Obesity and the Bidirectional Risk of Cancer and Cardiovascular Diseases in African Americans: Disparity vs. Ancestry.

Cardiovascular disease (CVD) and cancer often occur in the same individuals, in part due to the shared risk factors such as obesity. Obesity promotes adipose inflammation, which is pathogenically linked to both cardiovascular disease and cancer. Compared with Caucasians, the prevalence of obesity is significantly higher in African Americans (AA), who exhibit more pronounced inflammation and, in turn, suffer from a higher burden of CVD and cancer-related mortality. The mechanisms that underlie this association among obesity, inflammation, and the bidirectional risk of CVD and cancer, particularly in AA, remain to be determined. Socio-economic disparities such as lack of access to healthy and affordable food may promote obesity and exacerbate hypertension and other CVD risk factors in AA. In turn, the resulting pro-inflammatory milieu contributes to the higher burden of CVD and cancer in AA. Additionally, biological factors that regulate systemic inflammation may be contributory. Mutations in atypical chemokine receptor 1 (ACKR1), otherwise known as the Duffy antigen receptor for chemokines (DARC), confer protection against malaria. Many AAs carry a mutation in the gene encoding this receptor, resulting in loss of its expression. ACKR1 functions as a decoy chemokine receptor, thus dampening chemokine receptor activation and inflammation. Published and preliminary data in humans and mice genetically deficient in ACKR1 suggest that this common gene mutation may contribute to ethnic susceptibility to obesity-related disease, CVD, and cancer. In this narrative review, we present the evidence regarding obesity-related disparities in the bidirectional risk of CVD and cancer and also discuss the potential association of gene polymorphisms in AAs with emphasis on ACKR1.

Are We Ready to Integrate Artificial Intelligence Literacy into Medical School Curriculum: Students and Faculty Survey.

BACKGROUND: The effects of Artificial Intelligence (AI) technology applications are already felt in healthcare in general and in the practice of medicine in the disciplines of radiology, pathology, ophthalmology, and oncology. The expanding interface between digital data science, emerging AI technologies and healthcare is creating a demand for AI technology literacy in health professions. OBJECTIVE: To assess medical student and faculty attitudes toward AI, in preparation for teaching AI foundations and data science applications in clinical practice in an integrated medical education curriculum. METHODS: An online 15-question semi-structured survey was distributed among medical students and faculty. The questionnaire consisted of 3 parts: participant's background, AI awareness, and attitudes toward AI applications in medicine. RESULTS: A total of 121 medical students and 52 clinical faculty completed the survey. Only 30% of students and 50% of faculty responded that they were aware of AI topics in medicine. The majority of students (72%) and faculty (59%) learned about AI from the media. Faculty were more likely to report that they did not have a basic understanding of AI technologies (χ2, P = .031). Students were more interested in AI in patient care training, while faculty were more interested in AI in teaching training (χ2, P = .001). Additionally, students and faculty reported comparable attitudes toward AI, limited AI literacy and time constraints in the curriculum. There is interest in broad and deep AI topics. Our findings in medical learners and teaching faculty parallel other published professional groups' AI survey results. CONCLUSIONS: The survey conclusively proved interest among medical students and faculty in AI technology in general, and in its applications in healthcare and medicine. The study was conducted at a single institution. This survey serves as a foundation for other medical schools interested in developing a collaborative programming approach to address AI literacy in medical education.

A County-Level Analysis of Socioeconomic and Clinical Predictors of COVID-19 Incidence and Case-Fatality Rates in Georgia, March-September 2020.

OBJECTIVES: The global COVID-19 pandemic has affected various populations differently. We investigated the relationship between socioeconomic determinants of health obtained from the Robert Wood Johnson Foundation County Health Rankings and COVID-19 incidence and mortality at the county level in Georgia. METHODS: We analyzed data on COVID-19 incidence and case-fatality rates (CFRs) from the Georgia Department of Public Health from March 1 through August 31, 2020. We used repeated measures generalized linear mixed models to determine differences over time in Georgia counties among quartile health rankings of health outcomes, health behaviors, clinical care, social and economic factors, and physical environment. RESULTS: COVID-19 incidence per 100 000 population increased across all quartile county groups for all health rankings (range, 23.1-51.6 in May to 688.4-1062.0 in August). COVID-19 CFRs per 100 000 population peaked in April and May (range, 3312-6835) for all health rankings, declined in June and July (range, 827-5202), and increased again in August (range, 1877-3310). Peak CFRs occurred later in counties with low health rankings for health behavior and clinical care and in counties with high health rankings for social and economic factors and physical environment. All interactions between the health ranking quartile variables and month were significant (P < .001). County-level Gini indices were associated with significantly higher rates of COVID-19 incidence (P < .001) but not CFRs. CONCLUSIONS: From March through August 2020, COVID-19 incidence rose in Georgia's counties independent of health rankings categorization. Differences in time to peak CFRs differed at the county level based upon key health rankings. Public health interventions should incorporate unique strategies to improve COVID-19-related patient outcomes in these environments.

Machine Intelligence in Cardiovascular Medicine.

The computer science technology trend called artificial intelligence (AI) is not new. Both machine learning and deep learning AI applications have recently begun to impact cardiovascular medicine. Scientists working in the AI domain have long recognized the importance of data quality and provenance to AI algorithm efficiency and accuracy. A diverse array of cardiovascular raw data sources of variable quality-electronic medical records, radiological picture archiving and communication systems, laboratory results, omics, etc.-are available to train AI algorithms for predictive modeling of clinical outcomes (in-hospital mortality, acute coronary syndrome risk stratification, etc.), accelerated image interpretation (edge detection, tissue characterization, etc.) and enhanced phenotyping of heterogeneous conditions (heart failure with preserved ejection fraction, hypertension, etc.). A number of software as medical device narrow AI products for cardiac arrhythmia characterization and advanced image deconvolution are now Food and Drug Administration approved, and many others are in the pipeline. Present and future health professionals using AI-infused analytics and wearable devices have 3 critical roles to play in their informed development and ethical application in practice: (1) medical domain experts providing clinical context to computer and data scientists, (2) data stewards assuring the quality, relevance and provenance of data inputs, and (3) real-time and post-hoc interpreters of AI black box solutions and recommendations to patients. The next wave of so-called contextual adaption AI technologies will more closely approximate human decision-making, potentially augmenting cardiologists' real-time performance in emergency rooms, catheterization laboratories, imaging suites, and clinics. However, before such higher order AI technologies are adopted in the clinical setting and by healthcare systems, regulatory agencies, and industry must jointly develop robust AI standards of practice and transparent technology insertion rule sets.

The medical AI insurgency: what physicians must know about data to practice with intelligent machines.

Machine learning (ML) and its parent technology trend, artificial intelligence (AI), are deriving novel insights from ever larger and more complex datasets. Efficient and accurate AI analytics require fastidious data science-the careful curating of knowledge representations in databases, decomposition of data matrices to reduce dimensionality, and preprocessing of datasets to mitigate the confounding effects of messy (i.e., missing, redundant, and outlier) data. Messier, bigger and more dynamic medical datasets create the potential for ML computing systems querying databases to draw erroneous data inferences, portending real-world human health consequences. High-dimensional medical datasets can be static or dynamic. For example, principal component analysis (PCA) used within R computing packages can speed & scale disease association analytics for deriving polygenic risk scores from static gene-expression microarrays. Robust PCA of k-dimensional subspace data accelerates image acquisition and reconstruction of dynamic 4-D magnetic resonance imaging studies, enhancing tracking of organ physiology, tissue relaxation parameters, and contrast agent effects. Unlike other data-dense business and scientific sectors, medical AI users must be aware that input data quality limitations can have health implications, potentially reducing analytic model accuracy for predicting clinical disease risks and patient outcomes. As AI technologies find more health applications, physicians should contribute their health domain expertize to rules-/ML-based computer system development, inform input data provenance and recognize the importance of data preprocessing quality assurance before interpreting the clinical implications of intelligent machine outputs to patients.

Addressing the Diversity-Validity Dilemma Using Situational Judgment Tests.

PURPOSE: To examine the magnitudes of score differences across different demographic groups for three academic (grade point average [GPA], old Medical College Admission Test [MCAT], and MCAT 2015) and one nonacademic (situational judgment test [SJT]) screening measures and one nonacademic (multiple mini-interview [MMI]) interview measure (analysis 1), and the demographic implications of including an SJT in the screening stage for the pool of applicants who are invited to interview (analysis 2). METHOD: The authors ran the analyses using data from New York Medical College School of Medicine applicants from the 2015-2016 admissions cycle. For analysis 1, effect sizes (Cohen d) were calculated for GPA, old MCAT, MCAT 2015, CASPer (an online SJT), and MMI. Comparisons were made across gender, race, ethnicity (African American, Hispanic/Latino), and socioeconomic status (SES). For analysis 2, a series of simulations were conducted to estimate the number of underrepresented in medicine (UIM) applicants who would have been invited to interview with different weightings of GPA, MCAT, and CASPer scores. RESULTS: A total of 9,096 applicants were included in analysis 1. Group differences were significantly smaller or reversed for CASPer and MMI compared with the academic assessments (MCAT, GPA) across nearly all demographic variables/indicators. The simulations suggested that a higher weighting of CASPer may help increase gender, racial, and ethnic diversity in the interview pool; results for low-SES applicants were mixed. CONCLUSIONS: The inclusion of an SJT in the admissions process has the potential to widen access to medical education for a number of UIM groups.

How Cognitive Machines Can Augment Medical Imaging.

OBJECTIVE: Artificial intelligence (AI) neural networks rapidly convert disparate facts and data into highly predictive analytic models. Machine learning maps image-patient phenotype correlations opaque to standard statistics. Deep learning performs accurate image-derived tissue characterization and can generate virtual CT images from MRI datasets. Natural language processing reads medical literature and efficiently reconfigures years of PACS and electronic medical record information. CONCLUSION: AI logistics solve radiology informatics workflow pain points. Imaging professionals and companies will drive health care AI technology insertion. Data science and computer science will jointly potentiate the impact of AI applications for medical imaging.

Artificial Intelligence in Medical Practice: The Question to the Answer?

Computer science advances and ultra-fast computing speeds find artificial intelligence (AI) broadly benefitting modern society-forecasting weather, recognizing faces, detecting fraud, and deciphering genomics. AI's future role in medical practice remains an unanswered question. Machines (computers) learn to detect patterns not decipherable using biostatistics by processing massive datasets (big data) through layered mathematical models (algorithms). Correcting algorithm mistakes (training) adds to AI predictive model confidence. AI is being successfully applied for image analysis in radiology, pathology, and dermatology, with diagnostic speed exceeding, and accuracy paralleling, medical experts. While diagnostic confidence never reaches 100%, combining machines plus physicians reliably enhances system performance. Cognitive programs are impacting medical practice by applying natural language processing to read the rapidly expanding scientific literature and collate years of diverse electronic medical records. In this and other ways, AI may optimize the care trajectory of chronic disease patients, suggest precision therapies for complex illnesses, reduce medical errors, and improve subject enrollment into clinical trials.

A framework for evaluating student perceptions of health policy training in medical school.

PROBLEM: Nearly half of graduating medical students in the United States report that medical school provides inadequate instruction in topics related to health policy. Although most medical schools report some form of policy education, there lacks a standard for teaching core concepts and evaluating student satisfaction. APPROACH: Responses to the Association of American Medical College's Medical School Graduation Questionnaire were obtained for the years 2007-2008 and 2011-2012 and mapped to domains of training in health policy curricula for four domains: systems and principles; value and equity; quality and safety; and politics and law. Chi-square tests were used to test differences among unadjusted temporal trends. Multiple logistic regression models were fit to the outcome variables and adjusted for student characteristics, student preferences, and medical school characteristics. OUTCOMES: Compared with 2007-2008, students' perceptions of training in 2011-2012 increased on a relative basis by 11.7% for components within systems and principles, 2.8% for quality and safety, and 6.8% for value and equity. Components within politics and law had a composite decline of 4.8%. Multiple logistic regression models found higher odds of reporting satisfaction with training over time for all components within the domains of systems and principles, quality and safety, and value and equity (P < .01), with the exception of medical economics. NEXT STEPS: Medical student perceptions of training in health policy improved over time. Causal factors for these trends require further study. Despite improvement, nearly 40% of graduating medical students still report inadequate instruction in health policy.

Incremental cost-effectiveness of exercise echocardiography vs. SPECT imaging for the evaluation of stable chest pain.

AIMS: Technological advances in cardiac imaging have led to dramatic increases in test utilization and consumption of a growing proportion of cardiovascular healthcare costs. The opportunity costs of strategies favouring exercise echocardiography or SPECT imaging have been incompletely evaluated. METHODS AND RESULTS: We examined prognosis and cost-effectiveness of exercise echocardiography (n = 4884) vs. SPECT (n = 4637) imaging in stable, intermediate risk, chest pain patients. Ischaemia extent was defined as the number of vascular territories with echocardiographic wall motion or SPECT perfusion abnormalities. Cox proportional hazard models were employed to assess time to cardiac death or myocardial infarction (MI). Total cardiovascular costs were summed (discounted and inflation-corrected) throughout follow-up. A cost-effectiveness ratio < Dollars 50,000 per life year saved (LYS) was considered favourable for economic efficiency. The risk-adjusted 3-year death or MI rates classified by extent of ischaemia were similar, ranging from 2.3 to 8.0% for echocardiography and from 3.5 to 11.0% for SPECT (model chi2 = 216; P < 0.0001; interaction P = 0.24). Cost-effectiveness ratios for echocardiography were < Dollars 20,000/LYS when the annual risk of death or MI was < 2%. However, when yearly cardiac event rate were > 2%, cost-effectiveness ratios for echocardiography vs. SPECT were in the range of Dollars 66,686-Dollars 419,522/LYS. For patients with established coronary disease (i.e. > or = 2% annual event risk), SPECT ischaemia was associated with earlier and greater utilization of coronary revascularization (P < 0.0001) resulting in an incremental cost-effectiveness ratio of Dollars 32,381/LYS. CONCLUSION: Health care policies aimed at allocating limited resources can be effectively guided by applying clinical and economic outcomes evidence. A strategy aimed at cost-effective testing would support using echocardiography in low-risk patients with suspected coronary disease, whereas those higher risk patients benefit from referral to SPECT imaging.

Coronary artery disease: diagnostic and prognostic models for reducing patient risk.

Early and accurate diagnostic testing is a critical factor in the detection and optimal management of coronary artery disease (CAD); thus, noninvasive cardiac imaging has become a central tool for CAD evaluation. Currently, tests used for evaluating CAD include conventional resting and stress electrocardiogram, echocardiography, and myocardial perfusion imaging--the most widely used imaging test for evaluation of suspected myocardial ischemia. Emerging techniques for noninvasive assessment of myocardial perfusion and coronary angiography include cardiac computed tomography, cardiac magnetic resonance imaging, and positron emission tomography. The distinctive pathophysiology of atherosclerosis can be used together with imaging techniques to diagnose and assess risk for CAD. Imaging modalities for cardiac risk stratification include a diverse array of tools, such as noninvasive tests that visualize presymptomatic atherosclerosis to sophisticated radionuclide protocols that identify myocardial viability. Of the noninvasive imaging tests, gated SPECT is the most accurate method for risk stratification of CAD. Myocardial perfusion imaging with SPECT has improved accuracy and image quality such that a shift from diagnostic to prognostic use has occurred. Radionuclide myocardial perfusion imaging has played an important role in CAD evaluation since the introduction of thallium-201 (Tl-201) in the 1970s. Although Tl-201 has high sensitivity, specificity, and reproducibility, it also has physical properties that limit its use and affect image quality. Currently, Tc-99m tetrofosmin and sestamibi are the most commonly used agents for a variety of resting and stress protocols, both have similar diagnostic accuracy profiles. The field of nuclear cardiology has grown steadily over the past few decades, as more practitioners recognize its clinical applications and value in managing cardiovascular disease. There is abroad spectrum of noninvasive and invasive testing available for the diagnosis and management of patients with cardiovascular disease. Advances in computer technology have made sophisticated devices, such as the gated SPECT, a routine part of cardiology.