Machine learning for predicting the outcomes and risks of cardiovascular diseases in patients with hypertension: results of ESSE-RF in the Primorsky Krai

Aim. To assess the prospects of using artificial intelligence technologies in predicting the outcomes and risks of cardiovascular diseases (CVD) in patients with hypertension (HTN).Material and methods. A software application was created for data mining from respondent profiles in a semi-automatic mode; libraries with data preprocessing were analyzed. We analyzed the main and additional parameters (35) of CVD risk factors in 2131 people as a part of ESSE-RF study (2014-2019). To create a forecasting model, a high-level language Python 2.7 was used using object-oriented programming and exception handling with multithreading support. Using randomization, learning (n=488) and test (n=245) samples were formed, which included data from patients with an established diagnosis of HTN.Results. The prevalence of HTN among subjects was 34,39%. There were following significant factors for predicting CVD: anthropometric parameters, smoking, biochemical profile (total cholesterol, ApoA, ApoB, glucose, D-dimer, C-reactive protein). As a result of a 5-year follow-up, CVD was found in 235 people (32,06%) with HTN and 187 people (13,38%) without HTN; mortality rates were 1,27% in subjects with HTN and 1,12% — without HTN. The absolute mortality risk among participants with HTN (0,037) was significantly higher (p<0,05) than in patients without HTN (0,017). To create a neural network (NN), the basic Sequential model from the Keras library was used. During machine learning, 26 variables important for the CVD development were used as input and 9 neurons — as output, which corresponded to the number of established cardiovascular events. The created NN had a predictive value of up to 97,9%, which exceeded the SCORE value (34,9%).Conclusion. The data obtained indicate the importance of risk factor phenotyping using anthropometric markers and biochemical profile for determining their significance in the top 20 predictors of CVD. The Python-based machine learning provides CVD prediction according to standard risk assessments.

Download Full-text

Random Forests Highlight the Combined Effect of Environmental Heavy Metals Exposure and Genetic Damages for Cardiovascular Diseases

Applied Sciences ◽

10.3390/app11188405 ◽

2021 ◽

Vol 11 (18) ◽

pp. 8405

Author(s):

Alfonso Monaco ◽

Antonio Lacalamita ◽

Nicola Amoroso ◽

Armando D’Orta ◽

Andrea Del Buono ◽

...

Keyword(s):

Machine Learning ◽

Heavy Metals ◽

Cardiovascular Diseases ◽

Random Forests ◽

Classification Problem ◽

Classification Performance ◽

Cvd Risk ◽

Learning Framework ◽

Clinic Foundation ◽

Highly Correlated

Heavy metals are a dangerous source of pollution due to their toxicity, permanence in the environment and chemical nature. It is well known that long-term exposure to heavy metals is related to several chronic degenerative diseases (cardiovascular diseases, neoplasms, neurodegenerative syndromes, etc.). In this work, we propose a machine learning framework to evaluate the severity of cardiovascular diseases (CVD) from Human scalp hair analysis (HSHA) tests and genetic analysis and identify a small group of these clinical features mostly associated with the CVD risk. Using a private dataset provided by the DD Clinic foundation in Caserta, Italy, we cross-validated the classification performance of a Random Forests model with 90 subjects affected by CVD. The proposed model reached an AUC of 0.78 ± 0.01 on a three class classification problem. The robustness of the predictions was assessed by comparison with different cross-validation schemes and two state-of-the-art classifiers, such as Artificial Neural Network and General Linear Model. Thus, is the first work that studies, through a machine learning approach, the tight link between CVD severity, heavy metal concentrations and SNPs. Then, the selected features appear highly correlated with the CVD phenotype, and they could represent targets for future CVD therapies.

Download Full-text

Genome-wide DNA methylation analysis on C-reactive protein among Ghanaians suggests molecular links to the emerging risk of cardiovascular diseases

npj Genomic Medicine ◽

10.1038/s41525-021-00213-9 ◽

2021 ◽

Vol 6 (1) ◽

Author(s):

Felix P. Chilunga ◽

Peter Henneman ◽

Andrea Venema ◽

Karlijn A. C. Meeks ◽

Ana Requena-Méndez ◽

...

Keyword(s):

Cardiovascular Diseases ◽

Molecular Mechanisms ◽

Linear Models ◽

C Reactive Protein ◽

Cvd Risk ◽

Protein Marker ◽

Reactive Protein ◽

Emerging Risk ◽

Post Hoc ◽

Rodam Study

AbstractMolecular mechanisms at the intersection of inflammation and cardiovascular diseases (CVD) among Africans are still unknown. We performed an epigenome-wide association study to identify loci associated with serum C-reactive protein (marker of inflammation) among Ghanaians and further assessed whether differentially methylated positions (DMPs) were linked to CVD in previous reports, or to estimated CVD risk in the same population. We used the Illumina Infinium® HumanMethylation450 BeadChip to obtain DNAm profiles of blood samples in 589 Ghanaians from the RODAM study (without acute infections, not taking anti-inflammatory medications, CRP levels < 40 mg/L). We then used linear models to identify DMPs associated with CRP concentrations. Post-hoc, we evaluated associations of identified DMPs with elevated CVD risk estimated via ASCVD risk score. We also performed subset analyses at CRP levels ≤10 mg/L and replication analyses on candidate probes. Finally, we assessed for biological relevance of our findings in public databases. We subsequently identified 14 novel DMPs associated with CRP. In post-hoc evaluations, we found that DMPs in PC, BTG4 and PADI1 showed trends of associations with estimated CVD risk, we identified a separate DMP in MORC2 that was associated with CRP levels ≤10 mg/L, and we successfully replicated 65 (24%) of previously reported DMPs. All DMPs with gene annotations (13) were biologically linked to inflammation or CVD traits. We have identified epigenetic loci that may play a role in the intersection between inflammation and CVD among Ghanaians. Further studies among other Africans are needed to confirm our findings.

Download Full-text

Python the game changer in the field of Machine Learning, Data Science and IoT: A Review

International Journal for Research in Applied Science and Engineering Technology ◽

10.22214/ijraset.2021.37668 ◽

2021 ◽

Vol 9 (8) ◽

pp. 1827-1837

Author(s):

Prithwish Parial

Keyword(s):

Machine Learning ◽

Data Science ◽

Learning Experience ◽

Object Oriented Programming ◽

Machine Learning Techniques ◽

Learning Techniques ◽

High Level ◽

Learning Data ◽

Game Changer ◽

Oriented Programming Language

Abstract: Python is the finest, easily adoptable object-oriented programming language developed by Guido van Rossum, and first released on February 20, 1991 It is a powerful high-level language in the recent software world. In this paper, our discussion will be an introduction to the various Python tools applicable for Machine learning techniques, Data Science and IoT. Then describe the packages that are in demand of Data science and Machine learning communities, for example- Pandas, SciPy, TensorFlow, Theano, Matplotlib, etc. After that, we will move to show the significance of python for building IoT applications. We will share different codes throughout an example. To assistance, the learning experience, execute the following examples contained in this paper interactively using the Jupiter notebooks. Keywords: Machine learning, Real world programming, Data Science, IOT, Tools, Different packages, Languages- Python.

Download Full-text

Possible Approaches to Primary Prevention of Cardiovascular Diseases

Rational Pharmacotherapy in Cardiology ◽

10.20996/1819-6446-2020-10-12 ◽

2020 ◽

Vol 16 (5) ◽

pp. 831-841

Author(s):

V. N. Larina ◽

D. S. Mkrtychev ◽

V. A. Kuznetsova ◽

A. A. Tyazhelnikov

Keyword(s):

Cardiovascular Diseases ◽

Primary Prevention ◽

Heart Association ◽

Positive Trend ◽

Loss Assessment ◽

Cvd Risk ◽

Factors Affecting ◽

Glucose Levels ◽

Cholesterol Levels ◽

High Level

In recent years there is a positive trend in the development of preventive medicine, in particular, primary prevention of diseases. However, in most cases, patients seek help from a doctor after the manifestation of the disease, and therefore, early identification of risk factors (RF) remains relevant. Conduction of a large number of studies that are aimed at studying modifiable RF associated with the development of cardiovascular diseases (CVD), allowed the experts of the American Heart Association to develop recommendations “Life's Simple 7”, which makes it possible to structure methods of primary prevention of CVD and minimize the risk of their development. In 2019, experts from the American College of Cardiology presented a simplified version of these recommendations, to improve approaches to primary prevention and their effectiveness not only for doctors but also for patients. Thus, by involving the patient in the decision-making process about follow-up treatment, doctors can achieve a high level of compliance, which is essential for improving the prognosis. The “ABCDE” recommendations, in name of which are reflected the first letters of the leading CVD RF, include such paragraphs as RF assessment, the use of antiaggregating therapy, correction of blood pressure, cholesterol levels, smoking elimination, correction of high glucose levels and diabetes treatment, weight loss, assessment of social and economic factors affecting the morbidity in a particular patient. Despite the undoubted benefit of the “ABCDE” recommendations, some problems of primary prevention currently cannot be solved: the inability to accurately assess social and economic RF; the imperfection of the used CVD risk scales. The updated version of the recommendations allows not only to assess the existing RF of the patient, but also to effectively correct them. In addition, the patient himself can read the recommendations, which improves understanding of the primary prevention importance.

Download Full-text

Data search computing based on the similarity-dıfference metric

SCIENTIFIC WORK ◽

10.36719/2663-4619/60/46-60 ◽

2020 ◽

Vol 60 (11) ◽

pp. 46-60

Author(s):

Vugar Hajimahmud Abdullayev ◽

Keyword(s):

Machine Learning ◽

Social Computing ◽

Data Retrieval ◽

Critical Systems ◽

Primitive Elements ◽

Software Application ◽

Data Search ◽

Common Interests ◽

Social Critical ◽

High Level

Models, methods and algorithms for cyber-social computing and machine learning implies the use of the metric of similarity – difference of unitary coded information for processing big data in order to generate adequate actuator signals for controlling cyber-social critical systems. A set-theoretic method of data search is being developed based on the similarity – difference of the frequency parameters of primitive elements, which makes it possible to determine the similarity of objects, the strategy of transforming one object into another, and also to identify the level of common interests, conflicts. Computational architectures of cyber-social computing and metric search for key data are being created. The definitions of the fundamental concepts in the field of computing are given on the basis of metric relations between interacting processes and phenomena. A software application is proposed for calculating the similarity-differences of objects based on the formation of vectors of frequencies of two sets of primitive data. A high level of correlation of the application results with the well-known system for determining plagiarism is shown. Key words: computing, cybersocial computing, decision making, unitary data codes, similarity – difference, data retrieval, plagiarism

Download Full-text

Anthropometric risk factors of cardiovascular disease in adult population of Dharan Municipality, Eastern Nepal: Gender and age differences

Journal of Morphological Sciences ◽

10.4322/jms.111717 ◽

2017 ◽

Vol 34 (04) ◽

pp. 236-240

Author(s):

S. Shah ◽

S. Koirala ◽

L. Khanal ◽

B. Koirala

Keyword(s):

Risk Factors ◽

Cardiovascular Diseases ◽

Age Differences ◽

Statistical Significance ◽

Adult Population ◽

P Value ◽

Anthropometric Parameters ◽

Cvd Risk ◽

Gender And Age ◽

Gender And Age Differences

Abstract Introduction: Cardiovascular diseases (CVD) besides cancer are the most serious threat to the health and life of the population of both developed and developing countries. The aim of the study was to know the gender and age differences with anthropometric CVD risk factors among Nepalese adults of Dharan Municipality. Materials and Methods: A population based cross-sectional study was conducted using a pretested self-administered structured questionnaire on anthropometric parameters which can affect CVD. A systematic random sampling technique was applied to cover the estimated 280 households with 900 adult population. The parameters of anthropometric risk factors for cardiovascular diseases were Body Mass Index (BMI), Waist Hip Ratio (WHR) and Body Fat Percentage (BFP). The result was expressed as mean ± SD. Independent student t test and ANOVA were applied to find out the gender and age differences respectively. “P” value of < 0.05 was considered to indicate statistical significance. Results: The mean and SD of BMI, WHR and BFP were found to be 24.17±4.13, 1.01±2.83 and 26.91±7.15 respectively. The gender differences of BFP were found to be statistically significant, whereas for BMI and WHR were not significant (p>0.05). The age differences in BMI and BFP were statistically significant (<0.05) whereas for WHR was not found to be significant (p>0.05). The signiicant positive correlations were found among these parameters (p< 0.01). Conclusion:The results of this study emphasize the need for a comprehensive study (both lipid and anthropometric) for providing baseline data to prevent CVD in eastern Nepal.

Download Full-text

PNPLA3 genotype and fibrosis-4 index predict cardiovascular diseases of Japanese patients with histopathologically-confirmed NAFLD

BMC Gastroenterology ◽

10.1186/s12876-021-02020-z ◽

2021 ◽

Vol 21 (1) ◽

Author(s):

Norio Akuta ◽

Yusuke Kawamura ◽

Yasuji Arase ◽

Satoshi Saitoh ◽

Shunichiro Fujiyama ◽

...

Keyword(s):

Risk Factors ◽

Cardiovascular Diseases ◽

Liver Cancer ◽

Japanese Patients ◽

Incidence Rates ◽

Cvd Risk ◽

Alcoholic Fatty Liver ◽

Stage 4 ◽

High Level

Abstract Background Reliable noninvasive predictors of the top three causes of death [cardiovascular diseases (CVDs), malignancies, and liver-related events in patients with non-alcoholic fatty liver disease (NAFLD)] have not yet been determined. Methods We retrospectively investigated the incidence of three complications [CVDs, malignancy (except for liver cancer), and liver-related events] in 477 Japanese patients with histo-pathologically confirmed NAFLD for a median follow-up of 5.9 years. In addition to histological findings, we also investigated noninvasive predictors. Results A score of ≥ 2.67 for the noninvasive diagnosis of stage 4 fibrosis based on the Fibrosis-4 (FIB-4) index indicated a high level area under the receiver operating characteristic (AUROC) curve (0.90), sensitivity (82.9%), specificity (86.4%), and negative predictive value [(NPV) of 98.5%]. The yearly incidence rates of CVDs, malignancies, and liver-related events were found to be 1.04%, 0.83%, and 0.30%, respectively. Multivariate analysis identified a FIB-4 index ≥ 2.67 score as a significant and independent, noninvasive predictor of these three complications. Furthermore, the cumulative incidence rates of CVDs were significantly different among the three genotypes of PNPLA3. PNPLA3 genotype CC, chronic kidney disease (CKD), and FIB-4 index ≥ 2.67 was could be attributed to these three significant CVD risk factors. The rates of CVDs were significantly different among the three subgroups based on the combination of risk factors. In malignancy (except for liver cancer), the incidence rate of colon cancer was 25.0%; in particular, the rate in females was 53.8%. Conclusions Our results highlighted the importance of the PNPLA3 genotype and FIB-4 index ≥ 2.67 on the incidence of complications in Japanese patients with NAFLD, especially the incidence of CVDs. Early diagnosis, based on the presence of one or more risk factors, and early treatment might improve the prognosis for NAFLD patients.

Download Full-text

RISK FACTORS OF CARDIOVASCULAR DISEASES IN TEACHING STAFF AMONG PRIMARY SCHOOLS OF TOMSK CITY (THE DATA FROM MULTICENTER PROSPECTIVE PREVENTION STUDY)

CARDIOVASCULAR THERAPY AND PREVENTION ◽

10.15829/1728-8800-2015-2-68-73 ◽

2015 ◽

Vol 14 (2) ◽

pp. 68-73

Author(s):

N. S. Karamnova ◽

V. N. Serebryakova ◽

I. N. Trubacheva ◽

V. S. Kaveshnikov ◽

V. A. Vygodin ◽

...

Keyword(s):

Risk Factors ◽

Cardiovascular Diseases ◽

General Education ◽

Primary Schools ◽

Effective Control ◽

Teaching Staff ◽

Cvd Risk ◽

High Prevalence ◽

Insufficient Physical Activity ◽

High Level

Aim. To study the prevalence of risk factors (RF) of cardiovascular diseases (CVD) in the teaching staff of primary schools in Tomsk city. Material and methods. Totally 154 teachers studied (staff of 2 general education schools of Tomsk) at the age of 21-71 y. Mean age 46,6±10,9. Response — 84%. Results. Prevalence of arterial hypertension (AH) among the staff was 51,9%, and primary onset of AH — in 3,2%. Less than a half of teachers with AH took antihypertension therapy (42,8%) and just in 15% of those we found target levels of blood pressure during business day. Prevalence of smoking was 7,0%. We found a high prevalence of alimentary CVD risk factors. Two thirds of the staff were overweight (68,8%). Prevalence of excessive bodyweight was 33,1% and of obesity — 35,7%, of abdominal obesity — 40,2%. In teaching staff there was high level of hyperglycemia (12,1%), hypercholesterolemia (78,0%) and hypertriglyceridemia (32,0%). Insufficient physical activity was noted in 30% of teachers. Conclusion. The main points are found for the development and implementation of preventive strategies in the group — effective control over AH and correction of alimentary-dependent RF of CVD.

Download Full-text

BAND NN: A Deep Learning Framework For Energy Prediction and Geometry Optimization of Organic Small Molecules

10.26434/chemrxiv.9763094 ◽

2019 ◽

Author(s):

Siddhartha Laghuvarapu ◽

Yashaswi Pathak ◽

U. Deva Priyakumar

Keyword(s):

Machine Learning ◽

Density Functional ◽

Computational Cost ◽

Geometry Optimization ◽

Dft Methods ◽

Energy Prediction ◽

Machine Learning Model ◽

Equilibrium Structures ◽

High Level ◽

Non Equilibrium

Recent advances in artificial intelligence along with development of large datasets of energies calculated using quantum mechanical (QM)/density functional theory (DFT) methods have enabled prediction of accurate molecular energies at reasonably low computational cost. However, machine learning models that have been reported so far requires the atomic positions obtained from geometry optimizations using high level QM/DFT methods as input in order to predict the energies, and do not allow for geometry optimization. In this paper, a transferable and molecule-size independent machine learning model (BAND NN) based on a chemically intuitive representation inspired by molecular mechanics force fields is presented. The model predicts the atomization energies of equilibrium and non-equilibrium structures as sum of energy contributions from bonds (B), angles (A), nonbonds (N) and dihedrals (D) at remarkable accuracy. The robustness of the proposed model is further validated by calculations that span over the conformational, configurational and reaction space. The transferability of this model on systems larger than the ones in the dataset is demonstrated by performing calculations on select large molecules. Importantly, employing the BAND NN model, it is possible to perform geometry optimizations starting from non-equilibrium structures along with predicting their energies.

Download Full-text

An Association Between C-reactive Protein and Cardiovascular Diseases

Medical Journal of Western Black Sea ◽

10.29058/mjwbs.2017.2.5 ◽

2017 ◽

Vol 1 (2) ◽

pp. 66-75

Author(s):

Bilgehan DOĞRU ◽

Ayşe Ceylan HAMAMCIOĞLU ◽

Tuğçe YEŞİLTAŞ

Keyword(s):

Cardiovascular Diseases ◽

C Reactive Protein ◽

Reactive Protein

Download Full-text