Machine Learning Algorithms to Distinguish Myocardial Perfusion SPECT Polar Maps

Myocardial perfusion imaging (MPI) plays an important role in patients with suspected and documented coronary artery disease (CAD). Machine Learning (ML) algorithms have been developed for many medical applications with excellent performance. This study used ML algorithms to discern normal and abnormal gated Single Photon Emission Computed Tomography (SPECT) images. We analyzed one thousand and seven polar maps from a database of patients referred to a university hospital for clinically indicated MPI between January 2016 and December 2018. These studies were reported and evaluated by two different expert readers. The image features were extracted from a specific type of polar map segmentation based on horizontal and vertical slices. A senior expert reading was the comparator (gold standard). We used cross-validation to divide the dataset into training and testing subsets, using data augmentation in the training set, and evaluated 04 ML models. All models had accuracy >90% and area under the receiver operating characteristics curve (AUC) >0.80 except for Adaptive Boosting (AUC = 0.77), while all precision and sensitivity obtained were >96 and 92%, respectively. Random Forest had the best performance (AUC: 0.853; accuracy: 0,938; precision: 0.968; sensitivity: 0.963). ML algorithms performed very well in image classification. These models were capable of distinguishing polar maps remarkably into normal and abnormal.

Machine Learning Algorithms to Detect Sex in Myocardial Perfusion Imaging

Myocardial perfusion imaging (MPI) is an essential tool used to diagnose and manage patients with suspected or known coronary artery disease. Additionally, the General Data Protection Regulation (GDPR) represents a milestone about individuals' data security concerns. On the other hand, Machine Learning (ML) has had several applications in the most diverse knowledge areas. It is conceived as a technology with huge potential to revolutionize health care. In this context, we developed ML models to evaluate their ability to distinguish an individual's sex from MPI assessment. We used 260 polar maps (140 men/120 women) to train ML algorithms from a database of patients referred to a university hospital for clinically indicated MPI from January 2016 to December 2018. We tested 07 different ML models, namely, Classification and Regression Tree (CART), Naive Bayes (NB), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Adaptive Boosting (AB), Random Forests (RF) and, Gradient Boosting (GB). We used a cross-validation strategy. Our work demonstrated that ML algorithms could perform well in assessing the sex of patients undergoing myocardial scintigraphy exams. All the models had accuracy greater than 82%. However, only SVM achieved 90%. KNN, RF, AB, GB had, respectively, 88, 86, 85, 83%. Accuracy standard deviation was lower in KNN, AB, and RF (0.06). SVM and RF had had the best area under the receiver operating characteristic curve (0.93), followed by GB (0.92), KNN (0.91), AB, and NB (0.9). SVM and AB achieved the best precision. Our results bring some challenges regarding the autonomy of patients who wish to keep sex information confidential and certainly add greater complexity to the debate about what data should be considered sensitive to the light of the GDPR.

Artificial Intelligence in Cardiology: Concepts, Tools and Challenges - "The Horse is the One Who Runs, You Must Be the Jockey"

The recent advances at hardware level and the increasing requirement of personalization of care associated with the urgent needs of value creation for the patients has helped Artificial Intelligence (AI) to promote a significant paradigm shift in the most diverse areas of medical knowledge, particularly in Cardiology, for its ability to support decision-making and improve diagnostic and prognostic performance. In this context, the present work does a non-systematic review of the main papers published on AI in Cardiology, focusing on its main applications, potential impacts and challenges.

Association between B-type natriuretic peptide and within-visit blood pressure variability

BACKGROUND

Blood pressure variability (BPV) has been shown to predict cardiovascular events. Within-visit BPV is the simplest and easiest measure of BPV, but previous studies have shown conflicts as to whether within-visit BPV correlates with target organ damage. We aimed to evaluate whether within-visit BPV correlates with B-type natriuretic peptide (BNP) in a general population.

HYPOTHESIS

Within-visit BPV correlates with BNP in a general population.

METHODS

This was a cross-sectional study that included 633 individuals, randomly selected, age 45 to 99 years, registered in the primary care program from an urban medium-sized town. Patients were scheduled for a single-day visit that consisted of clinical evaluation and laboratory tests. Three blood pressure (BP) readings, 1 minute apart, were done, and within-visit BPV was determined as the coefficient of variation (CV) of the 3 BP measures. Our main outcome was to correlate BNP and within-visit BPV. A multivariable model was estimated using a generalized linear model to evaluate the independent effects of different variables on BNP levels.

RESULTS

The median age was 57 years. Median BNP was 16 pg/mL, and the median systolic and diastolic BP-CV were, respectively, 3.9% and 3.5%. There was a weak but positive correlation between BNP and both systolic BP-CV and diastolic BP-CV (r = 0.107 and P = 0.007 and r = 0.092 and P = 0.019, respectively). In multiple regression equation, systolic BP, diastolic BP-CV, body mass index, and estimated glomerular filtration rate were associated with BNP.

CONCLUSIONS

In the present study, there was a positive, albeit weak, correlation between within-visit BPV and BNP. In addition, diastolic BPV was associated with BNP even after adjustment for multiple confounders.

The Role of Renin-Angiotensin-Aldosterone System and Its New Components in Arterial Stiffness and Vascular Aging

Many cardiovascular diseases present renin-angiotensin-aldosterone system (RAAS) hyperactivity as an important pathophysiological mechanism to be target in the therapeutic approaches. Moreover, arterial stiffness is currently considered as a new independent risk factor for cardiovascular disease in different clinical conditions, including hypertension and chronic kidney disease. In fact, excessive stimulation of angiotensin type 1 (AT1) receptors, as well as mineralocorticoid receptors, results in cellular growth, oxidative stress and vascular inflammation, which may lead to arterial stiffness and accelerate the process of vascular aging. In the last decades, a vasoprotective axis of the RAAS has been discovered, and now it is well established that new components with antioxidant and anti-inflammatory properties play important roles promoting vasodilation, natriuresis and reducing collagen deposition, thus attenuating arterial stiffness and improving endothelial function. In this review, we will focus on these pathophysiological mechanisms and the relevance of RAAS inhibition by different strategies to increase arterial compliance and to decelerate vascular aging.

Distinct mortality profile in systemic sclerosis: a death certificate study in Rio de Janeiro, Brazil (2006-2015) using a multiple causes of death analysis

The objective of this study was to assess the mortality profile related to SSc in the state of Rio de Janeiro, Brazil. We retrospectively examined all registered deaths in the region (2006-2015 period) in which the diagnosis of SSc was mentioned on any line of the death certificates (underlying cause of death [UCD], n = 223; non-UCD, n = 151). Besides the analysis of gender, age, and the causes of death, we also compared the mortality from UCDs between individuals whose death causes included SSc (cases) and those whose death causes did not include SSc (deceased controls). For the latter comparison, we used the mortality odds ratio to approximate the cause-specific standardized mortality ratio. We identified 1495 death causes among the 374 SSc cases. The mean age at death of the SSc cases (85% women) was significantly lower than that of the controls (n = 1,294,117) (58.7 vs. 65.5 years, respectively). The main death causes were circulatory system diseases, infections, and respiratory diseases (36%, 34%, and 21% of SSc cases, respectively). Compared to the deceased controls, there were proportionally more deaths among the SSc cases from pulmonary arterial hypertension, lung fibrosis, septicemia, gastrointestinal hemorrhage, other systemic connective tissue diseases, and heart failure (for death age < 50 years). We confirmed the high burden of cardiovascular, respiratory, and infectious causes in this predominantly non-Caucasian sample of SSc patients. Of interest, the percentage of infection-related deaths in our report was about three times higher than that in SSc studies with predominantly Caucasian populations.

A Rare Case of Ascites due to Peritoneal Amyloidosis

BACKGROUND

The clinical manifestations of amyloidosis depend on the type of insoluble protein as well as the location of amyloid deposits in tissues or organs. In the gastrointestinal tract, the small intestine is the most common site of amyloid deposits, whereas peritoneal involvement and ascites are rare.

CASE REPORT

We report on a case of ascites due to peritoneal amyloidosis. A 65-year-old patient was admitted to our institution due to anasarca and pulmonary congestion, mimicking heart failure. We started the patient on diuretics and vasodilators. Despite improvement in pulmonary congestion and peripheral edema, his ascites was not reduced. Echocardiogram revealed restrictive cardiomyopathy and a speckle-tracking pattern suggestive of cardiac amyloidosis. Subcutaneous and peritoneal biopsies revealed amyloidosis.

CONCLUSIONS

Amyloidosis is rare in the peritoneum and is usually asymptomatic. Ascites occurs in only 20% of patients with peritoneal amyloidosis. We searched PubMed using "ascites" and "amyloidosis" and identified only eight case reports of amyloidosis with ascites. Physicians should be particularly careful in heart failure and anasarca cases when ascites is disproportional or not responsive to diuretic treatment. To date, there is no specific treatment for peritoneal amyloidosis.

Renin-Angiotensin System Blockade Associated with Statin Improves Endothelial Function in Diabetics

Background

Studies suggest that statins have pleiotropic effects, such as reduction in blood pressure, and improvement in endothelial function and vascular stiffness.

Objective

To analyze if prior statin use influences the effect of renin-angiotensin-aldosterone system inhibitors on blood pressure, endothelial function, and vascular stiffness.

Methods

Patients with diabetes and hypertension with office systolic blood pressure ≥ 130 mmHg and/or diastolic blood pressure ≥ 80 mmHg had their antihypertensive medications replaced by amlodipine during 6 weeks. They were then randomized to either benazepril or losartan for 12 additional weeks while continuing on amlodipine. Blood pressure (assessed with ambulatory blood pressure monitoring), endothelial function (brachial artery flow-mediated dilation), and vascular stiffness (pulse wave velocity) were evaluated before and after the combined treatment. In this study, a post hoc analysis was performed to compare patients who were or were not on statins (SU and NSU groups, respectively).

Results

The SU group presented a greater reduction in the 24-hour systolic blood pressure (from 134 to 122 mmHg, p = 0.007), and in the brachial artery flow-mediated dilation (from 6.5 to 10.9%, p = 0.003) when compared with the NSU group (from 137 to 128 mmHg, p = 0.362, and from 7.5 to 8.3%, p = 0.820). There was no statistically significant difference in pulse wave velocity (SU group: from 9.95 to 9.90 m/s, p = 0.650; NSU group: from 10.65 to 11.05 m/s, p = 0.586).

Conclusion

Combined use of statins, amlodipine, and renin-angiotensin-aldosterone system inhibitors improves the antihypertensive response and endothelial function in patients with hypertension and diabetes.

Nobel prizes: contributions to cardiology

The Nobel Prize was created by Alfred Nobel. The first prize was awarded in 1901 and Emil Adolf von Behring was the first laureate in medicine due to his research in diphtheria serum. Regarding cardiology, Nobel Prize's history permits a global comprehension of progress in pathophysiology, diagnosis and therapeutics of various cardiac diseases in last 120 years. The objective of this study was to review the major scientific discoveries contemplated by Nobel Prizes that contributed to cardiology. In addition, we also hypothesized why Carlos Chagas, one of our most important scientists, did not win the prize in two occasions. We carried out a non-systematic review of Nobel Prize winners, selecting the main studies relevant to heart diseaseamong the laureates. In the period between 1901 and 2013, 204 researches and 104 prizes were awarded in Nobel Prize, of which 16 (15%) studies were important for cardiovascular area. There were 33 (16%) laureates, and two (6%) were women. Fourteen (42%) were American, 15 (45%) Europeans and four (13%) were from other countries. There was only one winner born in Brazil, Peter Medawar, whose career was all in England. Reviewing the history of the Nobel Prize in physiology or medicine area made possible to identify which researchers and studies had contributed to advances in the diagnosis, prevention and treatment of cardiovascular diseases. Most winners were North Americans and Europeans, and male.