Coronary artery disease prediction using hidden Markov model based support vector machine

Periodontal disease, smoking, cardiovascular complications and mortality in type 1 diabetes

Introduction Type 2 diabetes mellitus is a major public health problem. Coronary artery disease (CAD) is the major cause of morbidity and mortality due to diabetes. A subset of these patients develops this complication relatively early. Identifying such patients at increased risk of CAD using an appropriate marker is important in prevention. Insulin resistance plays an important role in the development and progression of atherosclerotic CAD. Measurement of serum C-peptide may helpin identifying those with increased insulin resistance characterized by higher levels. This study was conducted to find the association between fasting C-peptide (FCP) levels and CAD in middle-aged patients with type 2 diabetes mellitus. Materials and methods This cross-sectional analytical study was done at the Medicine Outpatient Departmentof Government Medical College, Kannur, Kerala, South India. Type 2 diabetic patients aged 41-60 years were included. Patients on insulin, chronic liver disease, malignancy, critical illness,and an estimated glomerular filtration rate (eGFR) less than 45 ml/min/1.73 m2 were excluded. After a detailed clinical evaluation, blood samples were collected for FCP and other biochemicaltests. A chemiluminescent immunoassay was used for measuring serum C-peptide levels. An independent t-test compared mean FCP levels between those with and without CAD, while binary logistic regression identified CAD risk factors and strength of association. The receiver operating characteristic (ROC) curve determined the cutoff value for predicting CAD. Results Among the 229 patients included, 50 (21.83%) had evidence of CAD. On an independent t-test, patients with CAD had significantly higher serum FCP levels (2.36 ± 0.80 vs. 1.86 ± 0.58 ng/ml, p < 0.001) than those without CAD. On binary logistic regression analysis, the association of serum FCP levels with CAD was found to be statistically significant (p = 0.001), with odds of CAD increasing by 2.98 times for each unit increase in FCP levels. On the ROC curve, FCP was found to have a good predictive ability with an area under curve (AUC) of 0.71 (0.61-0.80); using a cutoff value of FCP ≥ 2.13 ng/ml, the sensitivity, specificity, and accuracy for predicting CAD were 70%, 69%, and 69.2%, respectively.When patients with CAD were categorized based on quartiles of FCP levels, the majority of patients with CAD were in upper quartiles (Q3 and Q4) with higher serum C-peptide levels, further indicating an association between elevated C-peptide levels and CAD. Conclusion In this study, a significant positive association was observed between serum FCP levels and CAD in type 2 diabetic patients aged 41-60 years. The odds of CAD increased nearly threefold for each unit rise in serum FCP levels. Patients with a higher risk of CAD can be identified by using an optimal cutoff value of serum FCP. These findings underscore the potential role of FCP in early identification of patients at increased risk for CAD.

Background:Depression is an established predictor of coronary artery disease (CAD) progression and mortality. “Somatic” symptoms of depression such as fatigue and sleep impairment overlap with symptoms of CAD and independently predict CAD events. Differentiating between “somatic” and “cognitive” depressive symptoms in at-risk patients may improve our understanding of the relationship between depression and CAD.Methods:The study utilized data from the Women’s Ischemia Syndrome Evaluation. Participants (N = 641; mean age = 58.0 [11.4] years) were enrolled to evaluate chest pain or suspected myocardial ischemia. They completed a battery of symptom and psychological questionnaires (including the Beck Depression Inventory [BDI]) at baseline, along with quantitative coronary angiography and other CAD diagnostic procedures. The BDI provided scores for total depression and for cognitive and somatic depressive symptom subscales.Results:Two hundred and fourteen (33.4%) women met criteria for obstructive CAD. Logistic regression models were used to examine relationships between depression symptoms and obstructive CAD. Neither BDI total scores (odds ratio [OR] =1.02, 95% confidence interval [CI], 0.99–1.05, P = 0.053) nor BDI cognitive scores (OR = 1.02, 95% CI, 1.00–1.04, P = 0.15) predicted CAD status. BDI somatic symptom scores, however, significantly predicted CAD status and remained statistically significant after controlling for age, race, and education (OR = 1.06, 95% CI, 1.01–1.12, P = 0.02).Conclusion:Among women with suspected myocardial ischemia, somatic but not cognitive depressive symptoms predicted an increased risk of obstructive CAD determined by coronary angiography. Consistent with prior reports, these results suggest a focus on somatic rather than cognitive depressive symptoms could offer additional diagnostic information.

In patients with uncomplicated diabetes, there is low probability of finding significant coronary artery disease (CAD) by noninvasive tests. Therefore, screening for its presence is not justified, and it is important to find reliable predictors of silent CAD to identify patients with uncomplicated diabetes for further screening. The relationship between lipoprotein(a) [Lp(a)], apolipoprotein(a) [apo(a)] polymorphism, and silent CAD has never been studied. We investigated the association of Lp(a) and apo(a) polymorphism with angiographically documented asymptomatic CAD in type 2 diabetic patients without evident complications. A total of 1,323 diabetic patients without any clinical and electrocardiographic evidence of CAD were evaluated. Of 121 patients with highly positive results of exercise electrocardiography (ECG) (n = 30) or positive results on exercise thallium scintigraphy (n = 91), 103 subjects showed angiographically documented CAD (CAD group). Of 1,106 patients with negative results on exercise ECG, 103 subjects without CAD (NO CAD group) were selected and matched by age, gender, and duration of diabetes to patients in the CAD group. In patients in the NO CAD group, results of exercise ECG, 48-h ambulatory ECG, and stress echocardiography were negative for CAD. The CAD group had higher Lp(a) levels (21.7 +/- 17.7 vs. 15.2 +/- 19.0 mg/dl; P = 0.0093) than the NO CAD group, and a percentage of subjects had at least one small apo(a) isoform (68.9 vs. 29.1%; P = 0.0000) higher than the NO CAD group. Logistic regression analysis showed that apo(a) phenotypes (odds ratio [OR] 8.13, 95% CI 3.65-21.23), microalbuminuria (5.38, 2.44-11.88), smoking (2.72, 1.31-5.64), and Lp(a) levels (2.41, 1.15-5.03) were predictors of asymptomatic CAD. Our investigation reports the first evidence of an independent association of Lp(a) and apo(a) polymorphism with asymptomatic CAD. This suggests that Lp(a) levels and apo(a) phenotypes could be used together with other risk factors as markers of asymptomatic CAD in patients with diabetes.

The incidence of nephropathy, a major predictor of coronary artery disease (CAD) in type 1 diabetes (T1D), has been strongly declining over the past several years. However, CAD rates do not appear to be declining commensurately. To identify similarities and differences in the CAD risk factor profile in subjects with and without microalbuminuria or worse (MA+), 20 year follow up data from the Pittsburgh Epidemiology of Diabetes Complications (EDC) study of childhood onset T1D (mean baseline age 28 and duration 19 yrs) were examined. MA+ was determined by kidney disease-related death, dialysis, transplant, or AER&gt;20 μg/mL in 2 of 3 timed urine collections. CAD was defined as angina, ischemic ECG changes, revascularization, confirmed MI, or CAD death. Cox proportional hazard models, adjusted for all the significant univariate risk factors, were determined separately for subjects who developed MA+ before their incident CAD event and for those who had CAD without preceding MA (MA-). As we have previously shown that CAD risk factors to vary by sex, we performed sex specific analyses within each MA group to determine if the predictors of CAD differ by sex and MA status. The results of the sex-specific analysis are listed in the table below. MA- MA+ Men (n=15 cases/84 β (CI) Men (n=65 cases/165) β (CI) nonHDL (mg/dl) 1.02 (1.0-1.04) Diabetes Duration (years) 1.12 (1.1-1.2) WHR 3.05 (.98-9.5) AER (log) 1.34 (1.2-1.5) Women (n=20 cases/92) Women (n=56 cases/172) Diabetes Duration (years) 1.11 (1.0-1.2) Diabetes Duration (years) 1.11 (1.07-1.15) Pulse Rate (bpm) 1.06 (1.0-1.1) Hypertension 1.85 (1.0-3.2) nonHDL (mg/dl) 1.01 (1.0-1.01) In the MA+ group, diabetes duration predicted CAD in both sexes, with AER (men) and hypertension and nonHDL (women) also contributing. However, in the MA- group, duration was only predictive in women, where pulse rate (p=0.016) was also predictive, while in men nonHDL, and to a small extent WHR (p=.054) predicted CAD. In conclusion, these data suggest that further factors beyond standard risk factors and renal disease are operative in T1D, particularly pulse rate in women.

Prior studies examining coronary atherosclerosis in the young have been limited by retrospective analyses in small cohorts. We examined the relationship between cardiovascular risk factors (RFs) and prevalence and severity of coronary atherosclerosis in a large, prospective, multinational registry of consecutive young individuals undergoing coronary computerized tomographic angiography (CCTA). Of 27 125 patients undergoing CCTA, 1635 young (<45 years) individuals without known coronary artery disease (CAD) or coronary anomalies were identified. Coronary plaque was assessed for any CAD, obstructive CAD (≥50% stenosis), and presence of calcified plaque (CP) and non-calcified plaque (NCP). Among 1635 subjects (70% men, age 38 ± 6 years), any CAD, obstructive CAD, CP, and NCP were observed in 19, 4, 5, and 8%, respectively. Compared with women, men demonstrated higher rates of any CAD (21 vs. 12%, P < 0.001), CP (6 vs. 3%, P = 0.01), and NCP (9 vs. 5%, P = 0.008), although no difference was observed for rates of obstructive CAD (5 vs. 4%, P = 0.46). Any CAD, obstructive CAD, and NCP were higher for young individuals with diabetes, hypertension, dyslipidaemia, current smoking, or family history of CAD; while only diabetes and dyslipidaemia were associated with CP. Increasing cardiovascular RFs was associated with a greater prevalence and extent and severity of CAD, with individuals with 0, 1, 2, ≥3 RFs manifesting a dose-response increase in any CAD (P < 0.001, for trend), obstructive CAD (P < 0.001, for trend), NCP (P < 0.001, for trend), and CP (P < 0.001, for trend). In multivariable analysis adjusting for sex and cardiovascular RFs, male sex was the strongest predictor for any CAD (odds ratio [OR] = 1.95, 95% confidence interval [CI] = 1.43-2.66, P < 0.001), CP (OR = 1.46, 95% CI = 1.08-1.98, P = 0.01), and NCP (OR = 1.33, 95% CI = 1.06-1.67, P = 0.01); family history of CAD was the strongest predictor for obstructive CAD (OR = 2.71, 95% CI = 1.65-4.45, P < 0.001). Any and obstructive CAD is present in 1 in 5 and 1 in 20 young individuals, respectively, with family history associated with the greatest risk of obstructive CAD.

Background Incidence of coronary artery disease (CAD) increases significantly in postmenopausal women, which is assumed to be an imbalance between serum androgen and estrogen levels. However, studies assessing serum sex hormones and CAD are few and have shown conflicting results.Objective To compare serum sex hormone levels and traditional risk factors among postmenopausal women with angiographically proven CAD and without CAD.Method The study included consecutive postmenopausal women undergoing coronary angiography in our institute from May 2016 to June 2017. The clinical and coronary angiographic data and traditional risk factors were assessed. Fasting serum levels of estradiol (E2), testosterone (T), sex hormone-binding globulin (SHBG), dehydroepiandrosterone sulfate (DHEA-S), and insulin were measured.Results A total of 121 postmenopausal women were included in the study; 69 were CAD and 52 without CAD. Single-vessel disease was most common (55.1%), followed by double-vessel disease (24.6%) and triple-vessel disease (20.3%). Women with CAD had significantly lower estradiol/testosterone (E2/T) ratio (3.7 ± 2.6 vs. 5.4 ± 4.2,p = 0.008) compared with non-CAD group. SHBG, DHEA-S, and insulin levels were similar in CAD and non-CAD groups. The serum level of estradiol predicted the E2/T ratio (r = 0.316,p &lt; 0.001) and positively associated with DHEA (r = 0.181,p = 0.047). Testosterone was negatively associated with E2/T ratio (r = – 0.682,p &lt; 0.001). There was no significant correlation of estrogen, testosterone, or E2/T ratio to lipid profile (total cholesterol, HDL, LDL) in women with CAD.Conclusion E2/T ratio was significantly lowered in postmenopausal women with CAD. E2/T ratio may be a used a predictor of CAD in postmenopausal women

Aims: Skin autofluorescence (SAF), a non-invasive measurement of the accumulation for glycation end products (AGEs) level of skin, has been reported as a useful marker for micro/macro vascular complications. However there is little data evaluating whether SAF can be the predictor of the patients with suspected a coronary artery disease (CAD) and indicating the coronary angiogram (CAG). The aim of this study was to evaluate association between SAF and CAD detected by CAG in the catheterization laboratory. Methods: From October/2013-March/2014, 333 patients with suspected the CAD and underwent CAG were enrolled prospectively. SAF were assessed with autofluorescence reader before CAG. Patients with the history of percutaneous coronary intervention, coronary artery bypass grafting were excluded. CAD was defined as if there was at least one coronary artery stenosis greater than or equal to 50% (AHA classification) by CAG. Association between SAF and existence of CAD was evaluated, and we also explored the relationship between SAF and number of diseased coronary artery. Results: Of the 333 patients, CAD was observed in 145 (43.5%) patients. In multivariate logistic regression analysis, age (odds ratio [OR]: 1.03; 95% con[[Unable to Display Character: &amp;#64257;]]dence interval [CI]: 1.00 to 1.06; p = 0.049), gender (male as reference) (odds ratio [OR]: 0.46; 95% con[[Unable to Display Character: &amp;#64257;]]dence interval [CI]: 0.28 to 0.75; p = 0.002), HbA1c (odds ratio [OR]: 1.45; 95% con[[Unable to Display Character: &amp;#64257;]]dence interval [CI]: 1.10 to 1.91; p = 0.008) and SAF (odds ratio [OR]: 4.85; 95% con[[Unable to Display Character: &amp;#64257;]]dence interval [CI]: 2.750 to 8.55; p &lt;0.001) were considered to be the predictors of CAD. Furthermore patients with higher SAF showed more number of coronary artery disease (Figure 1). When we use the response variable greater equal to 2.3 to determine CAD, sensitivity is 0.655 (0.572, 0.732) and specificity is 0.713 (0.642, 0.776). Conclusions: SAF was independent predictor of CAD and showed a possibility of being the surrogate for the CAD as non-invasive methodology.

INTRODUCTION: Cardiovascular diseases, particularly coronary artery disease (CAD), present a global health challenge, necessitating effective detection and diagnosis methods for early intervention. Various machine learning and deep learning approaches have emerged, utilizing diverse data sources such as electrocardiogram (ECG) signals and clinical features to enhance CAD detection. Additionally, circadian heart rate variability (HRV) has been explored as a potential diagnostic marker for CAD severity. This research aims to contribute to the burgeoning field of medical AI and its application in cardiology.OBJECTIVES: This study seeks to develop a Comprehensive Coronary Artery Disease Detection System integrating real-time heart rate monitoring and CAD prediction via an Android application. The objectives include seamless data transmission, efficient cloud-based data management, and the utilization of AI models, including ANNs, CNNs for ECG images, and hybrid models combining clinical and ECG data, to improve early CAD detection and management.METHODS: The system architecture involves integrating key sensors, an Arduino microcontroller, a Bluetooth module, and AI models to facilitate early CAD detection. An Android application complements the system, offering heart rate monitoring and CAD prediction using various data sources. Cloud computing is employed for efficient data management and analysis.RESULTS: The developed system successfully integrates cutting-edge technology to enhance CAD detection, achieving accurate and efficient results in real-time heart rate monitoring and CAD prediction.CONCLUSION: The Comprehensive Coronary Artery Disease Detection System, leveraging AI and cloud computing, contributes to proactive health monitoring and informed decision-making in CAD management and prevention, thereby addressing a critical need in cardiovascular health care.

Elevated oxidative stress among coronary artery disease patients on statin therapy: A cross sectional study

1015-75 Detection of Aortic Plaques is not Predictive of Coronary Artery Disease In the Elderly: A Biplane Transesophageal Echocardiographic Study

We have previously demonstrated associations between the urinary proteome profile and coronary artery disease (CAD) in cross-sectional studies. Here we evaluated the potential of urinary proteomics as a predictor of CAD in the Hypertension Associated Cardiovascular Disease (HACVD) sub-study population of the ASCOT study. Thirty-seven cases with the primary endpoint CAD (fatal CAD, non-fatal myocardial infarction and coronary revascularisation) but without established cardiovascular disease at baseline were selected and matched for sex and age within ±2 years to controls who had not reached a CAD endpoint during the study (median observation time, 5 years). A spot urine sample collected at 1 to 1.5 years post randomisation was analysed using capillary electrophoresis (CE) on-line coupled to Micro-TOF mass spectrometry (MS). A previously developed 238-marker CE-MS model for diagnosis of CAD (CAD 238 ) was assessed for its predictive potential. Sixty urine samples (32 cases; 28 controls; 88% male, mean age 64±5 years) passed quality control for proteomic analysis. There was a trend towards lower ("healthier") values in controls for the CAD model classifier (-0.432±0.326 vs -0.587±0.297, P =0.062), and the CAD model showed statistical significance on Kaplan-Meier survival analysis (Log Rank (Mantel-Cox) P =0.021). After unblinding we found 190 individual markers out of 1501 urinary peptides that separated cases and controls with an AUC&gt;0.6. Of these, 28 peptides including fragments of PGRC1, PTGDS, CO1A1, CO3A1, COGA1, PXDC2, B3GT6 and RET4 were also components of CAD 238 . A urinary proteome panel that was originally developed in a cross-sectional study predicts CAD endpoints independent of age and sex in a well controlled prospective study. Proteomic analysis may have the potential to detect subclinical cardiovascular damage that is associated with increased cardiovascular risk.

The precise clinical utility of lung 201Tl uptake in exercise SPECT myocardial perfusion imaging remains open to research. This study validates an optimal index for lung 201Tl uptake measurement and assesses its value in the prediction of higher-risk coronary artery disease (CAD). Three hundred and ninety-eight patients underwent exercise SPECT myocardial perfusion imaging. They were separated into derivation (n = 217) and validation (n = 186) groups, both including sub-populations of lower- and higher-risk CAD, according to coronary angiography. Another 56 individuals with a low probability of CAD comprised the control group. From a planar, anterior, post-exercise acquisition, the lung (L) to heart (H) maximal (L/H(max)), total (L/H(mean)) and background-subtracted total (L/H(net)) ratios were calculated. These were also adjusted for confounding variables, as assessed from the control group. ROC analysis was used to compare the L/H ratios and define thresholds of abnormality. The performance of the optimal index was assessed in the derivation group and was then tested in the validation population. Subsequently, it was compared with other scintigraphic, exercise electrocardiography and clinical variables. In the derivation group L/H(net) was a better discriminator for higher-risk CAD than both L/H(max) and L/H(mean). Similarly, the adjusted L/H(net) was a better discriminator than both the adjusted L/H(max) and the adjusted L/H(mean). No significant difference was attained between L/H(net) and the adjusted L/H(net). At the upper defined threshold of abnormality, sensitivity and specificity of L/H(net) in the detection of higher-risk CAD in the derivation and the validation cohorts were 52% and 92% versus 47% and 94%, respectively (p = ns). The results were similar at other defined thresholds. Moreover, L/H(net) was found to be a significant predictor of higher-risk CAD, superior to myocardial perfusion images, transient ventricular dilation measurements, and clinical and exercise testing variables (ROC analysis and logistic regression). By raising the threshold of abnormality of L/H(net), specificity and positive predictive value increased, whereas sensitivity and negative predictive value decreased. Lung 201Tl assessment assists substantially in the identification of higher-risk CAD in exercise SPECT myocardial perfusion imaging and this is best achieved by L/H(net). This index is a significant predictor of higher-risk CAD, superior to myocardial perfusion images, and its value is associated with the probability of a disease state.

To explore risk factors for coronary artery disease (CAD) among middle-aged women in Taiwan. Coronary artery disease is a leading cause of death among females. Risk factors for CAD vary due to differences in ethnicity, gender and age. However, few studies have documented risk factors among middle-aged women. We employed a cross-sectional, comparative study design. Sixty-five middle-aged women who were suspected of having CAD and who received cardiac catheterisation were purposively sampled and divided into a CAD group (with at least one coronary artery with >50% stenosis) and a control group, according to the results of catheterisation. Individual questionnaires regarding their medical history, blood test results, sociodemographic characteristics, metabolism, biomarkers and lifestyle risk factors were administered and quantified. The mean age of the 65 women (31 CAD and 34 controls) was 56·2years. Within the CAD group, there was a greater incidence of women with a history of diabetes mellitus (DM), increased fasting blood glucose and increased diastolic blood pressure. Comparatively fewer women within the CAD category used dietary supplements or had a lower level of physical activity. After adjusting for other confounders, it was discovered that women who used dietary supplements (OR=0·28; p=0·04) and engaged in physical activities (OR=0·16; p=0·02) were less likely to develop CAD. Use of dietary supplements and engaging in physical activities can significantly predict the incidence of CAD among middle-aged women in Taiwan. Middle-aged women should be encouraged to take appropriate dietary supplements and engage in physical activity in order to prevent CAD.

Evidence-based medicine has confirmed the role of psychosocial factors in the pathogenesis of many diseases, both cardiovascular (CVD) and metabolic. On the other hand, CVD patients often suffer from concomitant diseases. Depression was found to be an independent predictor of coronary artery disease (CAD) in many populations. To evaluate the association between depressive symptoms (DS) and CAD in the Polish adult population. A random sample of the Polish population (6392 men and 7153 women), aged 20-74 years, was examined in 2003-2005 for the presence of DS using the Beck Depression Inventory. In the examined population, CAD was found in 12.1% of men and 11.0% of women. Persons with CAD were older, more often finished their education at the level of primary school and lived in large communes, and more often had obesity, hypertension, diabetes and hyperlipidaemia compared to those without CAD. DS were found twice more often in persons with CAD compared to those without CAD, both in men and women. Subjects with DS were twice more likely to have CAD (men: odds ratio [OR] 2.14, 95% confidence interval [CI] 1.78-2.56; women: OR 2.03, 95% CI 1.70-2.43) and arrhythmia (women), and 1.5-fold more likely to report myocardial infarction and arrhythmia. A significant association between DS and CAD, myocardial infarction and arrhythmia independent of CVD risk factors was found in the Polish adult population.