Analysis and evaluation of explainable artificial intelligence on suicide risk assessment.

This study explores the effectiveness of Explainable Artificial Intelligence (XAI) for predicting suicide risk from medical tabular data. Given the common challenge of limited datasets in health-related Machine Learning (ML) applications, we use data augmentation in tandem with ML to enhance the identification of individuals at high risk of suicide. We use SHapley Additive exPlanations (SHAP) for XAI and traditional correlation analysis to rank feature importance, pinpointing primary factors influencing suicide risk and preventive measures. Experimental results show the Random Forest (RF) model is excelling in accuracy, F1 score, and AUC (>97% across metrics). According to SHAP, anger issues, depression, and social isolation emerge as top predictors of suicide risk, while individuals with high incomes, esteemed professions, and higher education present the lowest risk. Our findings underscore the effectiveness of ML and XAI in suicide risk assessment, offering valuable insights for psychiatrists and facilitating informed clinical decisions.

Clinical outcomes and health care costs of transferring rural Western Australians for invasive coronary angiography, and a cost-effective alternative care model: a retrospective cross-sectional study.

OBJECTIVES: To examine the severity of coronary artery disease (CAD) in people from rural or remote Western Australia referred for invasive coronary angiography (ICA) in Perth and their subsequent management; to estimate the cost savings were computed tomography coronary angiography (CTCA) offered in rural centres as a first line investigation for people with suspected CAD. DESIGN: Retrospective cohort study. SETTING, PARTICIPANTS: Adults with stable symptoms in rural and remote WA referred to Perth public tertiary hospitals for ICA evaluation during the 2019 calendar year. MAIN OUTCOME MEASURES: Severity and management of CAD (medical management or revascularisation); health care costs by care model (standard care or a proposed alternative model with local CTCA assessment). RESULTS: The mean age of the 1017 people from rural and remote WA who underwent ICA in Perth was 62 years (standard deviation, 13 years); 680 were men (66.9%), 245 were Indigenous people (24.1%). Indications for referral were non-ST elevation myocardial infarction (438, 43.1%), chest pain with normal troponin level (394, 38.7%), and other (185, 18.2%). After ICA assessment, 619 people were medically managed (60.9%) and 398 underwent revascularisation (39.1%). None of the 365 patients (35.9%) without obstructed coronaries (< 50% stenosis) underwent revascularisation; nine patients with moderate CAD (50-69% stenosis; 7%) and 389 with severe CAD (≥ 70% stenosis or occluded vessel; 75.5%) underwent revascularisation. Were CTCA used locally to determine the need for referral, 527 referrals could have been averted (53%), the ICA:revascularisation ratio would have improved from 2.6 to 1.6, and 1757 metropolitan hospital bed-days (43% reduction) and $7.3 million in health care costs (36% reduction) would have been saved. CONCLUSION: Many rural and remote Western Australians transferred for ICA in Perth have non-obstructive CAD and are medically managed. Providing CTCA as a first line investigation in rural centres could avert half of these transfers and be a cost-effective strategy for risk stratification of people with suspected CAD.

Capturing the pulse: a state-of-the-art review on camera-based jugular vein assessment.

Heart failure is associated with a rehospitalisation rate of up to 50% within six months. Elevated central venous pressure may serve as an early warning sign. While invasive procedures are used to measure central venous pressure for guiding treatment in hospital, this becomes impractical upon discharge. A non-invasive estimation technique exists, where the clinician visually inspects the pulsation of the jugular veins in the neck, but it is less reliable due to human limitations. Video and signal processing technologies may offer a high-fidelity alternative. This state-of-the-art review analyses existing literature on camera-based methods for jugular vein assessment. We summarize key design considerations and suggest avenues for future research. Our review highlights the neck as a rich imaging target beyond the jugular veins, capturing comprehensive cardiac signals, and outlines factors affecting signal quality and measurement accuracy. Addressing an often quoted limitation in the field, we also propose minimum reporting standards for future studies.

Reproducibility of cardiac magnetic resonance imaging in patients referred for the assessment of cardiac sarcoidosis; implications for clinical practice.

Cardiac sarcoidosis (CS) is an increasingly recognized condition, but cardiac magnetic resonance (CMR) image interpretation in these patients may be challenging as findings are often non-specific. The main objective of this study was to investigate the inter-reader agreement for the overall interpretation of CMR for the diagnosis of CS in an experienced reference center and investigate factors that may lead to discrepancies between readers. Consecutive patients undergoing CMR imaging to investigate for CS were included. CMR images were independently reviewed by two readers, blinded to all clinical, imaging and demographic information. The readers classified each scan as "consistent with cardiac sarcoidosis", "not consistent with cardiac sarcoidosis" or "indeterminate". Inter-reader agreement was assessed using κ-statistics. When there was disagreement on the overall interpretation, a third reader reviewed the images. Also, two readers independently commented on the presence of edema, presence of LGE (both ventricles) and quantified the extent of left ventricular LGE. 87 patients (43 women, mean age 54.3 ± 12.2 years) were included in the study. There was agreement regarding the overall interpretation in 72 of 87 (83%) CMR scans. The κ value was 0.64, indicating moderate agreement. There was similar moderate agreement in the interpretation of LGE parameters. In an experienced referral center, we found moderate agreement between readers in the interpretation of CMR in patients with suspected CS. Physicians should be aware of this inter-observer variability in interpretation of CMR studies in patients with suspected CS.

The echocardiographic assessment of the right ventricle in patients with arrhythmogenic right ventricular cardiomyopathy/dysplasia compared with athletes and matched controls.

BACKGROUND: There are discrepancies in the quantitative echocardiographic criteria for the right ventricle (RV) between the revised task force criteria (TFC) for Arrhythmogenic Right Ventricular Cardiomyopathy/Dysplasia (ARVC/D) and the guidelines for RV assessment endorsed by American Society of Echocardiography (ASE). Importantly, these criteria do not take into account potential adaptation of the RV to exercise. The goal of this study was to compare the revised TFC quantitative echocardiographic parameters in patients with ARVC/D, athletes and matched controls. METHODS: Echocardiographic parameters of the RV were retrospectively collected in patients who fulfilled the TFC for ARVC/D, an age- matched, sex-matched, and body surface area-matched control population, and athletes (defined as individuals who exercised for more than 7 hours per week). Patients with structural heart disease were excluded in the control and athlete groups. RESULTS: Twenty patients with ARVC/D, 11 athletes and 20 matched controls were included. There was no significant difference between ARVC/D patients and athletes with the exception of the parasternal long axis right ventricular outflow tract diameter. All parameters were significantly different between ARVC/D patients and the control group. Furthermore, when subjects were categorized into meeting 1 major revised TFC/abnormal ASE criteria or not, only ASE criteria were able to differentiate ARVC/D from control population. Both were unable to differentiate ARVC/D from athletes. CONCLUSIONS: Right ventricle quantitative echocardiographic criteria in the revised TFC are not specific for ARVC/D. Care should be taken in applying these criteria in athletes.

Cardiac Surgery in HIV Patients: State of the Art.

The clinical status of HIV infection has changed dramatically with the introduction of combined antiretroviral therapy. Patients with HIV are now living long enough to be susceptible to chronic illnesses, such as coronary disease and nonischemic cardiomyopathy, which can be consequences of the combined antiretroviral therapy treatment itself. Cardiovascular diseases are a major source of morbidity and mortality in HIV-positive patients. Increasingly, such patients might be candidates for the full range of cardiac surgical interventions, including coronary bypass, valve surgery, and heart transplantation. There has been a shift from offering palliative procedures such as pericardial window and balloon valvuloplasty, to more conventional and durable surgical therapies in HIV-positive patients. We herein provide an overview of the contemporary outcomes of cardiac surgery in this complex and unique patient population. We review some of the ethical issues around the selection and surgical care of HIV-positive patients. We also discuss strategies to best protect the surgical treatment team from the risks of HIV transmission. Finally, we highlight the need for involvement of dedicated infectious disease professionals in a multidisciplinary heart team approach, aiming at the comprehensive care of these unique and complex patients.

Sex Differences in Associations of Arterial Compliance With Coronary Artery Plaque and Calcification Burden.

BACKGROUND: Coronary artery disease is a leading killer of women. Arterial stiffness predicts myocardial infarction, and postmenopausal women have lower arterial compliance (AC) than men. We hypothesized that lower AC would be associated with greater burden of coronary artery plaque and calcification, and that these associations would be stronger in women than men. METHODS AND RESULTS: We evaluated 3639 consecutive adults without coronary artery disease history who had coronary computed tomography between 2006 and 2014. Coronary artery calcification was calculated using the Agatston method. Plaque extent was assessed by the number of arterial segments with visible plaque divided by the number of visualized segments ×100 (percent plaque score). Indexed AC was calculated as stroke volume index/central pulse pressure. We used step-wise multivariable linear regression to assess associations of log indexed AC with log (percent plaque score+1) and log (coronary artery calcification+1). Sex-specific models were performed if the interaction sex×AC was significant. Mean age was 57±11 years, 53% were men, and 71% were hypertensive. Interaction term sex×AC was significant for both models (P=0.008 for percent plaque score and 0.022 for coronary artery calcification). Lower indexed AC was associated with higher percent plaque score and coronary artery calcification in women (ß±SE: -0.231±0.113, P=0.042 and -0.334±0.166, P=0.044, respectively), but not in men (ß±SE: -0.062±0.104, P=0.551 and 0.114±0.173, P=0.510, respectively). CONCLUSIONS: Lower AC is associated with greater burden of coronary artery plaque and calcification in women, but not in men. Our findings highlight low AC as a correlate of more-advanced coronary artery disease and as a potential link to the worse cardiovascular outcomes in women.

Cardiac CT assessment of left ventricular mass in mid-diastasis and its prognostic value.

AIMS: To determine the influence of cardiac motion on measurements of left ventricular (LV) mass obtained with 64-slice computed tomography (CT) and to elucidate the prognostic value of LV mass on major adverse cardiac events (MACE) and all-cause mortality. Increased LV mass has been linked with MACE. Although Cardiac CT allows measurement of LV anatomy, it is susceptible to motion artefacts often requiring image acquisition during diastasis. There is a need to understand variability in LV mass measurements across phases of the cardiac cycle, and whether mid-diastolic measurements have prognostic value. METHODS AND RESULTS: The study comprised two equally sized cohorts of patients that had undergone retrospectively gated cardiac CT: patients who had MACE and/or all-cause death at follow-up and a matched (age, sex, and risk factors) event-free cohort. LV mass was measured at mid-diastole, end-diastole, and end-systole. Correlation and agreement between phases were determined. The incremental value of mid-diastolic hypertrophy (LVH) over the National Cholesterol Education Programme (NCEP) risk was performed for LV mass indices normalized to body surface area (LVMIBSA) or weight (LVMIWeight). Of 166 patients, 31.3% experienced MACE and 28.9% died of any cause (follow-up 22.9 ± 13.4 months). LV mass at all cardiac phases were strongly correlated (r > 0.94). Mean mid-diastolic LVMIBSA was higher in the cohort with events (93.7 vs. 80.7 g/m2, P= 0.008) as was LVMIWeight (2.26 vs. 1.88 g/kg, P = 0.001). LVMIBSA and LVMIWeight had prognostic value incremental to NCEP with 1.85 and 2.47 hazard ratios, respectively. CONCLUSIONS: Measurement of LV mass can be obtained by cardiac CT images obtained at mid-diastasis. LV mass measurements obtained at mid-diastasis have prognostic value.

Routine versus selective cardiac magnetic resonance in non-ischemic heart failure - OUTSMART-HF: study protocol for a randomized controlled trial (IMAGE-HF (heart failure) project 1-B).

BACKGROUND: Imaging has become a routine part of heart failure (HF) investigation. Echocardiography is a first-line test in HF given its availability and it provides valuable diagnostic and prognostic information. Cardiac magnetic resonance (CMR) is an emerging clinical tool in the management of patients with non-ischemic heart failure. Current ACC/AHA/CCS/ESC guidelines advocate its role in the detection of a variety of cardiomyopathies but there is a paucity of high quality evidence to support these recommendations.The primary objective of this study is to compare the diagnostic yield of routine cardiac magnetic resonance versus standard care (that is, echocardiography with only selective use of CMR) in patients with non-ischemic heart failure. The primary hypothesisis that the routine use of CMR will lead to a more specific diagnostic characterization of the underlying etiology of non-ischemic heart failure. This will lead to a reduction in the non-specific diagnoses of idiopathic dilated cardiomyopathy and HF with preserved ejection fraction. DESIGN: Tertiary care sites in Canada and Finland, with dedicated HF and CMR programs, will randomize consecutive patients with new or deteriorating HF to routine CMR or selective CMR. All patients will undergo a standard clinical echocardiogram and the interpreter will assign the most likely HF etiology. Those undergoing CMR will also have a standard examination and will be assigned a HF etiology based upon the findings. The treating physician's impression about non-ischemic HF etiology will be collected following all baseline testing (including echo ± CMR). Patients will be followed annually for 4 years to ascertain clinical outcomes, quality of life and cost. The expected outcome is that the routine CMR arm will have a significantly higher rate of infiltrative, inflammatory, hypertrophic, ischemic and 'other' cardiomyopathy than the selective CMR group. DISCUSSION: This study will be the first multicenter randomized, controlled trial evaluating the role of CMR in non-ischemic HF. Non-ischemic HF patients will be randomized to routine CMR in order to determine whether there are any gains over management strategies employing selective CMR utilization. The insight gained from this study should improve appropriate CMR use in HF. TRIAL REGISTRATION: NCT01281384.

A comprehensive assessment of cardiac structure and function in patients with treated malignant phase hypertension: the West Birmingham Malignant Hypertension project.

BACKGROUND: Previous studies have confirmed that cardiac structural and functional abnormalities exist in patients with malignant hypertension (MHT). The effect of long-term blood pressure control in MHT patients on cardiac structure and function is still unknown. METHODS: We performed detailed left ventricle (LV) assessment using two-dimensional (2DE) and three-dimensional (3DE) echocardiography, and tissue Doppler imaging (TDI) in patients with previous MHT (but now in stable phase) who were compared with patients with treated 'high risk' hypertension (HHT, but non-MHT) and healthy controls (HC). Vasodilator stress myocardial contrast echocardiography (in addition to wall motion analysis) was used to exclude significant coronary artery disease, as part of our comprehensive echocardiographic assessment. Septal and posterior wall thickness, LV mass index, LV volumes and ejection fraction, mitral valve inflow indices (E, A) mitral annular velocity (S, E') and left atrial volume index (LAVI), were calculated using 2DE, 3DE, and TDI. MHT patients had good blood pressure control for an average of 144months. RESULTS: A total of 95 subjects (MHT=15; HHT and HC=40 each) were studied. Both posterior and septal wall thickness were significantly higher in the MHT and hypertensive groups compared to normal controls with no difference between MHT and HHT. No significant difference in LV ejection fraction was found between the 3 groups. Increased LAVI (p<0.05 MHT vs. HC and HHT vs. HC), reduced 'S' velocity on TDI (p=0.05 MHT vs. HC and vs.HHT, p<0.001 HHT vs. HC) and higher E/E' (p=0.029 HHT vs. HC) and lower E/A ratio (p=0.001 MHT vs. HC, p<0.001 HHT vs. HC) values were detected in the two hypertensive groups. CONCLUSION: Despite long-term good blood pressure control, MHT patients have persistent structural and functional changes in LV function on echocardiography, comparable to that seen in HHT.

Assessment of cardiac computed tomography-myocardial perfusion imaging - promise and challenges - .

Cardiac computed tomography (CT) has evolved rapidly over the last decade into a reliable imaging modality for the non-invasive assessment of coronary artery disease. With the advancement in multi-detector CT technology, there has developed an increasing body of evidence that suggests that the role of cardiac CT can be extended to include functional assessment of the myocardium not only at rest but also during stress. Simultaneous anatomical and functional assessment approaches will have a number of advantages such as evaluation of the transmural extent of myocardial perfusion defects (including small subendocardial perfusion defects), reduced risk associated with multiple sources of radiation, and short image acquisition time. Although initial results hold some promise, CT myocardial perfusion imaging is a modality in the early stages of development and further work and studies are required to define, validate, and optimize this technique. This review will provide an overview of this novel perfusion imaging method, its underlying principles, evolution, limitations and future directions.

Endothelial dysfunction: methods of assessment & implications for cardiovascular diseases.

The endothelium is a thin monocellular layer lining the entire human vascular system, separating blood from interstitium. It plays a core role in the vascular tone by releasing a variety of vasoactive substances, such as nitric oxide (NO) and endothelin. In addition to regulating vasomotion, the healthy endothelium also has anti-thrombotic (through prostacyclins), anti-inflammatory (through developmental endothelial locus-1{Del-1}) and anti-proliferative (through NO and prostaglandin I2) properties. All such mechanisms are regulated by a strict balance amongst several agonist and antagonist biochemical substances secreted by the endothelium. Endothelial dysfunction (ED) is a systemic process in which the endothelium loses the ability/capacity to maintain vascular equilibrium. ED is strongly associated with cardiovascular risk factors/diseases and can be assessed by a number of invasive and non invasive methods. Strict physiological and/or pharmacological management of cardiovascular risk factors improves the functional status of the endothelium and reduces the risk of future cardiac events. This review will provide an overview of the modern perception of endothelial biology, the methods of its assessment and interaction of the endothelium with cardiovascular risk factors and prognosis.

Normal value of carotid intima-media thickness--a surrogate marker of atherosclerosis: quantitative assessment by B-mode carotid ultrasound.

BACKGROUND: Carotid intima-media thickness (IMT) is a surrogate marker of atherosclerosis and imparts prognostic information independent of traditional cardiovascular risk factors. Quantitative assessment of IMT using semiautomated border detection software is a new and easy technique that has been previously shown to be accurate, effective, and reproducible. The study is aimed to define the upper limit of carotid IMT at the common carotid artery (CCA) and its bifurcation among a healthy population in the United Kingdom. METHODS: Asymptomatic men and women aged 35 to 75 years, without evidence of clinical atherosclerosis, underwent B-mode carotid duplex ultrasound (Sonos 7500, Philips, Best, The Netherlands). Mean carotid IMT at the far wall of both left and right CCA were quantitatively determined using a semiautomated edge-detection algorithm (Q-lab 4, Philips). Healthy population was defined as participants with no cardiovascular disease and no evidence of diabetes mellitus or hypertension with a body mass index less than 30 kg/m2, serum cholesterol less than 6 mmol/L, and absence of carotid plaque on ultrasound. RESULTS: Of the 453 participants, 137 were found to be healthy. IMT measured at the bifurcation was found to be significantly higher compared with that at the CCA. Carotid IMT in both CCA and its bifurcation increased significantly with age. The upper limits (97.5 percentile) of IMT at CCA for participants age 35 to 39, 40 to 49, 50 to 59, and 60 years or older were 0.60, 0.64, 0.71, and 0.81 mm, respectively, whereas for that at bifurcation were 0.83, 0.77, 0.85, and 1.05 mm, respectively. CONCLUSION: This study demonstrated the value of IMT at CCA and its bifurcation in a healthy population in the United Kingdom using a semiautomated edge-detection software, which is easy to use and reproducible.

Cost effectiveness of the B type natriuretic peptide, electrocardiography, and portable echocardiography for the assessment of patients from the community with suspected heart failure.

To analyze the cost efficiency of guidelines proposed by the European Society of Cardiology for investigation of patients in the community with suspected heart failure (HF). The guidelines recommend electrocardiography (ECG) and/or measurement of N-terminal pro B type natriuretic peptide (NTproBNP) prior to referral for echocardiography. Portable echocardiography is a new but validated technique for the evaluation of HF. Accordingly, 137 suspected HF patients (mean age 71+/-13 years) from the community underwent ECG and NTproBNP estimation prior to portable echocardiography. Cost effective analysis for ECG, NTproBNP, portable echocardiography and a combination of these; to define valvular heart disease, right ventricular dysfunction and left ventricular systolic and diastolic dysfunction were compared. The cost of abnormal NTproBNP followed by portable echocardiography, abnormal ECG followed by portable echocardiography and portable echocardiography alone for the detection per case of left ventricular systolic dysfunction were 313 euro, 310 euro, and 296 euro respectively and that for detection per case of any of the aforementioned cardiac abnormalities were 198 euro, 223 euro, and 170 euro respectively. Portable echocardiography alone for the assessment of suspected HF patients resulted in a cost reduction of up to 1083 euro for the detection per case of cardiac abnormality. While a strategy where initial NTproBNP estimation is cost effective in detecting any causes of heart failure, portable echocardiography remains the most costeffective strategy to assess patients from the community with suspected heart failure.

Contrast echocardiography versus gated single photon emission computed tomography for the assessment of parameters of left ventricular remodeling after acute myocardial infarction.

BACKGROUND: Assessment of parameters of left ventricular (LV) remodeling after acute myocardial infarction (AMI) has both therapeutic and prognostic implication. Contrast echocardiography (CE) has the advantage of simultaneously assessing myocardial perfusion and LV remodeling. We aimed to evaluate the accuracy of CE to assess LV remodeling after AMI compared with technetium-99m sestamibi gated single photon emission computed tomography (SPECT). METHODS: Accordingly, 36 consecutive patients underwent gated SPECT, CE, and cardiovascular magnetic resonance imaging (CMR) 7 to 10 days after AMI. LV ejection fraction (LVEF), and LV end-systolic and end-diastolic volumes were assessed. RESULTS: Absolute differences for LVEF and LV end-diastolic volume between CMR and CE were significantly smaller than that between CMR and SPECT. CE estimate of LVEF more accurately classified patients into LVEF less than 35%, 35% to 45%, and greater than 45% (agreement = 83%, kappa = 0.66 with CMR) compared with SPECT (agreement = 61%, kappa = 0.36 with CMR). CONCLUSION: CE is more accurate than gated SPECT for the estimation of LV remodeling after AMI.

Accurate assessment of aortic stenosis with intravenous contrast.

We came across an interesting case of calcific aortic stenosis in which severity was inaccurately assessed on two-dimensional and Doppler echocardiogram resulting in catheterization. Use of intravenous transpulmonary contrast agent enhanced the Doppler signal enabling better quantification of the transvalvular gradient. Use of contrast in such difficult to image patients is very useful in establishing a correct diagnosis.