How AI drives innovation in cardiovascular medicine.

Medicine is entering a new era in which artificial intelligence (AI) and deep learning have a measurable impact on patient care. This impact is especially evident in cardiovascular medicine. While the purpose of this short opinion paper is not to provide an in-depth review of the many applications of AI in cardiovascular medicine, we summarize some of the important advances that have taken place in this domain.

Mobilizing data during a crisis: Building rapid evidence pipelines using multi-institutional real world data.

The COVID-19 pandemic generated tremendous interest in using real world data (RWD). Many consortia across the public and private sectors formed in 2020 with the goal of rapidly producing high-quality evidence from RWD to guide medical decision-making, public health priorities, and more. Experiences were gathered from five large consortia on rapid multi-institutional evidence generation during the COVID-19 pandemic. Insights have been compiled across five dimensions: consortium composition, governance structure and alignment of priorities, data sharing, data analysis, and evidence dissemination. The purpose of this piece is to offer guidance on building large-scale multi-institutional RWD analysis pipelines for future public health issues. The composition of each consortium was largely influenced by existing collaborations. A central set of priorities for evidence generation guided each consortium, however different approaches to governance emerged. Challenges surrounding limited access to clinical data due to various contributors were overcome in unique ways. While all consortia used different methods to construct and analyze patient cohorts ranging from centralized to federated approaches, all proved effective for generating meaningful real-world evidence. Actionable recommendations for clinical practice and public health agencies were made from translating insights from consortium analyses. Each consortium was successful in rapidly answering questions about COVID-19 diagnosis and treatment despite all taking slightly different approaches to data sharing and analysis. Leveraging RWD, leveraged in a manner that applies scientific rigor and transparency, can complement higher-level evidence and serve as an important adjunct to clinical trials to quickly guide policy and critical care, especially for a pandemic response.

Artificial Intelligence and Pain Medicine: An Introduction.

Artificial intelligence was introduced 60 years ago and has evolved immensely since that time. While artificial intelligence is found in nearly all aspects of our life, the use of artificial intelligence in the healthcare industry has only recently become apparent and more widely discussed. It is expected that artificial intelligence will allow improved disease recognition, treatment optimization, cost and time savings, product development, decision making, and marketing. For pain medicine specifically, these same benefits will be translatable and we can expect better disease recognition and treatment selection. As adoption occurs with this impressive technology, it will be imperative for the pain medicine community to be informed on proper definitions and expected use cases for artificial intelligence. Our objective was to provide pain medicine physicians an overview of artificial intelligence, including important definitions to aid understanding, and to offer potential clinical applications pertinent to the specialty.

A Nationwide Network of Health AI Assurance Laboratories.

Importance: Given the importance of rigorous development and evaluation standards needed of artificial intelligence (AI) models used in health care, nationwide accepted procedures to provide assurance that the use of AI is fair, appropriate, valid, effective, and safe are urgently needed. Observations: While there are several efforts to develop standards and best practices to evaluate AI, there is a gap between having such guidance and the application of such guidance to both existing and new AI models being developed. As of now, there is no publicly available, nationwide mechanism that enables objective evaluation and ongoing assessment of the consequences of using health AI models in clinical care settings. Conclusion and Relevance: The need to create a public-private partnership to support a nationwide health AI assurance labs network is outlined here. In this network, community best practices could be applied for testing health AI models to produce reports on their performance that can be widely shared for managing the lifecycle of AI models over time and across populations and sites where these models are deployed.

Data-Driven and Technology-Enabled Trial Innovations Toward Decentralization of Clinical Trials: Opportunities and Considerations.

Traditional trial designs have well-recognized inefficiencies and logistical barriers to participation. Decentralized trials and digital health solutions have been suggested as potential solutions and have certainly risen to the challenge during the pandemic. Clinical trial designs are now increasingly data driven. The use of distributed clinical data networks and digitization has helped to fundamentally upgrade existing research systems. A trial design may vary anywhere from fully decentralized to hybrid to traditional on-site. Various decentralization components are available for stakeholders to increase the reach and pace of their trials, such as electronic informed consent, remote interviews, administration, outcome assessment, monitoring, and laboratory and imaging modalities. Furthermore, digital health technologies can be included to enrich study conduct. However, careful consideration is warranted, including assessing verification and validity through usability studies and having various contingencies in place through dedicated risk assessment. Selecting the right combination depends not just on the ability to handle patient care and the medical know-how but also on the availability of appropriate technologic infrastructure, skills, and human resources. Throughout this process, quality of evidence generation and physician-patient relation must not be undermined. Here we also address some knowledge gaps, cost considerations, and potential impact of decentralization and digitization on inclusivity, recruitment, engagement, and retention. Last, we mention some future directions that may help drive the necessary change in the right direction.

Expanding repertoire of SARS-CoV-2 deletion mutations contributes to evolution of highly transmissible variants.

The emergence of highly transmissible SARS-CoV-2 variants and vaccine breakthrough infections globally mandated the characterization of the immuno-evasive features of SARS-CoV-2. Here, we systematically analyzed 2.13 million SARS-CoV-2 genomes from 188 countries/territories (up to June 2021) and performed whole-genome viral sequencing from 102 COVID-19 patients, including 43 vaccine breakthrough infections. We identified 92 Spike protein mutations that increased in prevalence during at least one surge in SARS-CoV-2 test positivity in any country over a 3-month window. Deletions in the Spike protein N-terminal domain were highly enriched for these 'surge-associated mutations' (Odds Ratio = 14.19, 95% CI 6.15-32.75, p value = 3.41 × 10-10). Based on a longitudinal analysis of mutational prevalence globally, we found an expanding repertoire of Spike protein deletions proximal to an antigenic supersite in the N-terminal domain that may be one of the key contributors to the evolution of highly transmissible variants. Finally, we generated clinically annotated SARS-CoV-2 whole genome sequences from 102 patients and identified 107 unique mutations, including 78 substitutions and 29 deletions. In five patients, we identified distinct deletions between residues 85-90, which reside within a linear B cell epitope. Deletions in this region arose contemporaneously on a diverse background of variants across the globe since December 2020. Overall, our findings based on genomic-epidemiology and clinical surveillance suggest that the genomic deletion of dispensable antigenic regions in SARS-CoV-2 may contribute to the evasion of immune responses and the evolution of highly transmissible variants.

Prospective evaluation of smartwatch-enabled detection of left ventricular dysfunction.

Although artificial intelligence (AI) algorithms have been shown to be capable of identifying cardiac dysfunction, defined as ejection fraction (EF) ≤ 40%, from 12-lead electrocardiograms (ECGs), identification of cardiac dysfunction using the single-lead ECG of a smartwatch has yet to be tested. In the present study, a prospective study in which patients of Mayo Clinic were invited by email to download a Mayo Clinic iPhone application that sends watch ECGs to a secure data platform, we examined patient engagement with the study app and the diagnostic utility of the ECGs. We digitally enrolled 2,454 unique patients (mean age 53 ± 15 years, 56% female) from 46 US states and 11 countries, who sent 125,610 ECGs to the data platform between August 2021 and February 2022; 421 participants had at least one watch-classified sinus rhythm ECG within 30 d of an echocardiogram, of whom 16 (3.8%) had an EF ≤ 40%. The AI algorithm detected patients with low EF with an area under the curve of 0.885 (95% confidence interval 0.823-0.946) and 0.881 (0.815-0.947), using the mean prediction within a 30-d window or the closest ECG relative to the echocardiogram that determined the EF, respectively. These findings indicate that consumer watch ECGs, acquired in nonclinical environments, can be used to identify patients with cardiac dysfunction, a potentially life-threatening and often asymptomatic condition.

Moving towards vertically integrated artificial intelligence development.

Substantial interest and investment in clinical artificial intelligence (AI) research has not resulted in widespread translation to deployed AI solutions. Current attention has focused on bias and explainability in AI algorithm development, external validity and model generalisability, and lack of equity and representation in existing data. While of great importance, these considerations also reflect a model-centric approach seen in published clinical AI research, which focuses on optimising architecture and performance of an AI model on best available datasets. However, even robustly built models using state-of-the-art algorithms may fail once tested in realistic environments due to unpredictability of real-world conditions, out-of-dataset scenarios, characteristics of deployment infrastructure, and lack of added value to clinical workflows relative to cost and potential clinical risks. In this perspective, we define a vertically integrated approach to AI development that incorporates early, cross-disciplinary, consideration of impact evaluation, data lifecycles, and AI production, and explore its implementation in two contrasting AI development pipelines: a scalable "AI factory" (Mayo Clinic, Rochester, United States), and an end-to-end cervical cancer screening platform for resource poor settings (Paps AI, Mbarara, Uganda). We provide practical recommendations for implementers, and discuss future challenges and novel approaches (including a decentralised federated architecture being developed in the NHS (AI4VBH, London, UK)). Growth in global clinical AI research continues unabated, and introduction of vertically integrated teams and development practices can increase the translational potential of future clinical AI projects.

Addressing racial disparities in surgical care with machine learning.

There is ample evidence to demonstrate that discrimination against several population subgroups interferes with their ability to receive optimal surgical care. This bias can take many forms, including limited access to medical services, poor quality of care, and inadequate insurance coverage. While such inequalities will require numerous cultural, ethical, and sociological solutions, artificial intelligence-based algorithms may help address the problem by detecting bias in the data sets currently being used to make medical decisions. However, such AI-based solutions are only in early development. The purpose of this commentary is to serve as a call to action to encourage investigators and funding agencies to invest in the development of these digital tools.

Assessing socioeconomic bias in machine learning algorithms in health care: a case study of the HOUSES index.

OBJECTIVE: Artificial intelligence (AI) models may propagate harmful biases in performance and hence negatively affect the underserved. We aimed to assess the degree to which data quality of electronic health records (EHRs) affected by inequities related to low socioeconomic status (SES), results in differential performance of AI models across SES. MATERIALS AND METHODS: This study utilized existing machine learning models for predicting asthma exacerbation in children with asthma. We compared balanced error rate (BER) against different SES levels measured by HOUsing-based SocioEconomic Status measure (HOUSES) index. As a possible mechanism for differential performance, we also compared incompleteness of EHR information relevant to asthma care by SES. RESULTS: Asthmatic children with lower SES had larger BER than those with higher SES (eg, ratio = 1.35 for HOUSES Q1 vs Q2-Q4) and had a higher proportion of missing information relevant to asthma care (eg, 41% vs 24% for missing asthma severity and 12% vs 9.8% for undiagnosed asthma despite meeting asthma criteria). DISCUSSION: Our study suggests that lower SES is associated with worse predictive model performance. It also highlights the potential role of incomplete EHR data in this differential performance and suggests a way to mitigate this bias. CONCLUSION: The HOUSES index allows AI researchers to assess bias in predictive model performance by SES. Although our case study was based on a small sample size and a single-site study, the study results highlight a potential strategy for identifying bias by using an innovative SES measure.

A proposal for developing a platform that evaluates algorithmic equity and accuracy.

We are at a pivotal moment in the development of healthcare artificial intelligence (AI), a point at which enthusiasm for machine learning has not caught up with the scientific evidence to support the equity and accuracy of diagnostic and therapeutic algorithms. This proposal examines algorithmic biases, including those related to race, gender and socioeconomic status, and accuracy, including the paucity of prospective studies and lack of multisite validation. We then suggest solutions to these problems. We describe the Mayo Clinic, Duke University, Change Healthcare project that is evaluating 35.1 billion healthcare records for bias. And we propose 'Ingredients' style labels and an AI evaluation/testing system to help clinicians judge the merits of products and services that include algorithms. Said testing would include input data sources and types, dataset population composition, algorithm validation techniques, bias assessment evaluation and performance metrics.

Surveillance of Safety of 3 Doses of COVID-19 mRNA Vaccination Using Electronic Health Records.

Importance: Recent reports on waning of COVID-19 vaccine-induced immunity have led to the approval and rollout of additional doses and booster vaccinations. Individuals at increased risk of SARS-CoV-2 infection are receiving additional vaccine doses in addition to the regimen that was tested in clinical trials. Risks and adverse event profiles associated with additional vaccine doses are currently not well understood. Objective: To evaluate the safety of third-dose vaccination with US Food and Drug Administration (FDA)-approved COVID-19 mRNA vaccines. Design, Setting, and Participants: This cohort study was conducted using electronic health record (EHR) data from December 2020 to October 2021 from the multistate Mayo Clinic Enterprise. Participants included all 47 999 individuals receiving 3-dose COVID-19 mRNA vaccines within the study setting who met study inclusion criteria. Participants were divided into 2 cohorts by vaccine brand administered and served as their own control groups, with no comparison made between cohorts. Data were analyzed from September through November 2021. Exposures: Three doses of an FDA-authorized COVID-19 mRNA vaccine, BNT162b2 or mRNA-1273. Main Outcomes and Measures: Vaccine-associated adverse events were assessed via EHR report. Adverse event risk was quantified using the percentage of study participants who reported the adverse event within 14 days after each vaccine dose and during a 14-day control period, immediately preceding the first vaccine dose. Results: Among 47 999 individuals who received 3-dose COVID-19 mRNA vaccines, 38 094 individuals (21 835 [57.3%] women; median [IQR] age, 67.4 [52.5-76.5] years) received BNT162b2 (79.4%) and 9905 individuals (5099 [51.5%] women; median [IQR] age, 67.7 [59.5-73.9] years) received mRNA-1273 (20.6%). Reporting of severe adverse events remained low after the third vaccine dose, with rates of pericarditis (0.01%; 95% CI, 0%-0.02%), anaphylaxis (0%; 95% CI, 0%-0.01%), myocarditis (0%; 95% CI, 0%-0.01%), and cerebral venous sinus thrombosis (no individuals) consistent with results from earlier studies. Significantly more individuals reported low-severity adverse events after the third dose compared with after the second dose, including fatigue (2360 individuals [4.92%] vs 1665 individuals [3.47%]; P < .001), lymphadenopathy (1387 individuals [2.89%] vs 995 individuals [2.07%]; P < .001), nausea (1259 individuals [2.62%] vs 979 individuals [2.04%]; P < .001), headache (1185 individuals [2.47%] vs 992 individuals [2.07%]; P < .001), arthralgia (1019 individuals [2.12%] vs 816 individuals [1.70%]; P < .001), myalgia (956 individuals [1.99%] vs 784 individuals [1.63%]; P < .001), diarrhea (817 individuals [1.70%] vs 595 individuals [1.24%]; P < .001), fever (533 individuals [1.11%] vs 391 individuals [0.81%]; P < .001), vomiting (528 individuals [1.10%] vs 385 individuals [0.80%]; P < .001), and chills (224 individuals [0.47%] vs 175 individuals [0.36%]; P = .01). Conclusions and Relevance: This study found that although third-dose vaccination against SARS-CoV-2 infection was associated with increased reporting of low-severity adverse events, risk of severe adverse events remained comparable with risk associated with the standard 2-dose regime. These findings suggest the safety of third vaccination doses in individuals who were eligible for booster vaccination at the time of this study.

The digital transformation of hepatology: The patient is logged in.

The rise in innovative digital health technologies has led a paradigm shift in health care toward personalized, patient-centric medicine that is reaching beyond traditional brick-and-mortar facilities into patients' homes and everyday lives. Digital solutions can monitor and detect early changes in physiological data, predict disease progression and health-related outcomes based on individual risk factors, and manage disease intervention with a range of accessible telemedicine and mobile health options. In this review, we discuss the unique transformation underway in the care of patients with liver disease, specifically examining the digital transformation of diagnostics, prediction and clinical decision-making, and management. Additionally, we discuss the general considerations needed to confirm validity and oversight of new technologies, usability and acceptability of digital solutions, and equity and inclusivity of vulnerable populations.

Enabling Research and Clinical Use of Patient-Generated Health Data (the mindLAMP Platform): Digital Phenotyping Study.

BACKGROUND: There is a growing need for the integration of patient-generated health data (PGHD) into research and clinical care to enable personalized, preventive, and interactive care, but technical and organizational challenges, such as the lack of standards and easy-to-use tools, preclude the effective use of PGHD generated from consumer devices, such as smartphones and wearables. OBJECTIVE: This study outlines how we used mobile apps and semantic web standards such as HTTP 2.0, Representational State Transfer, JSON (JavaScript Object Notation), JSON Schema, Transport Layer Security (version 1.3), Advanced Encryption Standard-256, OpenAPI, HTML5, and Vega, in conjunction with patient and provider feedback to completely update a previous version of mindLAMP. METHODS: The Learn, Assess, Manage, and Prevent (LAMP) platform addresses the abovementioned challenges in enhancing clinical insight by supporting research, data analysis, and implementation efforts around PGHD as an open-source solution with freely accessible and shared code. RESULTS: With a simplified programming interface and novel data representation that captures additional metadata, the LAMP platform enables interoperability with existing Fast Healthcare Interoperability Resources-based health care systems as well as consumer wearables and services such as Apple HealthKit and Google Fit. The companion Cortex data analysis and machine learning toolkit offer robust support for artificial intelligence, behavioral feature extraction, interactive visualizations, and high-performance data processing through parallelization and vectorization techniques. CONCLUSIONS: The LAMP platform incorporates feedback from patients and clinicians alongside a standards-based approach to address these needs and functions across a wide range of use cases through its customizable and flexible components. These range from simple survey-based research to international consortiums capturing multimodal data to simple delivery of mindfulness exercises through personalized, just-in-time adaptive interventions.

Comparative effectiveness of mRNA-1273 and BNT162b2 against symptomatic SARS-CoV-2 infection.

BACKGROUND: mRNA coronavirus disease 2019 (COVID-19) vaccines are safe and effective, but increasing reports of breakthrough infections highlight the need to vigilantly monitor and compare the effectiveness of these vaccines. METHODS: We retrospectively compared protection against symptomatic infection conferred by mRNA-1273 and BNT162b2 at Mayo Clinic sites from December 2020 to September 2021. We used a test-negative case-control design to estimate vaccine effectiveness (VE) and to compare the odds of symptomatic infection after full vaccination with mRNA-1273 versus BNT162b2, while adjusting for age, sex, race, ethnicity, geography, comorbidities, and calendar time of vaccination and testing. FINDINGS: Both vaccines were highly effective over the study duration (VEmRNA-1273: 84.1%, 95% confidence interval [CI]: 81.6%-86.2%; VEBNT162b2: 75.6%, 95% CI: 72.2%-78.7%), but their effectiveness was reduced during July-September (VEmRNA-1273: 75.6%, 95% CI: 70.1%-80%; VEBNT162b2: 63.5%, 95% CI: 55.8%-69.9%) as compared to December-May (VEmRNA-1273: 93.7%, 95% CI: 90.4%-95.9%; VEBNT162b2: 85.7%, 95% CI: 81.4%-88.9%). Adjusted for demographic characteristics, clinical comorbidities, time of vaccination, and time of testing, the odds of experiencing a symptomatic breakthrough infection were lower after full vaccination with mRNA-1273 than with BNT162b2 (odds ratio: 0.60; 95% CI: 0.55-0.67). CONCLUSIONS: Both mRNA-1273 and BNT162b2 strongly protect against symptomatic severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. It is imperative to continue monitoring and comparing available vaccines over time and with respect to emerging variants to inform public and global health decisions. FUNDING: This study was funded by nference.

Best practices in the real-world data life cycle.

With increasing digitization of healthcare, real-world data (RWD) are available in greater quantity and scope than ever before. Since the 2016 United States 21st Century Cures Act, innovations in the RWD life cycle have taken tremendous strides forward, largely driven by demand for regulatory-grade real-world evidence from the biopharmaceutical sector. However, use cases for RWD continue to grow in number, moving beyond drug development, to population health and direct clinical applications pertinent to payors, providers, and health systems. Effective RWD utilization requires disparate data sources to be turned into high-quality datasets. To harness the potential of RWD for emerging use cases, providers and organizations must accelerate life cycle improvements that support this process. We build on examples obtained from the academic literature and author experience of data curation practices across a diverse range of sectors to describe a standardized RWD life cycle containing key steps in production of useful data for analysis and insights. We delineate best practices that will add value to current data pipelines. Seven themes are highlighted that ensure sustainability and scalability for RWD life cycles: data standards adherence, tailored quality assurance, data entry incentivization, deploying natural language processing, data platform solutions, RWD governance, and ensuring equity and representation in data.