ABSTRACT
Gene expression in human tissue has primarily been studied on the transcriptional level, largely neglecting translational regulation. Here, we analyze the translatomes of 80 human hearts to identify new translation events and quantify the effect of translational regulation. We show extensive translational control of cardiac gene expression, which is orchestrated in a process-specific manner. Translation downstream of predicted disease-causing protein-truncating variants appears to be frequent, suggesting inefficient translation termination. We identify hundreds of previously undetected microproteins, expressed from lncRNAs and circRNAs, for which we validate the protein products in vivo. The translation of microproteins is not restricted to the heart and prominent in the translatomes of human kidney and liver. We associate these microproteins with diverse cellular processes and compartments and find that many locate to the mitochondria. Importantly, dozens of microproteins are translated from lncRNAs with well-characterized noncoding functions, indicating previously unrecognized biology.
Subject(s)
Myocardium/metabolism , Protein Biosynthesis , Adolescent , Adult , Aged , Animals , Codon/genetics , Female , Gene Expression Regulation , HEK293 Cells , Humans , Infant , Male , Mice , Mice, Inbred C57BL , Middle Aged , Open Reading Frames/genetics , RNA, Circular/genetics , RNA, Circular/metabolism , RNA, Long Noncoding/genetics , RNA, Long Noncoding/metabolism , RNA, Messenger/genetics , RNA, Messenger/metabolism , Rats , Ribosomes/genetics , Ribosomes/metabolism , Young AdultABSTRACT
Natural language processing techniques are having an increasing impact on clinical care from patient, clinician, administrator, and research perspective. Among others are automated generation of clinical notes and discharge letters, medical term coding for billing, medical chatbots both for patients and clinicians, data enrichment in the identification of disease symptoms or diagnosis, cohort selection for clinical trial, and auditing purposes. In the review, an overview of the history in natural language processing techniques developed with brief technical background is presented. Subsequently, the review will discuss implementation strategies of natural language processing tools, thereby specifically focusing on large language models, and conclude with future opportunities in the application of such techniques in the field of cardiology.
Subject(s)
Artificial Intelligence , Cardiology , Humans , Natural Language Processing , Patient DischargeABSTRACT
BACKGROUND: The effect of haemodynamic monitoring of pulmonary artery pressure has predominantly been studied in the USA. There is a clear need for randomised trial data from patients treated with contemporary guideline-directed-medical-therapy with long-term follow-up in a different health-care system. METHODS: MONITOR-HF was an open-label, randomised trial, done in 25 centres in the Netherlands. Eligible patients had chronic heart failure of New York Heart Association class III and a previous heart failure hospitalisation, irrespective of ejection fraction. Patients were randomly assigned (1:1) to haemodynamic monitoring (CardioMEMS-HF system, Abbott Laboratories, Abbott Park, IL, USA) or standard care. All patients were scheduled to be seen by their clinician at 3 months and 6 months, and every 6 months thereafter, up to 48 months. The primary endpoint was the mean difference in the Kansas City Cardiomyopathy Questionnaire (KCCQ) overall summary score at 12 months. All analyses were by intention-to-treat. This trial was prospectively registered under the clinical trial registration number NTR7673 (NL7430) on the International Clinical Trials Registry Platform. FINDINGS: Between April 1, 2019, and Jan 14, 2022, we randomly assigned 348 patients to either the CardioMEMS-HF group (n=176 [51%]) or the control group (n=172 [49%]). The median age was 69 years (IQR 61-75) and median ejection fraction was 30% (23-40). The difference in mean change in KCCQ overall summary score at 12 months was 7·13 (95% CI 1·51-12·75; p=0·013) between groups (+7·05 in the CardioMEMS group, p=0·0014, and -0·08 in the standard care group, p=0·97). In the responder analysis, the odds ratio (OR) of an improvement of at least 5 points in KCCQ overall summary score was OR 1·69 (95% CI 1·01-2·83; p=0·046) and the OR of a deterioration of at least 5 points was 0·45 (0·26-0·77; p=0·0035) in the CardioMEMS-HF group compared with in the standard care group. The freedom of device-related or system-related complications and sensor failure were 97·7% and 98·8%, respectively. INTERPRETATION: Haemodynamic monitoring substantially improved quality of life and reduced heart failure hospitalisations in patients with moderate-to-severe heart failure treated according to contemporary guidelines. These findings contribute to the aggregate evidence for this technology and might have implications for guideline recommendations and implementation of remote pulmonary artery pressure monitoring. FUNDING: The Dutch Ministry of Health, Health Care Institute (Zorginstituut), and Abbott Laboratories.
Subject(s)
Heart Failure , Hemodynamic Monitoring , Humans , Aged , Pulmonary Artery , Hemodynamic Monitoring/adverse effects , Quality of Life , Heart Failure/drug therapy , Chronic DiseaseABSTRACT
BACKGROUND: We explored the association between use of renin-angiotensin system inhibitors and beta-blockers, with mortality/morbidity in 5 previously identified clusters of patients with heart failure with preserved ejection fraction (HFpEF). METHODS AND RESULTS: We analyzed 20,980 patients with HFpEF from the Swedish HF registry, phenotyped into young-low comorbidity burden (12%), atrial fibrillation-hypertensive (32%), older-atrial fibrillation (24%), obese-diabetic (15%), and a cardiorenal cluster (17%). In Cox proportional hazard models with inverse probability weighting, there was no heterogeneity in the association between renin-angiotensin system inhibitor use and cluster membership for any of the outcomes: cardiovascular (CV) mortality, all-cause mortality, HF hospitalisation, CV hospitalisation, or non-CV hospitalisation. In contrast, we found a statistical interaction between beta-blocker use and cluster membership for all-cause mortality (Pâ¯=â¯.03) and non-CV hospitalisation (Pâ¯=â¯.001). In the young-low comorbidity burden and atrial fibrillation-hypertensive cluster, beta-blocker use was associated with statistically significant lower all-cause mortality and non-CV hospitalisation and in the obese-diabetic cluster beta-blocker use was only associated with a statistically significant lower non-CV hospitalisation. The interaction between beta-blocker use and cluster membership for all-cause mortality could potentially be driven by patients with improved EF. However, patient numbers were diminished when excluding those with improved EF and the direction of the associations remained similar. CONCLUSIONS: In patients with HFpEF, the association with all-cause mortality and non-CV hospitalisation was heterogeneous across clusters for beta-blockers. It remains to be elucidated how heterogeneity in HFpEF could influence personalized medicine and future clinical trial design.
Subject(s)
Atrial Fibrillation , Diabetes Mellitus , Heart Failure , Humans , Heart Failure/drug therapy , Heart Failure/epidemiology , Renin/therapeutic use , Stroke Volume , Atrial Fibrillation/drug therapy , Atrial Fibrillation/epidemiology , Adrenergic beta-Antagonists/therapeutic use , Diabetes Mellitus/drug therapy , Obesity/drug therapy , Angiotensins/therapeutic useABSTRACT
BACKGROUND: Heart failure (HF) with preserved or mildly reduced ejection fraction includes a heterogenous group of patients. Reclassification into distinct phenogroups to enable targeted interventions is a priority. This study aimed to identify distinct phenogroups, and compare phenogroup characteristics and outcomes, from electronic health record data. METHODS: 2,187 patients admitted to five UK hospitals with a diagnosis of HF and a left ventricular ejection fraction ≥ 40% were identified from the NIHR Health Informatics Collaborative database. Partition-based, model-based, and density-based machine learning clustering techniques were applied. Cox Proportional Hazards and Fine-Gray competing risks models were used to compare outcomes (all-cause mortality and hospitalisation for HF) across phenogroups. RESULTS: Three phenogroups were identified: (1) Younger, predominantly female patients with high prevalence of cardiometabolic and coronary disease; (2) More frail patients, with higher rates of lung disease and atrial fibrillation; (3) Patients characterised by systemic inflammation and high rates of diabetes and renal dysfunction. Survival profiles were distinct, with an increasing risk of all-cause mortality from phenogroups 1 to 3 (p < 0.001). Phenogroup membership significantly improved survival prediction compared to conventional factors. Phenogroups were not predictive of hospitalisation for HF. CONCLUSIONS: Applying unsupervised machine learning to routinely collected electronic health record data identified phenogroups with distinct clinical characteristics and unique survival profiles.
Subject(s)
Electronic Health Records , Heart Failure , Stroke Volume , Ventricular Function, Left , Humans , Heart Failure/physiopathology , Heart Failure/diagnosis , Heart Failure/mortality , Female , Male , Aged , Middle Aged , Risk Assessment , United Kingdom/epidemiology , Risk Factors , Prognosis , Aged, 80 and over , Databases, Factual , Unsupervised Machine Learning , Hospitalization , Time Factors , Comorbidity , Cause of Death , Phenotype , Data MiningABSTRACT
PURPOSEOF REVIEW: Guideline-directed medical therapy (GDMT) underuse is common in heart failure (HF) patients. Digital solutions have the potential to support medical professionals to optimize GDMT prescriptions in a growing HF population. We aimed to review current literature on the effectiveness of digital solutions on optimization of GDMT prescriptions in patients with HF. RECENT FINDINGS: We report on the efficacy, characteristics of the study, and population of published digital solutions for GDMT optimization. The following digital solutions are discussed: teleconsultation, telemonitoring, cardiac implantable electronic devices, clinical decision support embedded within electronic health records, and multifaceted interventions. Effect of digital solutions is reported in dedicated studies, retrospective studies, or larger studies with another focus that also commented on GDMT use. Overall, we see more studies on digital solutions that report a significant increase in GDMT use. However, there is a large heterogeneity in study design, outcomes used, and populations studied, which hampers comparison of the different digital solutions. Barriers, facilitators, study designs, and future directions are discussed. There remains a need for well-designed evaluation studies to determine safety and effectiveness of digital solutions for GDMT optimization in patients with HF. Based on this review, measuring and controlling vital signs in telemedicine studies should be encouraged, professionals should be actively alerted about suboptimal GDMT, the researchers should consider employing multifaceted digital solutions to optimize effectiveness, and use study designs that fit the unique sociotechnical aspects of digital solutions. Future directions are expected to include artificial intelligence solutions to handle larger datasets and relieve medical professional's workload.
Subject(s)
Heart Failure , Telemedicine , Humans , Heart Failure/drug therapy , Artificial Intelligence , Retrospective Studies , Prescriptions , Stroke VolumeABSTRACT
To raise the quality of clinical artificial intelligence (AI) prediction modelling studies in the cardiovascular health domain and thereby improve their impact and relevancy, the editors for digital health, innovation, and quality standards of the European Heart Journal propose five minimal quality criteria for AI-based prediction model development and validation studies: complete reporting, carefully defined intended use of the model, rigorous validation, large enough sample size, and openness of code and software.
Subject(s)
Artificial Intelligence , Software , Humans , HeartABSTRACT
Artificial intelligence (AI) is increasingly being utilized in healthcare. This article provides clinicians and researchers with a step-wise foundation for high-value AI that can be applied to a variety of different data modalities. The aim is to improve the transparency and application of AI methods, with the potential to benefit patients in routine cardiovascular care. Following a clear research hypothesis, an AI-based workflow begins with data selection and pre-processing prior to analysis, with the type of data (structured, semi-structured, or unstructured) determining what type of pre-processing steps and machine-learning algorithms are required. Algorithmic and data validation should be performed to ensure the robustness of the chosen methodology, followed by an objective evaluation of performance. Seven case studies are provided to highlight the wide variety of data modalities and clinical questions that can benefit from modern AI techniques, with a focus on applying them to cardiovascular disease management. Despite the growing use of AI, further education for healthcare workers, researchers, and the public are needed to aid understanding of how AI works and to close the existing gap in knowledge. In addition, issues regarding data access, sharing, and security must be addressed to ensure full engagement by patients and the public. The application of AI within healthcare provides an opportunity for clinicians to deliver a more personalized approach to medical care by accounting for confounders, interactions, and the rising prevalence of multi-morbidity.
Subject(s)
Artificial Intelligence , Cardiovascular System , Humans , Algorithms , Machine Learning , Delivery of Health CareABSTRACT
Telemedicine in heart failure (HF) management may positively impact health outcomes, but varied effects in studies hinder guidance in HF guidelines. Evidence on the effectiveness of telemedicine in HF subpopulations is limited. We conducted a scoping review to evaluate and synthesise evidence on the effectiveness of telemedicine across HF subpopulations that could guide telemedicine strategies in routine practice. Meta-analyses concerning randomised controlled trials (RCTs) with subgroup analyses on telemedicine effectives were identified in PubMed. We identified 15 RCTs, encompassing 21 different subgroups based on characteristics of HF patients. Findings varied across studies and no definite evidence was found about which patients benefit most from telemedicine. Subgroup definitions were inconsistent, not always a priori defined and subgroups contained few patients. Some studies found heterogeneous effects of telemedicine on mortality and hospitalisation across subgroups defined by: New York Heart Association (NYHA) classification, previous HF decompensation, implantable device, concurrent depression, time since hospital discharge and duration of HF. Patients represented in the RCTs were mostly male, aged 65-75 years, with HF with reduced ejection fraction and NYHA class II/III. Traditional RCTs have not been able to provide clinicians with guidance; continuous real-world evidence generation could enhance monitoring and identify who benefits from telemedicine.
ABSTRACT
Randomised clinical trials (RCTs) are vital for medical progress. Unfortunately, 'traditional' RCTs are expensive and inherently slow. Moreover, their generalisability has been questioned. There is considerable overlap in routine health care data (RHCD) and trial-specific data. Therefore, integration of RHCD in an RCT has great potential, as it would reduce the effort and costs required to collect data, thereby overcoming some of the major downsides of a traditional RCT. However, use of RHCD comes with other challenges, such as privacy issues, as well as technical and practical barriers. Here, we give a current overview of related initiatives on national cardiovascular registries (Netherlands Heart Registration, Heart4Data), showcasing the interrelationships between and the relevance of the different registries for the practicing physician. We then discuss the benefits and limitations of RHCD use in the setting of a pragmatic RCT from a cardiovascular perspective, illustrated by a case study in heart failure.
ABSTRACT
IntroductionPatients who have experienced an acute coronary syndrome (ACS) are at risk of a recurrent event, but their level of risk varies. Because of their close temporal relationship with vascular injury, longitudinal measurements of circulating endothelial cells (CECs) carry potential to improve individual risk assessment.MethodsWe conducted an explorative nested case-control study within our multicenter, prospective, observational biomarker study (BIOMArCS) of 844 ACS patients. Following an index ACS, high-frequency blood sampling was performed during 1-year follow-up. CECs were identified using flow cytometric analyses in 15 cases with recurrent event, and 30 matched controls.ResultsCases and controls had a median (25th-75thpercentile) age of 64.1 (58.1-75.1) years and 80% were men. During the months preceding the endpoint, the mean (95%CI) CEC concentration in cases was persistently higher than in controls (12.8 [8.2-20.0] versus 10.0 [7.0-14.4] cells/ml), although this difference was non-significant (P = 0.339). In controls, the mean cell concentration was significantly (P = 0.030) lower in post 30-day samples compared to samples collected within one day after index ACS: 10.1 (7.5-13.6) versus 17.0 (10.8-26.6) cells/ml. Similar results were observed for CEC subsets co-expressing CD133 and CD309 (VEGFR-2) or CD106 (VCAM-1).ConclusionDespite their close relation to vascular damage, no increase in cell concentrations were found prior to the occurrence of a secondary adverse cardiac event.
Subject(s)
Acute Coronary Syndrome , Male , Humans , Middle Aged , Aged , Female , Acute Coronary Syndrome/diagnosis , Endothelial Cells , Prospective Studies , Case-Control Studies , BiomarkersABSTRACT
Whilst the Randomised Controlled Trial remains the gold standard for deriving robust causal estimates of treatment efficacy, too often a traditional design proves prohibitively expensive or cumbersome when it comes to assessing questions regarding the comparative effectiveness of routinely used treatments. As a result, patients experience variation in practice as clinicians lack the evidence needed to personalise treatments effectively. This variation may be classified as unwarranted, where existing evidence is ignored, or legitimate where in the absence of evidence, clinicians rely on experience, expert opinion, and inferred principles from basic science to make decisions. We argue that within the right ethical and technological framework, legitimate variation can be transformed into a mechanism for evidence generation and learning. Learning Health Systems which harness existing variation in practice, represent a novel approach for generating evidence from everyday clinical practice. The development of these systems has gained traction due to the increased availability of modern Electronic Health Record Systems. However, despite their promise, overcoming hurdles to successfully integrating clinical trials within Learning Health Systems has proven challenging. This article describes the origins of integrated clinical trials and explores two main barriers to their further implementation - how best to obtain informed consent from patients to participate in routine comparative effectiveness research, and how to automate and integrate randomisation into a clinical workflow. Having described these barriers, we present a potential solution in the form of a research pipeline using a novel form of flexible point-of-care randomisation to allow clinicians and patients to participate in studies where there is clinical equipoise.
Subject(s)
Electronic Health Records , Point-of-Care Systems , Humans , Research Design , Learning , Informed ConsentABSTRACT
Height is a highly heritable, classic polygenic trait with approximately 700 common associated variants identified through genome-wide association studies so far. Here, we report 83 height-associated coding variants with lower minor-allele frequencies (in the range of 0.1-4.8%) and effects of up to 2 centimetres per allele (such as those in IHH, STC2, AR and CRISPLD2), greater than ten times the average effect of common variants. In functional follow-up studies, rare height-increasing alleles of STC2 (giving an increase of 1-2 centimetres per allele) compromised proteolytic inhibition of PAPP-A and increased cleavage of IGFBP-4 in vitro, resulting in higher bioavailability of insulin-like growth factors. These 83 height-associated variants overlap genes that are mutated in monogenic growth disorders and highlight new biological candidates (such as ADAMTS3, IL11RA and NOX4) and pathways (such as proteoglycan and glycosaminoglycan synthesis) involved in growth. Our results demonstrate that sufficiently large sample sizes can uncover rare and low-frequency variants of moderate-to-large effect associated with polygenic human phenotypes, and that these variants implicate relevant genes and pathways.
Subject(s)
Body Height/genetics , Gene Frequency/genetics , Genetic Variation/genetics , ADAMTS Proteins/genetics , Adult , Alleles , Cell Adhesion Molecules/genetics , Female , Genome, Human/genetics , Glycoproteins/genetics , Glycoproteins/metabolism , Glycosaminoglycans/biosynthesis , Hedgehog Proteins/genetics , Humans , Intercellular Signaling Peptides and Proteins/genetics , Intercellular Signaling Peptides and Proteins/metabolism , Interferon Regulatory Factors/genetics , Interleukin-11 Receptor alpha Subunit/genetics , Male , Multifactorial Inheritance/genetics , NADPH Oxidase 4 , NADPH Oxidases/genetics , Phenotype , Pregnancy-Associated Plasma Protein-A/metabolism , Procollagen N-Endopeptidase/genetics , Proteoglycans/biosynthesis , Proteolysis , Receptors, Androgen/genetics , Somatomedins/metabolismABSTRACT
BACKGROUND: Late right heart failure (LRHF) is a common complication during long-term left ventricular assist device (LVAD) support. We aimed to identify risk factors for LRHF after LVAD implantation. METHODS: Patients undergoing primary LVAD implantation between 2006 and 2019 and surviving the perioperative period were included for this study (n = 261). Univariate Cox proportional hazards analysis was used to assess the association of clinical covariates and LRHF, stratified for device type. Variables with p < 0.10 entered the multivariable model. In a subset of patients with complete echocardiography or right catheterization data, this multivariable model was extended. Postoperative cardiopulmonary exercise test data were compared in patients with and without LRHF. RESULTS: Nineteen percentage of patients suffered from LRHF after a median of 12 months, of which 67% required hospitalization. A history of atrial fibrillation (AF) (HR: 2.06 [1.08-3.93], p = 0.029), a higher preoperative body mass index (BMI) (HR: 1.07 [1.01-1.13], p = 0.023), and intensive care unit (ICU) duration (HR: 1.03 [1.00-1.06], p = 0.025) were independent predictors of LHRF in the multivariable model. A significant relation between the severity of tricuspid regurgitation (TR) and LRHF (HR: 1.91 [1.13-3.21], p = 0.016) was found in patients with echocardiographic data. Patients with LRHF demonstrated a lower maximal workload and peak VO2 at 6 months postoperatively. CONCLUSION: A history of AF, BMI, and longer ICU stay may help identify patients at high risk for LRHF. Severity of TR was significantly related to LRHF in a subset of patients.
Subject(s)
Heart Failure , Heart-Assist Devices , Tricuspid Valve Insufficiency , Humans , Incidence , Treatment Outcome , Retrospective Studies , Heart Failure/epidemiology , Heart Failure/etiology , Risk Factors , Heart-Assist Devices/adverse effectsABSTRACT
Genome-wide association studies (GWASs) have identified at least 10 single-nucleotide polymorphisms (SNPs) associated with papillary thyroid cancer (PTC) risk. Most of these SNPs are common variants with small to moderate effect sizes. Here we assessed the combined genetic effects of these variants on PTC risk by using summarized GWAS results to build polygenic risk score (PRS) models in three PTC study groups from Ohio (1,544 patients and 1,593 controls), Iceland (723 patients and 129,556 controls), and the United Kingdom (534 patients and 407,945 controls). A PRS based on the 10 established PTC SNPs showed a stronger predictive power compared with the clinical factors model, with a minimum increase of area under the receiver-operating curve of 5.4 percentage points (P ≤ 1.0 × 10-9). Adding an extended PRS based on 592,475 common variants did not significantly improve the prediction power compared with the 10-SNP model, suggesting that most of the remaining undiscovered genetic risk in thyroid cancer is due to rare, moderate- to high-penetrance variants rather than to common low-penetrance variants. Based on the 10-SNP PRS, individuals in the top decile group of PRSs have a close to sevenfold greater risk (95% CI, 5.4-8.8) compared with the bottom decile group. In conclusion, PRSs based on a small number of common germline variants emphasize the importance of heritable low-penetrance markers in PTC.
Subject(s)
Biomarkers, Tumor/genetics , Genetic Predisposition to Disease , Multifactorial Inheritance , Thyroid Cancer, Papillary/genetics , Thyroid Neoplasms/genetics , Adult , Case-Control Studies , Cohort Studies , DNA Mutational Analysis , Female , Genome-Wide Association Study , Humans , Iceland/epidemiology , Male , Middle Aged , Models, Genetic , Penetrance , Polymorphism, Single Nucleotide , Predictive Value of Tests , ROC Curve , Risk Assessment/methods , Risk Factors , Thyroid Cancer, Papillary/epidemiology , Thyroid Cancer, Papillary/pathology , Thyroid Gland/pathology , Thyroid Neoplasms/epidemiology , Thyroid Neoplasms/pathology , United Kingdom/epidemiology , United States/epidemiologyABSTRACT
REVIEW PURPOSE: This systematic review aims to summarise clustering studies in heart failure (HF) and guide future clinical trial design and implementation in routine clinical practice. FINDINGS: 34 studies were identified (n = 19 in HF with preserved ejection fraction (HFpEF)). There was significant heterogeneity invariables and techniques used. However, 149/165 described clusters could be assigned to one of nine phenotypes: 1) young, low comorbidity burden; 2) metabolic; 3) cardio-renal; 4) atrial fibrillation (AF); 5) elderly female AF; 6) hypertensive-comorbidity; 7) ischaemic-male; 8) valvular disease; and 9) devices. There was room for improvement on important methodological topics for all clustering studies such as external validation and transparency of the modelling process. The large overlap between the phenotypes of the clustering studies shows that clustering is a robust approach for discovering clinically distinct phenotypes. However, future studies should invest in a phenotype model that can be implemented in routine clinical practice and future clinical trial design. HF = heart failure, EF = ejection fraction, HFpEF = heart failure with preserved ejection fraction, HFrEF = heart failure with reduced ejection fraction, CKD = chronic kidney disease, AF = atrial fibrillation, IHD = ischaemic heart disease, CAD = coronary artery disease, ICD = implantable cardioverter-defibrillator, CRT = cardiac resynchronization therapy, NT-proBNP = N-terminal pro b-type natriuretic peptide, BMI = Body Mass Index, COPD = Chronic obstructive pulmonary disease.
ABSTRACT
This article presents some of the most important developments in the field of digital medicine that have appeared over the last 12 months and are related to cardiovascular medicine. The article consists of three main sections, as follows: (i) artificial intelligence-enabled cardiovascular diagnostic tools, techniques, and methodologies, (ii) big data and prognostic models for cardiovascular risk protection, and (iii) wearable devices in cardiovascular risk assessment, cardiovascular disease prevention, diagnosis, and management. To conclude the article, the authors present a brief further prospective on this new domain, highlighting existing gaps that are specifically related to artificial intelligence technologies, such as explainability, cost-effectiveness, and, of course, the importance of proper regulatory oversight for each clinical implementation.
Subject(s)
Cardiovascular System , Wearable Electronic Devices , Artificial Intelligence , Big Data , Humans , Precision MedicineABSTRACT
The increasing volume and richness of healthcare data collected during routine clinical practice have not yet translated into significant numbers of actionable insights that have systematically improved patient outcomes. An evidence-practice gap continues to exist in healthcare. We contest that this gap can be reduced by assessing the use of nudge theory as part of clinical decision support systems (CDSS). Deploying nudges to modify clinician behaviour and improve adherence to guideline-directed therapy represents an underused tool in bridging the evidence-practice gap. In conjunction with electronic health records (EHRs) and newer devices including artificial intelligence algorithms that are increasingly integrated within learning health systems, nudges such as CDSS alerts should be iteratively tested for all stakeholders involved in health decision-making: clinicians, researchers, and patients alike. Not only could they improve the implementation of known evidence, but the true value of nudging could lie in areas where traditional randomized controlled trials are lacking, and where clinical equipoise and variation dominate. The opportunity to test CDSS nudge alerts and their ability to standardize behaviour in the face of uncertainty may generate novel insights and improve patient outcomes in areas of clinical practice currently without a robust evidence base.
Subject(s)
Decision Support Systems, Clinical , Learning Health System , Artificial Intelligence , Delivery of Health Care , Electronic Health Records , HumansABSTRACT
The medical field has seen a rapid increase in the development of artificial intelligence (AI)-based prediction models. With the introduction of such AI-based prediction model tools and software in cardiovascular patient care, the cardiovascular researcher and healthcare professional are challenged to understand the opportunities as well as the limitations of the AI-based predictions. In this article, we present 12 critical questions for cardiovascular health professionals to ask when confronted with an AI-based prediction model. We aim to support medical professionals to distinguish the AI-based prediction models that can add value to patient care from the AI that does not.
Subject(s)
Artificial Intelligence , Cardiovascular Diseases , Health Personnel , Humans , SoftwareABSTRACT
AIMS: Arrhythmogenic right ventricular dysplasia/cardiomyopathy (ARVC) is characterized by ventricular arrhythmias (VAs) and sudden cardiac death (SCD). We aimed to develop a model for individualized prediction of incident VA/SCD in ARVC patients. METHODS AND RESULTS: Five hundred and twenty-eight patients with a definite diagnosis and no history of sustained VAs/SCD at baseline, aged 38.2 ± 15.5 years, 44.7% male, were enrolled from five registries in North America and Europe. Over 4.83 (interquartile range 2.44-9.33) years of follow-up, 146 (27.7%) experienced sustained VA, defined as SCD, aborted SCD, sustained ventricular tachycardia, or appropriate implantable cardioverter-defibrillator (ICD) therapy. A prediction model estimating annual VA risk was developed using Cox regression with internal validation. Eight potential predictors were pre-specified: age, sex, cardiac syncope in the prior 6 months, non-sustained ventricular tachycardia, number of premature ventricular complexes in 24 h, number of leads with T-wave inversion, and right and left ventricular ejection fractions (LVEFs). All except LVEF were retained in the final model. The model accurately distinguished patients with and without events, with an optimism-corrected C-index of 0.77 [95% confidence interval (CI) 0.73-0.81] and minimal over-optimism [calibration slope of 0.93 (95% CI 0.92-0.95)]. By decision curve analysis, the clinical benefit of the model was superior to a current consensus-based ICD placement algorithm with a 20.3% reduction of ICD placements with the same proportion of protected patients (P < 0.001). CONCLUSION: Using the largest cohort of patients with ARVC and no prior VA, a prediction model using readily available clinical parameters was devised to estimate VA risk and guide decisions regarding primary prevention ICDs (www.arvcrisk.com).