methodology, interview and survey design, data processing, visualization, analysis, and writing the original draft. Ethics approval and participants’ consent Informed consent was obtained from all participants involved in the study. The research adhered to relevant guidelines and regulations, including those governing research practices at Deggendorf Institute of Technology. As the study did not involve interventions falling under German law's requirement for additional ethics approval, such approval was not sought. Funding Declaration The author declares that no external funding was received for the research, authorship, and/or publication of this manuscript. Conflict of interest Author have no conflict of interest relevant to the context of this article to declare References Mariachiara Di Cesare, Honor Bixby, Thomas Gaziano, Lisa Hadeed, Chodziwadziwa Kabudula, Diana Vaca McGhie, et al. World Heart Report 2023: Confronting the World’s Number One Killer. World Heart Federation. Geneva, Switzerland: World Heart Federation; 2023. Roth GA, Mensah GA, Johnson CO, Addolorato G, Ammirati E, Baddour LM, et al. Global Burden of Cardiovascular Diseases and Risk Factors, 1990–2019: Update From the GBD 2019 Study. Vol. 76, Journal of the American College of Cardiology. 2020. Langlais ÉL, Thériault-Lauzier P, Marquis-Gravel G, Kulbay M, So DY, Tanguay JF, et al. Novel Artificial Intelligence Applications in Cardiology: Current Landscape, Limitations, and the Road to Real-World Applications. J Cardiovasc Transl Res. 2023;16(3):513–25. Krittanawong C, Johnson KW, Rosenson RS, Wang Z, Aydar M, Baber U, et al. Deep learning for cardiovascular medicine: A practical primer. Vol. 40, European Heart Journal. 2019. Karatzia L, Aung N, Aksentijevic D. Artificial intelligence in cardiology: Hope for the future and power for the present. Vol. 9, Frontiers in Cardiovascular Medicine. 2022. Feeny AK, Chung MK, Madabhushi A, Attia ZI, Cikes M, Firouznia M, et al. Artificial Intelligence and Machine Learning in Arrhythmias and Cardiac Electrophysiology. Vol. 13, Circulation: Arrhythmia and Electrophysiology. 2020. Siontis KC, Noseworthy PA, Attia ZI, Friedman PA. Artificial intelligence-enhanced electrocardiography in cardiovascular disease management. Nat Rev Cardiol. 2021;18(7):465–78. Al-Zaiti SS, Martin-Gill C, Zègre-Hemsey JK, Bouzid Z, Faramand Z, Alrawashdeh MO, et al. Machine learning for ECG diagnosis and risk stratification of occlusion myocardial infarction. Nat Med. 2023;29(7):1804–13. Lee H, Yang HL, Ryu HG, Jung CW, Cho YJ, Yoon S Bin, et al. Real-time machine learning model to predict in-hospital cardiac arrest using heart rate variability in ICU. NPJ Digit Med. 2023;6(1):215. Schepart A, Burton A, Durkin L, Fuller A, Charap E, Bhambri R, et al. Artificial intelligence–enabled tools in cardiovascular medicine: A survey of current use, perceptions, and challenges. Cardiovasc Digit Health J. 2023;4(3). Scheetz J, Rothschild P, McGuinness M, Hadoux X, Soyer HP, Janda M, et al. A survey of clinicians on the use of artificial intelligence in ophthalmology, dermatology, radiology and radiation oncology. Sci Rep. 2021;11(1):5193. Castagno S, Khalifa M. Perceptions of Artificial Intelligence Among Healthcare Staff: A Qualitative Survey Study. Front Artif Intell. 2020;3. Tables Table 1: Challenges of AI in Cardiovascular Care Main Theme Sub-Theme Quotes from Interview Participants Data Data access “The first one is access. Training an AI model requires a lot of data and it is very difficult in many cases to acquire a lot of data”- PT1 “Data is key for training AI algorithms and having sufficient data for training is one of the key challenges” - PT2 Data integration “Another challenge lies in connecting data points, so not just integrating EHR data with ECG data, imaging data, vital signs, and voice analysis. So not just looking at one data source but interpreting everything in the context of another”- PT1 Regulations Unclear regulations “Regulatory approval for medical AI is more complex. The process is often long and requires evidence that the AI’s recommendations are at least as good as, if not better than that of a human expert.”- PT5 Static regulations “Another problem is that regulatory bodies ask AI companies to freeze their algorithms for approval. I think this is a bit of a hurdle and also an inconsistency because they want safer and better products but then they hinder them because they don’t know how to or because they require freezing.” - PT1 Infrastructure Lack of infrastructure “The infrastructure, in terms of know-how and competency of staff, the technology, the service, and performance, is largely missing for AI in cardiovascular care.” - PT4 “Many healthcare facilities use legacy systems that aren’t readily compatible with the latest AI technology”- PT5 Rural-urban divide “At this stage, AI progress is concentrated on large university hospitals or well-equipped hospitals, and on the startup market in urban areas, but not in the outskirts or rural areas.” -PT5 Main Theme Sub-Theme Quotes from Interview Participants Knowledge - “When I spoke to some healthcare workers, I noticed they didn’t know anything. They hadn’t even tried ChatGPT and had no hands-on experience with AI. This is one side of the spectrum”- PT4 “I see that there is some knowledge available on what it means. The basic terms are quite common these days, like on convolutional neural networks, but most people have limited understanding of what it means.”- PT2 “We generally see a lack of education and training on how to work with AI, or interpret its recommendations.”- PT5 Transparency - “One of the major challenges is explainable AI and understanding machine reasoning and also in terms of legal issues and liability. This is still one of the biggest hurdles that must be defined and needs to be understood.”- PT2 “There is a black-box phenomenon for different AI solutions in healthcare. This black-box phenomenon- you can’t have it in medical treatment.”- PT4 Ethics Fairness “Of-course there is the issue of biases in terms of patient recruitment for training data. I think there are some good practices that need to be followed in order to ensure that the training datasets are representative of the population that is being treated or diagnosed. So, I think that’s an aspect that needs to be accounted for”- PT1. Responsibility “There is the question of accountability when AI is used in diagnosis and treatment- Who is responsible if something goes wrong?”- PT5. Accountability ” Even though we have dedicated guidelines indicating and structuring how data treatment should be performed, some care providers do not follow these recommendations or report the model development process”- PT2. Main Theme Sub-Theme Quotes from Interview Participants Change Management Impact analysis “Medical and economic impact is the core of digital health and also AI models. We have to quantify what we bring from a medical perspective and an economic perspective”- PT3 “Regarding the impact measurement, I rarely see complete views. I call this the net present value. It’s basically a comprehensive view on what the algorithm does and what the whole infrastructure requires”- PT3 Plans “Certainly, change management is obligatory, especially when introducing a new way of working in clinical practice. One of the major challenges is not the technical part but how to introduce new solutions to an established way of working, that fits in clinical practice without an extra cost. Most healthcare institutions lack this plan”- PT2 Acceptance - “Minor barriers could be acceptance by patients, and physicians. People are interested and believe that the technology can do better but there are still people that don’t want their data analyzed or stored.”- PT1 “Experts in their field have an established way of working and some of them are struggling with new techniques and not seeing AI as a threat”- PT2 “Healthcare professionals may be skeptical or uncomfortable relying on AI for decisionmaking. This resistance can stem from concerns about job security, or doubts about AI capabilities.”- PT5 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 13 May, 2024 Submission checks completed at journal 08 May, 2024 Editor assigned by journal 08 May, 2024 First submitted to journal 05 May, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4370656","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":301819999,"identity":"ef1024ac-d1e5-4376-8ff7-01f466e64192","order_by":0,"name":"Valentine Idakwo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDUlEQVRIiWNgGAWjYFAC5gYgIQGkGQwYgUw5kNiBB3i1gNQlILQYg7UkENYCZoG1JIIsZcCnxeD4wcbHhT8s8vjZmTd+nNlmkz4/7PBDoC12croNOLScSWw2npEgUSzZzFYsubEtLXfj7TQDoJZkY7MD2LVINiS2SfMkSCRuOMxjIPlw2+HcjbMTQFoOJG7DpaX/YftvkJb9h3mMfwK1pBvOTv+AVwu/RGIbM9gWZh4zyY3bDifIS+fgt4Vf4mGzNE+aROKMw2xlljP/pRlukM4pOJBggNsvbPzJBz/z2NQl9vcf3nyz54yNvPzs9M0fPlTYyeHSggkMwCoNiFUOAvINpKgeBaNgFIyCkQAAPDti6z3DYGwAAAAASUVORK5CYII=","orcid":"","institution":"Deggendorf Institute of Technology, European Campus Rottal-Inn (ECRI)","correspondingAuthor":true,"prefix":"","firstName":"Valentine","middleName":"","lastName":"Idakwo","suffix":""}],"badges":[],"createdAt":"2024-05-05 07:24:00","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4370656/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4370656/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":56678154,"identity":"e0837a74-9dd2-41e9-ab91-18ac54abbb21","added_by":"auto","created_at":"2024-05-17 16:38:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":741551,"visible":true,"origin":"","legend":"\u003cp\u003eJob Distribution of Survey Participants\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-4370656/v1/10ba7a4c0c11980fe7a64184.png"},{"id":56679506,"identity":"a4e9ff67-a525-4556-abe7-ecce3f905078","added_by":"auto","created_at":"2024-05-17 16:46:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":571050,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of Participant Views on Data Access by AI Use Frequency\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-4370656/v1/c3aec96217bb14ebecb90688.png"},{"id":56678155,"identity":"e594021a-5963-4be3-b9a3-c78b5828a43d","added_by":"auto","created_at":"2024-05-17 16:38:04","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":593151,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of Organizational AI Readiness by Location\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-4370656/v1/739eca2720541d2d06a948d0.png"},{"id":56678157,"identity":"ec86d1cf-b5dd-4b28-93c7-a0afd89b141f","added_by":"auto","created_at":"2024-05-17 16:38:05","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":729110,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of AI Knowledge Levels among Survey Participants\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-4370656/v1/45cb83282e751326c82d0d03.png"},{"id":56678158,"identity":"fec481e1-7457-4f96-9f69-3130b4ed2129","added_by":"auto","created_at":"2024-05-17 16:38:05","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":970588,"visible":true,"origin":"","legend":"\u003cp\u003eSurvey Participants’ Views on AI\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-4370656/v1/7bdea4528acb5dc0761136dc.png"},{"id":56678159,"identity":"4a77a475-67d5-43d2-a993-053ef5a8d49c","added_by":"auto","created_at":"2024-05-17 16:38:05","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":3850488,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of Findings on Challenges of AI in Cardiovascular Care\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-4370656/v1/cbae60cbcab2c28e8c2b8393.png"},{"id":56679508,"identity":"7fee3479-2670-4ae1-aac4-906d4e6e01b6","added_by":"auto","created_at":"2024-05-17 16:46:21","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3469324,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4370656/v1/1ab501d5-30ba-4b63-89cc-1a5bd97fb394.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"From Diagnosis to Management: Unveiling the Challenges of Artificial Intelligence Solutions in Cardiovascular Healthcare","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCardiovascular diseases (CVDs) are the leading cause of mortality globally, accounting for nearly 19\u0026nbsp;million deaths in 2019 (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). This number is projected to increase due to the ageing demographics and the lack of healthcare workers (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Artificial Intelligence (AI) offers a path in helping clinicians and healthcare institutions manage the burden of CVDs through risk stratification, early diagnosis, clinical decision support, patient monitoring and personalized medicine. Some studies have shown promising results in several applications of AI in cardiovascular care (\u003cspan additionalcitationids=\"CR4 CR5 CR6\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Examples of these studies include the use of a machine learning (ML) model to predict the occurrence of in-hospital cardiac arrests using the variability in heart rates and the use of an ML model for the risk stratification of occlusion myocardial infarction (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite these promising results, the adoption of AI in clinical practice is significantly low due to the novelty of the field and the existence of several challenges hindering its adoption (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Although cardiovascular diseases (CVDs) have a significant impact on healthcare, there is a scarcity of experimental studies investigating the challenges associated with adopting AI in clinical cardiovascular care. A mixed-methods research, by Schepart et al. (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e), identified the existence of 5 major challenges of AI in cardiovascular care: insufficient knowledge, limited useability, limited funding, inadequate compatibility with electronic health records, and trust issues. It is important to note that the scope of Schepart et al.'s study was confined to cardiologists and health IT specialists, which may not reflect the perspectives of the broader healthcare community.\u003c/p\u003e \u003cp\u003eThe goal of this study is to better understand the challenges hindering the adoption of AI in cardiovascular care through a mixed-methods study of healthcare workers and stakeholders involved in cardiovascular care. A larger group including doctors, nurses, medical researchers, hospital administrators, health IT specialists, cardiovascular assistants, and other stakeholders in cardiovascular care was adopted to offer a more comprehensive view of the challenges faced. Involving a diverse cohort of healthcare professionals provides a multifaceted understanding of the challenges, ensuring that solutions are robust and widely applicable. By understanding these challenges, targeted interventions could be developed to enhance the adoption of AI, ultimately improving patient outcomes in cardiovascular care.\u003c/p\u003e"},{"header":"Methodology","content":"\u003cp\u003e\u003cem\u003eStudy Design\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn this study, a sequential mixed methods research design, with initial semi-structured interviews of several stakeholders in cardiovascular care followed by a self-administered online survey of a wider group, was used. \u0026nbsp;Results from the interviews were used to develop the quantitative survey. This approach allowed the capturing of in-depth perspectives from key stakeholders in cardiovascular care and the corroboration of these findings with quantitative evidence from a wider community of healthcare professionals.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQualitative Study\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eParticipant selection and recruitment\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMedical Doctors, nurses, health IT specialists, clinical scientists, medical researchers, hospital administrators, and cardiovascular AI specialists were eligible to participate in the interviews. The primary inclusion factors were:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eKnowledge of AI in cardiovascular care\u003c/li\u003e\n\u003cli\u003eOver 3 years of experience in cardiovascular care or AI solutions in cardiovascular care.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eParticipant recruitment was conducted using purposive sampling. This approach was chosen to ensure that the participants could provide in-depth and informed perspectives directly relevant to the research objectives. Eligible participants were recruited and identified by the researchers and invited to participate in the research via email and LinkedIn\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData collection\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eA semi-structured interview guide was developed following an extensive literature review of the challenges of AI in medical care. The interview guide was tested on a demographically similar subset of the target population. Feedback from this group led to refinements in question wording and order, optimizing clarity and response efficacy. All interviews lasted 30-40 minutes and were conducted online, via video conferencing, with one interviewer in November 2023. The interviewer asked follow-up questions for clarification and to allow participants to elaborate on the responses provided. The interviews were recorded, transcribed verbatim, and inductively analyzed to identify major themes and additional sub-themes.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe transcribed text was subjected to inductive thematic analysis. This entailed a systematic process of coding the data in multiple rounds, allowing themes to emerge directly from the participants' narratives without the constraint of pre-existing theoretical frameworks. Initial codes were generated by reading and re-reading the transcripts, which were then grouped into potential themes reflecting the core meanings evident in the data. These themes were reviewed and refined iteratively, ensuring they accurately represented the views of participants.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQuantitative Study\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eParticipant selection and recruitment\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMedical Doctors, nurses, health IT specialists, clinical scientists, medical researchers, hospital administrators, and cardiovascular AI specialists were eligible to participate in the surveys. A mix of purposive sampling and snowball sampling was used to ensure that participants met the eligible professional criteria and to reach a larger portion of the target population. Participants were invited to participate in the survey via email and LinkedIn.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData collection\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eA 10-minute online survey was created using Soscisurvey (Version 3.5.00) developed by SoSci Survey GmbH based on the key themes identified in the qualitative analysis. The questionnaire was developed in English and pretested with 7 individuals who were demographically similar to the target population. Feedback from the pretest indicated that certain questions in the survey were ambiguous and lacked clarity. Consequently, the highlighted questions were refined to be more precise and understandable. A screener question was employed to identify and filter out inattentive responses from participants. The survey link was shared with the participants and was online for 6 weeks from November to December 2023. The survey results were subsequently downloaded in Comma-Separated Values (CSV) format and subsequently analyzed.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe data was initially cleaned to remove incomplete responses and responses that incorrectly answered the screener. As responses were numerically coded automatically be the survey platform, additional coding was not necessary. Descriptive statistics were employed to gain insights into the characteristics of respondents and for univariate analysis. The non-parametric tests, Kruskal-Wallis and Mann-Whitney with Bonferroni correction, were used for pairwise comparisons. The analysis was done using Python libraries: Pandas (version 1.4.2), SciPy (version 1.8.1), and NumPy (version 1.23.3) due to their versatility, efficiency, and extensive capabilities for handling and analyzing large datasets. Statistical significance was set at \u003cem\u003ep-\u003c/em\u003evalue \u003cem\u003e\u0026le;\u003c/em\u003e 0.05.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQualitative Results\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDescription of participants\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFive experts representing diverse sectors within the cardiovascular care domain were interviewed: PT1, a medical doctor and AI researcher; PT2, the CEO of an AI-focused CVD company; PT3, a clinical scientist specializing in CVD care; PT4, a digital health hospital consultant; and PT5, an interventional cardiologist with experience in AI-assisted cardiovascular procedures.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIdentified Challenges\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eEight major challenges were identified by participants during the interviews. The findings are summarized in Table 1 and described below.\u003c/p\u003e\n\u003col\u003e\n\u003cli\u003e\u003cem\u003eData-Related Challenges\u003c/em\u003e: Participants identified several issues related to data that existed and needed to be addressed. According to them, data is vital for training AI algorithms in cardiovascular care and issues with data affect the quality of the algorithms developed. Challenges related to data integration and data access were highlighted by participants. Data for training and validating AI models is scarce. However, this data scarcity is artificial as participants acknowledged that data exist in different cardiovascular institutions. Issues like data fragmentation, data silos, differences in regulation, and variations in annotation make accessing this data difficult. Participants also highlighted that cardiovascular care requires multimodal data for the diagnosis of CVDs. However, existing AI solutions do not allow for the integration of multimodal data in the course of treatment, posing a challenge to their use in cardiovascular care.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eRegulatory Challenges:\u003c/em\u003e Participants also highlighted unclear regulations and the inflexibility of existing regulations as the primary regulatory challenges to AI in cardiovascular care. Existing AI regulations are vague and do not have specific medical or cardiovascular applications. This lack of clarity results in a long and arduous process for obtaining regulatory approval for AI solutions in cardiovascular care. Additionally, participants indicated that existing regulations require developers to freeze their algorithms to obtain regulatory approval and developers have to apply for a new approval once changes are made to the algorithms. Participants believed that this poses a challenge as AI algorithms require constant training with real-world data to be safer and more efficient. Requiring regulatory approval each time an algorithm is trained results in an overly burdensome process.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eInfrastructural Challenges: \u003c/em\u003eParticipants also acknowledged the dearth of human and technological infrastructure for the integration of AI in cardiovascular care. Having individuals with the right knowledge to develop and maintain AI infrastructure is key to AI use in clinical practice. However, participants indicated that most healthcare institutions lack the necessary IT department, equipped with an understanding of how AI systems work. In addition, healthcare facilities largely use legacy systems that are incompatible with the latest AI technology and participants believe that this hampers the use of AI in cardiovascular care. Participants also highlighted the existence of a rural-urban divide regarding infrastructure. While healthcare institutions in large urban areas are actively researching and investing in improving their AI infrastructure, healthcare institutions in rural areas are largely not focused on AI infrastructure.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eKnowledge Challenges:\u003c/em\u003e Participants highlighted a bidirectional knowledge gap between healthcare professionals and developers. They noted that healthcare professionals lack sufficient understanding of AI to effectively communicate its mechanisms to patients, while developers often lack the medical expertise necessary for creating clinically relevant applications. Participants also acknowledged the lack of AI training in medical and medically-affiliated curricula as one of the primary reasons for this gap. However, they also acknowledged the existence of advanced professional courses for individuals willing to improve their knowledge of AI.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eTransparency Challenges: \u003c/em\u003eA lack of explainability of existing AI solutions was identified as a primary challenge of AI in cardiovascular care by participants. They indicated that the transparency of AI models is key for legal and liability protection and existing AI models in cardiovascular care do not offer sufficient explanation of their decision-making process and functionality. Explainability is also key for regulatory compliance and entry into healthcare. Hence, the existence of black-box models does not spark confidence among clinicians and regulators, which results in slower adoption in clinical practice.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eEthical Challenges:\u003c/em\u003e Participants highlighted fairness in data collection, lack of accountability, and vagueness of responsibility as the primary ethical challenges faced by AI in cardiovascular care. They indicated that training datasets used in the development of AI solutions are not representative of the target populations. Hence biases are replicated in the results of these models. They also believe that questions on responsibility pose a challenge to the integration of AI in clinical care. While traditional medical care places responsibility primarily on clinicians when negative outcomes occur, this is not clearly outlined when an AI algorithm is primarily responsible for the decisions made. Hence, reluctance exists amongst clinicians on the use of AI in cardiovascular care. Participants also believe that AI developers are not transparent enough in reporting the developmental process and the steps leading to the creation of the final product. They believe that this lack of accountability diminishes the trust of clinicians and their willingness to adopt AI solutions in cardiovascular care.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eChange Management Challenges\u003c/em\u003e: Participants acknowledged the lack of quality change management plans and the lack of medical and economic impact analysis for AI solutions. They believe that the integration of AI in cardiovascular care is significantly slowed as healthcare institutions lack a comprehensive and coherent plan on how to integrate AI in cardiovascular care. One of the key aspects of change management plans is the impact analysis. Participants indicated the lack of in-depth impact analysis that offers genuine insights into the effect of AI solutions on clinical care and administration. This lack of coherent change management plans creates hurdles for medical institutions and governments in the integration of AI solutions in cardiovascular care.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eAcceptance Challenges\u003c/em\u003e: Participants acknowledged that acceptance by healthcare professionals and patients was a key factor in the integration of AI in cardiovascular care. They believed that some skepticism exists among healthcare professionals on the use of AI in cardiovascular care due to the reluctance to change established ways of working, concerns about job security, and lack of trust in AI solutions. It is important to note that participants do not acknowledge this as being the majority opinion but they acknowledged that the minority opinion still impacts the integration of AI in cardiovascular care.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cstrong\u003e[Table 1]\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQuantitative Results\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDescription of participants\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOf the 134 individuals who initiated the survey, 94 (70.1%) completed the survey in its entirety. However, three participants (3.2% of completed surveys) provided incorrect responses to the screener and were subsequently excluded from the analysis. The decision to exclude three participants who provided incorrect responses to the screener was made to maintain data integrity and ensure the reliability of the study findings. Thus, a total of 91 valid responses were obtained for analysis. Although the majority of participants (n= 55, 60.4%) were doctors, the responses also included nurses, medical researchers, health IT specialists, hospital administrators, medical assistants, and cardiovascular technologists. The distribution of participants across different job descriptions is visualized in Figure 1. The majority of participants were from Europe (n = 56, 61.5% of valid responses), followed by Africa (n = 24, 26.4%), Asia (n = 7, 7.7%), and North America (n = 6, 6.6%). Notably, no responses were received from Australia or South America.\u003c/p\u003e\n\u003cp\u003eParticipants were also asked about their frequency of using AI for work. The responses indicated that 42.6% (n = 38) reported never using AI, 23.6% (n=21) reported monthly usage, 16.9% (n=15) reported weekly usage and another 16.9% (n=15) reported daily usage.\u003c/p\u003e\n\u003col\u003e\n\u003cli\u003e\u003cem\u003eData-related Challenges\u003c/em\u003e: In terms of data access, the majority of respondents (51.5%) reported encountering difficulty (45.3% found it difficult, while 6.2% found it very difficult) in accessing cardiovascular health data for AI analysis and interpretation. 1.6% found data access to be very easy and 12.5% found it to be easy. 34.4% of respondents maintained a neutral position on data access. A pairwise comparison with Kruskal Wallis test indicated the presence of a statistically significant correlation [H(4) = 9.10, \u003cem\u003ep\u003c/em\u003e = 0.028] between the frequency of use and the opinion regarding data access. Further exploration with \u003cem\u003epost-hoc\u003c/em\u003e Mann-Whitney test with Bonferroni correction indicated the presence of a statistical difference (\u003cem\u003ep=\u003c/em\u003e038) in data access views between daily AI users and monthly AI users. Figure 2 shows a box plot visualization in which daily AI users consistently rated data access as more challenging, compared to the more varied responses of monthly AI users. On the compatibility of different data sources for CVD analysis, 36.5% indicated that data sources for CVD analysis were incompatible (4.8% found them highly incompatible and 31.7% found them incompatible). 36.5% maintained a neutral position while 23.8% and 3.2% found data sources for CVD care to be compatible and highly compatible respectively.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eRegulatory Challenges\u003c/em\u003e: To measure the clarity of existing regulations for AI solutions in cardiovascular care, participants were asked about the transparency and comprehensibility of existing regulations. The majority of respondents (55%) indicated that existing regulations were not transparent or comprehensible (41.7% expressed disagreement and 13.3% expressed strong disagreement). 35% maintained a neutral position and 10% agreed that existing regulations were transparent and comprehensible. Notably, no participant expressed strong agreement. Participants were also asked about the adequacy of the current regulatory framework in facilitating the safe and efficient implementation of AI solutions in cardiovascular care. 40.7% and 5.1% expressed disagreement and strong disagreement with the notion that existing AI regulations were adequate. 37.3% remained neutral and 16.9% expressed agreement. Notably, there were no participants who strongly agreed.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eInfrastructural challenges\u003c/em\u003e: Regarding organizational readiness for dealing with the introduction of AI in cardiovascular care, 24.7% of respondents indicated that their organizations were not at all equipped. 39.3% and 30.2% of respondents considered their organizations to be slightly equipped and moderately equipped respectively. Only 5.6% indicated that their organizations were very equipped. Notably, none of the participants regarded their organizations as extremely equipped. The Kruskal-Wallis test showed a significant difference [H(3)= 7.893, \u003cem\u003ep\u003c/em\u003e= 0.048] in organizational readiness based on geographical location. \u003cem\u003ePost-hoc\u003c/em\u003e analysis using Mann-Whitney test with Bonferroni correction indicated a marginal difference (\u003cem\u003ep\u003c/em\u003e= 0.063) between the respondents in Europe and Africa. The boxplot visualization in Figure 3 shows that respondents from Africa consistently indicated lower levels of organizational readiness compared to the varied responses of respondents in Europe.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eKnowledge Challenges:\u003c/em\u003e The question asking participants to rate their knowledge of AI showed less than optimal levels in the majority (68.2%) of participants ( 3.3% had very poor knowledge, 18.7% had below-average knowledge, and 46.2% had average knowledge). The frequency and distribution of participants\u0026rsquo; knowledge levels are shown in Figure 4. Additionally, participants were asked if they would be willing to take courses to improve their knowledge of AI. 78% of respondents were willing to take courses to improve their knowledge of AI, 7.7% were not willing to take additional courses and 14.3% were unsure.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eTransparency Challenges:\u003c/em\u003e On the importance of transparency, 82.4% of respondents indicated that AI systems needed to offer some form of explainability in their decision-making process (50.5% expressed agreement and 31.9% expressed strong agreement). 12.1% maintained a neutral position while 4.4% and 1.1% expressed disagreement and strong disagreement respectively. Additionally, on the impact of transparency on user trust, the majority of respondents (62%) indicated that they needed to understand the clinical decision-making process of AI systems to trust their recommendations (36.3% agreed and 26.4% strongly agreed). Conversely, 20.9% of respondents did not perceive understanding the decision-making process of AI systems as impactful on their trust in their recommendations. Specifically, 15.4% disagreed and 5.5% agreed. 16.5% of respondents were neutral.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eEthical Challenges:\u003c/em\u003e On fairness, participants were asked the importance of AI solutions considering patient diversity in cardiovascular care. The majority of respondents (52.2%) considered it extremely important for AI solutions to consider patient diversity. 18.9% considered it to be important and 18.9% maintained a neutral position. 7% and 3.3% of respondents considered it to be slightly unimportant and not important at all respectively. Subsequently, participants were asked if existing AI solutions adequately addressed the diversity of patient population. 10.2% of participants responded in the affirmative, 38.6% maintained a neutral position and 51.2% responded in the negative.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eChange Management Challenges:\u003c/em\u003e3% and 34.2% of respondents considered having an organizational plan for the integration of AI in cardiovascular care to be important and extremely important respectively. 19.7% of respondents took a neutral position. 10.5% and 5.3% indicated that having a plan was slightly unimportant and not important at all respectively. When asked about the quality of their organizational plans for the integration of AI in cardiovascular care, none of the respondents indicated that their organizational plans were fully optimized and comprehensive. 6.6% and 19.7% indicated that their organizational plans were well-developed and moderately developed respectively. 34.2% indicated that their organizational plans were poorly developed while 39.5% indicated that their organizational plans were non-existent.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eAcceptance Challenges\u003c/em\u003e: Participants were asked several questions to gauge their perspective of AI. Figure 5 contains a comprehensive overview of the distribution of responses.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis mixed-methods study was conducted to identify the gaps in AI solutions for the diagnosis and management of CVDs. The interviews and surveys included diverse health professionals involved in the care of cardiovascular patients. This diverse group was selected to offer a comprehensive view of the challenges. To my knowledge, this is the first study to assess such a diverse group of health professionals in cardiovascular care. In the interviews, several gaps were identified that aided the development of the ensuing quantitative research. Specifically, participants identified data-related, regulatory, infrastructural, knowledge, transparency, ethical, change management, and acceptance challenges. Identifying these gaps underscores the complexity of integrating AI solutions into the diagnosis and management of CVDs.\u003c/p\u003e \u003cp\u003eOn data, participants highlighted the importance of vast amounts of data in the introduction of AI in cardiovascular. However, challenges related to accessing this data and integrating data from different sources persist. The position on data access was corroborated by the subsequent survey of the larger population as a majority of participants indicated the existence of some difficulty in data access. The survey also showed that increasing the use of AI exposes users to more of the challenges involved in accessing data needed for AI development. On the compatibility of different data sources, the responses of participants were more varied. A substantial portion of respondents indicated the presence of compatibility issues and a little under a third of respondents found data sources to be compatible, indicating a diversity of experiences on this issue. Additionally, a significant portion of respondents chose to maintain a neutral stance, indicating a sizable segment of the surveyed population withholding strong opinions or possibly lacking sufficient knowledge or experience to form definitive judgments on data compatibility.\u003c/p\u003e \u003cp\u003eUnclear regulations and static regulations were identified by interviewees as the primary regulatory challenges being faced by AI in cardiovascular care. These regulatory challenges hinder the seamless integration of AI in cardiovascular care. The lack of clarity of regulations was corroborated by the survey as the majority of respondents believed that existing regulations were not transparent and comprehensible. Over a third of respondents maintained a neutral position indicating the absence of strong opinions or the lack of sufficient knowledge on the topic. Additionally, a significant portion of respondents also indicated that existing regulations are not adequate to facilitate the safe and efficient implementation of AI solutions in cardiovascular care. Further research is required to understand the specific reasons behind this view on adequacy.\u003c/p\u003e \u003cp\u003eThe successful introduction of new technology in cardiovascular care is dependent on the existence of the right infrastructure to aid in its integration. Interviewees indicated that existing healthcare institutions lacked the necessary human and technological infrastructure for the successful integration of AI in cardiovascular care. The existence of a rural-urban divide regarding infrastructure was also highlighted in the interview and would require additional research to understand the reasons behind this divide. The position on the lack of infrastructure was also highlighted in the survey as over 90% of respondents indicated that their organizations were either not equipped at all or insufficiently equipped. The survey also showed that participants in Africa consistently reported lower levels of organizational readiness than their counterparts in Europe, indicating that different geographical locations would need to adopt different methods to address this challenge.\u003c/p\u003e \u003cp\u003eInterviewees also indicated a prevalent knowledge gap between healthcare professionals and developers. They indicated that medically-affiliated curricula lack relevant AI-related training. Hence, medical professionals lack the necessary understanding of AI to effectively work with it or communicate its mechanisms to patients. Exploring the reasons behind the absence of AI-related training in medical curricula, such as competing educational priorities or limited faculty expertise in AI, would provide valuable insights into the systemic challenges contributing to the knowledge gap. The interview findings were similar to the findings of the survey, in which most participants reported average or less than average knowledge of AI. However, the survey also revealed that the majority of respondents were also willing to take additional courses to improve their knowledge of AI.\u003c/p\u003e \u003cp\u003eInterviewers highlighted the existence of transparency gaps with the prevalence of unexplainable models impacting the level of confidence among clinicians and regulators. Although survey respondents overwhelmingly agreed on the importance of explainability, there was a little less consensus on whether the explainability of AI impacts their level of trust in the recommendations offered by these solutions. Furthermore, while the survey results do not explicitly confirm the existence of transparency challenges, they reveal that consensus has not been achieved on the impact of transparency or the level of transparency required for fostering trust in AI recommendations. Further elaboration on the lack of consensus regarding the impact of explainability on trust in AI recommendations could include exploring factors contributing to this discrepancy, such as varying levels of familiarity with AI technology among healthcare professionals or differing perspectives on the importance of interpretability versus predictive accuracy.\u003c/p\u003e \u003cp\u003eFairness in data collection, lack of accountability and lack of clarity regarding responsibility were highlighted in the interviews as the primary ethical challenges. All three issues impact the level of trust that healthcare professionals have in AI solutions. Bias in data collection also results in faulty models as these biases are replicated in the model output. The survey results show that healthcare professionals understand the importance of fairness for AI solutions. However, the majority of respondents also indicated that existing solutions do not adequately address fairness in the patient population. Future research might explore potential strategies for enhancing accountability and clarity regarding responsibility in AI-driven healthcare systems, as well as developing methodologies to mitigate bias in data collection and ensure fairness in patient population representation.\u003c/p\u003e \u003cp\u003eInterviewees also indicated the lack of quality change management plans and the lack of medical and economic impact analysis as key change management challenges for AI in cardiovascular care. The lack of a comprehensive change management plan for healthcare institutions results in a chaotic adoption process and diminishes the enthusiasm for adoption. Additionally, the lack of impact analysis diminishes the trust of stakeholders in introducing AI solutions into their established workflow. The lack of plans was also highlighted in the surveys as the majority of respondents indicated that their organizational plans for AI introduction were either non-existent or insufficiently developed despite overwhelmingly agreeing on the importance of having an organizational plan. Future research needs to explore the unique requirements and challenges of AI change management plans and impact analysis in healthcare settings, particularly in comparison to the protocols and considerations involved in the introduction of new medications.\u003c/p\u003e \u003cp\u003eInterviewees also indicated the existence of some skepticism amongst end users for AI solutions in cardiovascular care. They highlighted that although this skepticism was not the majority position, it was impactful as overwhelming acceptance is required for the introduction of AI in cardiovascular care. This position was also confirmed in the in survey with the majority of respondents holding positive views on the impact of AI on the diagnosis and treatment of cardiovascular diseases, treatment errors, worker shortage, and patient benefits. On the trustworthiness of AI solutions, most participants maintained a neutral position underscoring the lack of positive or negative leanings on the trustworthiness of AI. Despite most participants indicating that AI should be used in the diagnosis and treatment of CVDs, about one-sixth of respondents indicated that AI should not be used in cardiovascular care, highlighting a divergence of opinions within the surveyed population regarding the appropriateness or efficacy of AI solutions in this context. The summary of findings is contained in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe findings of this study are similar to other studies exploring the challenges of AI in cardiovascular care and AI in general. The survey by Schepart et al.(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e) also indicated the existence of minority skepticism toward AI from cardiologists and health IT professionals. Additionally, the 80% of respondents who highlighted the importance of transparency in trusting AI solutions from that study indicate a larger consensus on transparency than that expressed in this study. The lack of knowledge in AI by medical professionals was also highlighted in several studies of AI in cardiovascular care and medicine as a whole indicating that these problems are not confined to cardiovascular care (\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Our study extends the findings of challenges of AI in cardiovascular care by sampling a more diverse group of healthcare professionals involved in cardiovascular care.\u003c/p\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eThe sample size and sample distribution are limitations to this study. The relatively small sample size for both the interviews and the survey limits the generalizability of the conclusions generated. Despite this limitation, the focused nature of the sample allows for in-depth exploration of perspectives and insights from a diverse range of healthcare professionals involved in cardiovascular care, providing rich qualitative data and nuanced perspectives on the challenges and opportunities associated with the integration of AI solutions in this context. Additionally, our survey sample composition presents a skew, with the majority of respondents being doctors. This overrepresentation of one group might introduce bias and limit the perspectives of other healthcare professionals. While the over-representation of doctors might introduce bias, it positions the study to capture in-depth insights from some of the primary users of cardiovascular AI technology. The geographical skew of participants toward Europe potentially limits the global applicability of the research conclusions. These limitations highlight the need for a cautious interpretation of the study\u0026rsquo;s results and suggest the opportunity for extensive research to expand upon the findings.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, this mixed-methods study provides valuable insights into the challenges of AI solutions in cardiovascular care. Through interviews and surveys with diverse healthcare professionals, key data-related, regulatory, transparency, acceptance, change management, knowledge, and infrastructural gaps were identified. These findings emphasize the complexity and multifaceted nature of implementing AI in cardiovascular settings. Future research should focus on developing tailored approaches to address the identified challenges.\u003c/p\u003e \u003cp\u003eOverall, this study contributes to the growing body of literature on AI in healthcare by providing nuanced insights into the complexities of integrating AI solutions in cardiovascular care. By addressing these challenges and leveraging the opportunities presented by AI, we can strive towards more effective and personalized approaches to the diagnosis, management, and treatment of cardiovascular diseases, ultimately improving patient outcomes and advancing the field of cardiovascular medicine.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eSupplementary information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe supplementary documents for this article can be found online at:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eI thank all the experts who took part in the data collection and research development for this study. I also specifically thank Mrs. Stefanie Werkman, Professor Dr. Stefan K\u0026auml;\u0026auml;b, Professor Dr. Solveig Vieluf, and Dr. Christian G\u0026ouml;lz for their role in pre-testing and data collection. Additionally, I would like to thank Professor Dr. Georgi Chaltikyan and Professor Mouzhi Ge for their support on this research.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eVI: conceptualization, methodology, interview and survey design, data processing, visualization, analysis, and writing the original draft.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and participants\u0026rsquo; consent\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eInformed consent was obtained from all participants involved in the study. The research adhered to relevant guidelines and regulations, including those governing research practices at Deggendorf Institute of Technology. As the study did not involve interventions falling under German law\u0026apos;s requirement for additional ethics approval, such approval was not sought.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Declaration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author declares that no external funding was received for the research, authorship, and/or publication of this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAuthor have no conflict of interest relevant to the context of this article to declare\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eMariachiara Di Cesare, Honor Bixby, Thomas Gaziano, Lisa Hadeed, Chodziwadziwa Kabudula, Diana Vaca McGhie, et al. World Heart Report 2023: Confronting the World\u0026rsquo;s Number One Killer. World Heart Federation. Geneva, Switzerland: World Heart Federation; 2023.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoth GA, Mensah GA, Johnson CO, Addolorato G, Ammirati E, Baddour LM, et al. Global Burden of Cardiovascular Diseases and Risk Factors, 1990\u0026ndash;2019: Update From the GBD 2019 Study. Vol. 76, Journal of the American College of Cardiology. 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLanglais \u0026Eacute;L, Th\u0026eacute;riault-Lauzier P, Marquis-Gravel G, Kulbay M, So DY, Tanguay JF, et al. Novel Artificial Intelligence Applications in Cardiology: Current Landscape, Limitations, and the Road to Real-World Applications. J Cardiovasc Transl Res. 2023;16(3):513\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrittanawong C, Johnson KW, Rosenson RS, Wang Z, Aydar M, Baber U, et al. Deep learning for cardiovascular medicine: A practical primer. Vol. 40, European Heart Journal. 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaratzia L, Aung N, Aksentijevic D. Artificial intelligence in cardiology: Hope for the future and power for the present. Vol. 9, Frontiers in Cardiovascular Medicine. 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeeny AK, Chung MK, Madabhushi A, Attia ZI, Cikes M, Firouznia M, et al. Artificial Intelligence and Machine Learning in Arrhythmias and Cardiac Electrophysiology. Vol. 13, Circulation: Arrhythmia and Electrophysiology. 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSiontis KC, Noseworthy PA, Attia ZI, Friedman PA. Artificial intelligence-enhanced electrocardiography in cardiovascular disease management. Nat Rev Cardiol. 2021;18(7):465\u0026ndash;78.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl-Zaiti SS, Martin-Gill C, Z\u0026egrave;gre-Hemsey JK, Bouzid Z, Faramand Z, Alrawashdeh MO, et al. Machine learning for ECG diagnosis and risk stratification of occlusion myocardial infarction. Nat Med. 2023;29(7):1804\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee H, Yang HL, Ryu HG, Jung CW, Cho YJ, Yoon S Bin, et al. Real-time machine learning model to predict in-hospital cardiac arrest using heart rate variability in ICU. NPJ Digit Med. 2023;6(1):215.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchepart A, Burton A, Durkin L, Fuller A, Charap E, Bhambri R, et al. Artificial intelligence\u0026ndash;enabled tools in cardiovascular medicine: A survey of current use, perceptions, and challenges. Cardiovasc Digit Health J. 2023;4(3).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScheetz J, Rothschild P, McGuinness M, Hadoux X, Soyer HP, Janda M, et al. A survey of clinicians on the use of artificial intelligence in ophthalmology, dermatology, radiology and radiation oncology. Sci Rep. 2021;11(1):5193.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCastagno S, Khalifa M. Perceptions of Artificial Intelligence Among Healthcare Staff: A Qualitative Survey Study. Front Artif Intell. 2020;3.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable\u0026nbsp;1: Challenges of AI in Cardiovascular Care\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"25.124792013311147%\" valign=\"top\"\u003e\n \u003cp\u003eMain Theme\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.96339434276206%\" valign=\"top\"\u003e\n \u003cp\u003eSub-Theme\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.91181364392679%\" valign=\"top\"\u003e\n \u003cp\u003eQuotes from Interview Participants\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"25.124792013311147%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eData\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.96339434276206%\" valign=\"top\"\u003e\n \u003cp\u003eData access\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.91181364392679%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;The first one is access. Training an AI model requires a lot of data and it is very difficult in many cases to acquire a lot of data\u0026rdquo;- PT1\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Data is key for training AI algorithms and having sufficient data for training is one of the key challenges\u0026rdquo; - PT2\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"29.333333333333332%\" valign=\"top\"\u003e\n \u003cp\u003eData integration\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"70.66666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Another challenge \u0026nbsp;lies in connecting data points, so not just integrating EHR data with ECG data, imaging data, vital signs, and voice analysis. So not just looking at one data source but interpreting everything in the context of another\u0026rdquo;- PT1\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"25.124792013311147%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eRegulations\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.96339434276206%\" valign=\"top\"\u003e\n \u003cp\u003eUnclear regulations\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.91181364392679%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Regulatory approval for medical AI is more complex. The process is often long and requires evidence that the AI\u0026rsquo;s recommendations are at least as good as, if not better than that of a human expert.\u0026rdquo;- PT5\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"29.333333333333332%\" valign=\"top\"\u003e\n \u003cp\u003eStatic regulations\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"70.66666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Another problem is that regulatory bodies ask AI companies to freeze their algorithms for approval. I think this is a bit of a hurdle and also an inconsistency because they want safer and better products but then they hinder them because they don\u0026rsquo;t know how to or because they require freezing.\u0026rdquo; - PT1\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"25.124792013311147%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eInfrastructure\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.96339434276206%\" valign=\"top\"\u003e\n \u003cp\u003eLack of infrastructure\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.91181364392679%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;The infrastructure, in terms of know-how and competency of staff, the technology, the service, and performance, is largely missing for AI in cardiovascular care.\u0026rdquo; - PT4\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Many healthcare facilities use legacy systems that aren\u0026rsquo;t readily compatible with the latest\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eAI technology\u0026rdquo;- PT5\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"29.333333333333332%\" valign=\"top\"\u003e\n \u003cp\u003eRural-urban divide\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"70.66666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;At this stage, AI progress is concentrated on large university hospitals or well-equipped hospitals, and on the startup market in urban areas, but not in the outskirts or rural areas.\u0026rdquo; -PT5\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"25.124792013311147%\" valign=\"top\"\u003e\n \u003cp\u003eMain Theme\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.62728785357737%\" valign=\"top\"\u003e\n \u003cp\u003eSub-Theme\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"51.24792013311148%\" valign=\"top\"\u003e\n \u003cp\u003eQuotes from Interview Participants\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"25.124792013311147%\" valign=\"top\"\u003e\n \u003cp\u003eKnowledge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.62728785357737%\" valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"51.24792013311148%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;When I spoke to some healthcare workers, I noticed they didn\u0026rsquo;t know anything. They hadn\u0026rsquo;t even tried ChatGPT and had no hands-on experience with AI. This is one side of the spectrum\u0026rdquo;- PT4\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;I see that there is some knowledge available on what it means. The basic terms are quite common these days, like on convolutional neural networks, but most people have limited understanding of what it means.\u0026rdquo;- PT2\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;We generally see a lack of education and training on how to work with AI, or interpret its recommendations.\u0026rdquo;- PT5\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"25.124792013311147%\" valign=\"top\"\u003e\n \u003cp\u003eTransparency\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.62728785357737%\" valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"51.24792013311148%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;One of the major challenges is explainable AI and understanding machine reasoning and also in terms of legal issues and liability. This is still one of the biggest hurdles that must be defined and needs to be understood.\u0026rdquo;- PT2\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;There is a black-box phenomenon for different AI solutions in healthcare. This black-box phenomenon- you can\u0026rsquo;t have it in medical treatment.\u0026rdquo;- PT4\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"25.124792013311147%\" rowspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003eEthics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.62728785357737%\" valign=\"top\"\u003e\n \u003cp\u003eFairness\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"51.24792013311148%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Of-course there is the issue of biases in terms of patient recruitment for training data. I think there are some good practices that need to be followed in order to ensure that the training datasets are representative of the population that is being treated or diagnosed. So, I think that\u0026rsquo;s an aspect that needs to be accounted for\u0026rdquo;- PT1.\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.555555555555557%\" valign=\"top\"\u003e\n \u003cp\u003eResponsibility\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.44444444444444%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;There is the question of accountability when AI is used in diagnosis and treatment- Who is responsible if something goes wrong?\u0026rdquo;- PT5.\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.555555555555557%\" valign=\"top\"\u003e\n \u003cp\u003eAccountability\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.44444444444444%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026rdquo; Even though we have dedicated guidelines indicating and structuring how data treatment should be performed, some care providers do not follow these recommendations or report the model development process\u0026rdquo;- PT2.\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"26.622296173044926%\" valign=\"top\"\u003e\n \u003cp\u003eMain Theme\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.62728785357737%\" valign=\"top\"\u003e\n \u003cp\u003eSub-Theme\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"49.750415973377706%\" valign=\"top\"\u003e\n \u003cp\u003eQuotes from Interview Participants\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"26.622296173044926%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eChange Management\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.62728785357737%\" valign=\"top\"\u003e\n \u003cp\u003eImpact analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"49.750415973377706%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Medical and economic impact is the core of digital health and also AI models. We have to quantify what we bring from a medical perspective and an economic perspective\u0026rdquo;- PT3\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Regarding the impact measurement, I rarely see complete views. I call this the net present value. It\u0026rsquo;s basically a comprehensive view on what the algorithm does and what the whole infrastructure requires\u0026rdquo;- PT3\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"32.19954648526077%\" valign=\"top\"\u003e\n \u003cp\u003ePlans\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"67.80045351473923%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Certainly, change management is obligatory, especially when introducing a new way of working in clinical practice. One of the major challenges is not the technical part but how to introduce new solutions to an established way of working, that fits in clinical practice without an extra cost. Most healthcare institutions lack this plan\u0026rdquo;- PT2\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"26.622296173044926%\" valign=\"top\"\u003e\n \u003cp\u003eAcceptance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.62728785357737%\" valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"49.750415973377706%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Minor barriers could be acceptance by patients, and physicians. People are interested and believe that the technology can do better but there are still people that don\u0026rsquo;t want their data analyzed or stored.\u0026rdquo;- PT1\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Experts in their field have an established way of working and some of them are struggling with new techniques and not seeing AI as a threat\u0026rdquo;- PT2\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026ldquo;Healthcare professionals may be skeptical or uncomfortable relying on AI for decisionmaking. This resistance can stem from concerns about job security, or doubts about AI capabilities.\u0026rdquo;- PT5\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-digital-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [BMC Digital Health](https://bmcdigitalhealth.biomedcentral.com/)","snPcode":"44247","submissionUrl":"https://submission.nature.com/new-submission/44247/3","title":"BMC Digital Health","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-4370656/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4370656/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCardiovascular diseases (CVDs) are the leading cause of mortality in the world. Artificial Intelligence (AI) offers an opportunity to improve the quality of care provided to cardiovascular patients due to its ability to handle large and complex data. Despite promising results obtained in several studies, widespread adoption of AI in cardiovascular care is lacking due to the existence of some gaps. The goal of this study is to analyze the existing challenges faced by AI solutions in cardiovascular care. This study adopted a mixed-methods research approach, combining semi-structured interviews with responses from a self-administered online survey. A total of 5 interviews were conducted and 91 valid survey responses were obtained. Survey respondents included doctors, nurses, medical researchers, health I specialists, hospital administrators, and other clinically affiliated participants working with cardiovascular patients. Participants identified 8 major challenges: data-related challenges, regulatory challenges, infrastructural challenges, gaps in knowledge, transparency challenges, ethical challenges, issues with change management, and acceptance challenges. These gaps hinder the adoption of AI in cardiovascular care and taking proactive measures to address these challenges is key to fostering AI adoption.\u003c/p\u003e","manuscriptTitle":"From Diagnosis to Management: Unveiling the Challenges of Artificial Intelligence Solutions in Cardiovascular Healthcare","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-05-17 16:37:59","doi":"10.21203/rs.3.rs-4370656/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-05-13T09:48:00+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-05-08T09:55:23+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-05-08T09:55:23+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Digital Health","date":"2024-05-05T07:13:50+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-digital-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [BMC Digital Health](https://bmcdigitalhealth.biomedcentral.com/)","snPcode":"44247","submissionUrl":"https://submission.nature.com/new-submission/44247/3","title":"BMC Digital Health","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f4a2cb3b-977e-4fb0-8f74-216f833ddd9f","owner":[],"postedDate":"May 17th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-06-25T14:23:09+00:00","versionOfRecord":[],"versionCreatedAt":"2024-05-17 16:37:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4370656","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4370656","identity":"rs-4370656","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}