Interpreting Wide-Complex Tachycardia using Artificial Intelligence.

BACKGROUND: Adopting artificial intelligence in medicine may improve speed and accuracy in patient diagnosis. We sought to develop an artificial intelligence (AI) algorithm to interpret wide complex tachycardia (WCT) electrocardiograms (ECG) and compare its diagnostic accuracy to cardiologists. METHODS: Using 3330 WCT ECGs (2906 SVT and 424 VT), we created a training/validation (3131) and test set (199 ECGs). A convolutional neural network (CNN) structure using a modification of differentiable architecture search (DARTS), ZeroLess-DARTS, was developed to differentiate between SVT and VT. RESULTS: The mean accuracy of electrophysiology (EP) cardiologists was 92.5% with a sensitivity of 91.7%, specificity of 93.4%, positive predictive value of 93.7%, negative predictive value of 91.7%. NonEP cardiologists had an accuracy of 73.2 ± 14.4% with a sensitivity, specificity, positive and negative predictive value of 59.8 ± 18.2%, 93.8 ± 3.7%, 93.6 ± 2.3%, and 73.2 ± 14.4%, respectively. AI had superior sensitivity and accuracy (91.9% and 93.0%, respectively) than NonEP cardiologists, and had similar performance of EP cardiologists. Mean time to interpret each ECG varied between 10.1-13.8 seconds for EP cardiologists and 3.1 -16.6 seconds for NonEP cardiologists. Conversely AI required a mean of 0.0092 ± 0.0035 seconds for each ECG interpretation. CONCLUSIONS: AI appears to diagnose WCT with superior accuracy than Cardiologists and similar to those of Electrophysiologists. Using AI to assist with ECG interpretations may improve patient care.

Telehealth during COVID-19 restrictions in patients with cardiovascular disease: impact on medication prescriptions and patient satisfaction.

BACKGROUND AND AIMS: Telehealth plays an integral part in healthcare delivery. The impact of telehealth and the COVID-19 pandemic on medication prescribing and patient satisfaction with telehealth in cardiology clinics remains unknown. METHODS: A retrospective study of cardiology clinic patients at an Australian tertiary hospital was conducted; 630 patients seen before the COVID-19 pandemic (0.6% telehealth) and 678 during the pandemic (91.2% telehealth) were included. Medication changes, new prescriptions and time to obtaining prescriptions after clinic were compared. To evaluate patients' experiences, cardiology clinic patients reviewed during the pandemic were prospectively invited to participate in an electronic survey sent to their mobile phones. RESULTS: The overall rates of medication changes made in the clinic between the prepandemic and the pandemic periods did not differ significantly (26.9% vs 25.8%). Compared with prepandemic, new cardiac medication prescriptions during clinic were significantly less (9.3% vs 2.5%; P < 0.0001) and recommendations to general practitioners (GP) to initiate cardiac medications were significantly more (2.6% vs 9.1%; P < 0.0001). Time to obtaining new prescriptions was significantly longer in the pandemic cohort (median 0 days (range: 0-32) vs 10.5 days (range: 0-231); P < 0.0001). Two hundred forty-three (32.7%) patients participated in the survey; 50% reported that telehealth was at least as good as face-to-face consultations. Most patients (61.5%) were satisfied with telehealth and most (62.9%) wished to see telehealth continued postpandemic. CONCLUSION: Telehealth during the COVID-19 pandemic was associated with greater reliance on GP to prescribe cardiac medications and delays in obtaining prescriptions among cardiology clinic patients. Although most patients were satisfied with telehealth services, nearly half of the cardiac patients expressed preference towards traditional face-to-face consultations.

Early SGLT2 inhibitor use is associated with improved left atrial strain following acute coronary syndrome.

AIM: Left atrial (LA) strain, a novel marker of LA function, reliably predicts diastolic dysfunction. SGLT2 inhibitors improve heart failure outcomes, but limited data exists regarding their use in the immediate aftermath of acute coronary syndrome (ACS). We studied the effect of empagliflozin on LA strain in patients with type 2 diabetes (T2D) and ACS. METHODS: Patients with ACS and T2D were identified and empagliflozin was initiated in eligible patients prior to discharge. Patients not initiated on empagliflozin were analysed as a comparator group. A blinded investigator assessed LA strain using baseline and 3-6 month follow-up echocardiograms. RESULTS: Forty-four participants (n = 22 each group) were included. Baseline characteristics and LA strain were similar in the two groups. LA reservoir, conduit and contractile strain increased in empagliflozin group (28.0 ± 8.4% to 34.6 ± 12.2% p < 0.001, 14.5 ± 5.4% to 16.7 ± 7.0% p = 0.034, 13.5 ± 5.2% to 17.9 ± 7.2% p = 0.005, respectively) but remained unchanged in comparison group (29.2 ± 6.7% to 28.8 ± 7.0%, 12.8 ± 4.2% to 13.3 ± 4.7%, 16.7 ± 5.3% to 15.5 ± 4.5%, respectively, p = NS). The difference in change between groups was significant for LA reservoir (p = 0.003) and contractile strain (p = 0.005). CONCLUSION: In patients with ACS and T2D, addition of empagliflozin to standard ACS therapy prior to discharge is associated with improved LA function.

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.

Obesity in patients without diabetes undergoing coronary artery bypass graft surgery: Time to select patients for semaglutide.

Patients undergoing coronary artery bypass graft (CABG) surgery require intensive secondary prevention. Semaglutide reduced cardiovascular events in patients with cardiovascular disease and overweight or obesity but without diabetes in the SELECT trial. In this real-world study of 1386 patients without diabetes undergoing CABG surgery in an Australian hospital, approximately 1 in 2 patients (53.3 %) were potentially eligible for semaglutide based on the SELECT trial criteria. These findings highlight that a significant percentage of this very high-risk cohort merit receiving semaglutide for weight management and cardiovascular risk reduction. The implications for optimal care, healthcare costs and clinical guidelines require further evaluation.

Screening tools for the detection of clinically silent cardiac sarcoidosis.

BACKGROUND: clinically silent cardiac sarcoidosis (CS) may be associated with adverse outcomes, hence the rationale for screening patients with extracardiac sarcoidosis. The optimal screening strategy has not been clearly defined. METHODS: patients with extra-cardiac sarcoidosis were prospectively included and underwent screening consisting of symptom history, electrocardiography (ECG), transthoracic echocardiogram, Holter, and signal-averaged ECG (SAECG). Cardiac magnetic resonance (CMR) was performed in all patients. Clinically silent CS was defined as CMR demonstrating late gadolinium enhancement (LGE) in a pattern compatible with CS according to a majority of independent and blinded CMR experts. Significant cardiac involvement was defined as the presence of LGE ≥6% and/or a positive fluorodeoxyglucose-positron emission tomography. RESULTS: among the 129 patients included, clinically silent CS was diagnosed in 29/129 (22.5%), and 19/129 patients (14.7%) were classified as CS with significant cardiac involvement. There was a strong association between hypertension and CS (p < 0.05). Individual screening tools provided low diagnostic yield; however, combination of tests performed better, for example, a normal Holter and a normal SAECG had negative predictive values of 91.7%. We found consistently better diagnostic accuracy for the detection of CS with significant cardiac involvement. CONCLUSION: clinically silent CS and CS with significant cardiac involvement were found in 22.5% and 14.7% of patients with extra-cardiac sarcoidosis. The association with hypertension raises the possibility that some cases of hypertensive cardiomyopathy may be mistaken for CS. Screening with readily available tools, for example Holter and SAECG, may help identifying patients without CS where additional CMR is not needed.

Atherosclerosis as the Damocles' sword of human evolution: insights from nonhuman ape-like primates, ancient human remains, and isolated modern human populations.

Atherosclerosis is the leading cause of death worldwide, and the predominant risk factors are advanced age and high-circulating low-density lipoprotein cholesterol (LDL-C). However, the findings of atherosclerosis in relatively young mummified remains and a lack of atherosclerosis in chimpanzees despite high LDL-C call into question the role of traditional cardiovascular risk factors. The inflammatory theory of atherosclerosis may explain the discrepancies between traditional risk factors and observed phenomena in current literature. Following the divergence from chimpanzees several millennia ago, loss of function mutations in immune regulatory genes and changes in gene expression have resulted in an overactive human immune system. The ubiquity of atherosclerosis in the modern era may reflect a selective pressure that enhanced the innate immune response at the cost of atherogenesis and other chronic disease states. Evidence provided from the fields of genetics, evolutionary biology, and paleoanthropology demonstrates a sort of circular dependency between inflammation, immune system functioning, and evolution at both a species and cellular level. More recently, the role of proinflammatory stimuli, somatic mutations, and the gene-environment effect appear to be underappreciated elements in the development and progression of atherosclerosis. Neurobiological stress, metabolic syndrome, and traditional cardiovascular risk factors may instead function as intermediary links between inflammation and atherosclerosis. Therefore, considering evolution as a mechanistic process and atherosclerosis as part of the inertia of evolution, greater insight into future preventative and therapeutic interventions for atherosclerosis can be gained by examining the past.

Evaluation of stable chest pain following emergency department presentation: Impact of first-line cardiac computed tomography diagnostic strategy in an Australian setting.

OBJECTIVE: International guidelines provide increasing support for computed tomography coronary angiography (CTCA) in investigating chest pain. A pathway utilising CTCA first-line for outpatient stable chest pain evaluation was implemented in an Australian ED. METHODS: In pre-post design, the impact of the pathway was prospectively assessed over 6 months (August 2021 to January 2022) and compared with a 6-month pre-implementation group (February 2021 to July 2021). CTCA was recommended first-line in suspected stable cardiac chest pain, followed by chest pain clinic review. Predefined criteria were provided recommending functional testing in select patients. The impact of CTCA versus functional testing was evaluated. Data were obtained from digital medical records. RESULTS: Three hundred and fifteen patients were included, 143 pre-implementation and 172 post-implementation. Characteristics were similar except age (pre-implementation: 58.9 ± 12.0 vs post-implementation: 62.8 ± 12.3 years, P = 0.004). Pathway-guided management resulted in higher first-line CTCA (73.3% vs 46.2%, P < 0.001), lower functional testing (30.2% vs 56.6%, P < 0.001) and lower proportion undergoing two non-invasive tests (4.7% vs 10.5%, P = 0.047), without increasing investigation costs or invasive coronary angiography (ICA) (pre-implementation: 13.3% vs post-implementation: 9.3%, P = 0.263). In patients undergoing CTCA, 40.7% had normal coronaries and 36.2% minimal/mild disease, with no difference in disease burden post-implementation. More medication changes occurred following CTCA compared with functional testing (aspirin: P = 0.005, statin: P < 0.001). In patients undergoing ICA, revascularisation to ICA ratio was higher following CTCA compared with functional testing (91.7% vs 18.2%, P < 0.001). No 30-day myocardial infarction or death occurred. CONCLUSIONS: The pathway increased CTCA utilisation and reduced downstream investigations. CTCA was associated with medication changes and improved ICA efficiency.

Gut microbiota and metabolomics profiles in patients with chronic stable angina and acute coronary syndrome.

Cardiovascular disease (CVD) is the leading cause of death worldwide. The gut microbiota and its associated metabolites may be involved in the development and progression of CVD, although the mechanisms and impact on clinical outcomes are not fully understood. This study investigated the gut microbiome profile and associated metabolites in patients with chronic stable angina (CSA) and acute coronary syndrome (ACS) compared with healthy controls. Bacterial alpha diversity in stool from patients with ACS or CSA was comparable to healthy controls at both baseline and follow-up visits. Differential abundance analysis identified operational taxonomic units (OTUs) assigned to commensal taxa differentiating patients with ACS from healthy controls at both baseline and follow-up. Patients with CSA and ACS had significantly higher levels of trimethylamine N-oxide compared with healthy controls (CSA: 0.032 ± 0.023 mmol/L, P < 0.01 vs. healthy, and ACS: 0.032 ± 0.023 mmol/L, P = 0.02 vs. healthy, respectively). Patients with ACS had reduced levels of propionate and butyrate (119 ± 4 vs. 139 ± 5.1 µM, P = 0.001, and 14 ± 4.3 vs. 23.5 ± 8.1 µM, P < 0.001, respectively), as well as elevated serum sCD14 (2245 ± 75.1 vs. 1834 ± 45.8 ng/mL, P < 0.0001) and sCD163 levels (457.3 ± 31.8 vs. 326.8 ± 20.7 ng/mL, P = 0.001), compared with healthy controls at baseline. Furthermore, a modified small molecule metabolomic and lipidomic signature was observed in patients with CSA and ACS compared with healthy controls. These findings provide evidence of a link between gut microbiome composition and gut bacterial metabolites with CVD. Future time course studies in patients to observe temporal changes and subsequent associations with gut microbiome composition are required to provide insight into how these are affected by transient changes following an acute coronary event.NEW & NOTEWORTHY The study found discriminative microorganisms differentiating patients with acute coronary syndrome (ACS) from healthy controls. In addition, reduced levels of certain bacterial metabolites and elevated sCD14 and sCD163 were observed in patients with ACS compared with healthy controls. Furthermore, modified small molecule metabolomic and lipidomic signatures were found in both patient groups. Although it is not known whether these differences in profiles are associated with disease development and/or progression, the findings provide exciting options for potential new disease-related mechanism(s) and associated therapeutic target(s).

Advances in cellular and tissue-based imaging techniques for sarcoid granulomas.

Sarcoidosis embodies a complex inflammatory disorder spanning multiple systems, with its origin remaining elusive. It manifests as the infiltration of inflammatory cells that coalesce into distinctive noncaseous granulomas within afflicted organs. Unraveling this disease necessitates the utilization of cellular or tissue-based imaging methods to both visualize and characterize the biochemistry of these sarcoid granulomas. Although hematoxylin and eosin stain, standard in routine use alongside cytological stains have found utility in diagnosis within clinical contexts, special stains such as Masson's trichrome, reticulin, methenamine silver, and Ziehl-Neelsen provide additional varied perspectives of sarcoid granuloma imaging. Immunohistochemistry aids in pinpointing specific proteins and gene expressions further characterizing these granulomas. Finally, recent advances in spatial transcriptomics promise to divulge profound insights into their spatial orientation and three-dimensional (3-D) molecular mapping. This review focuses on a range of preexisting imaging methods employed for visualizing sarcoid granulomas at the cellular level while also exploring the potential of the latest cutting-edge approaches like spatial transcriptomics and matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI-MSI), with the overarching goal of shedding light on the trajectory of sarcoidosis research.

Attitudes towards artificial intelligence in emergency medicine.

OBJECTIVE: To assess Australian and New Zealand emergency clinicians' attitudes towards the use of artificial intelligence (AI) in emergency medicine. METHODS: We undertook a qualitative interview-based study based on grounded theory. Participants were recruited through ED internal mailing lists, the Australasian College for Emergency Medicine Bulletin, and the research teams' personal networks. Interviews were transcribed, coded and themes presented. RESULTS: Twenty-five interviews were conducted between July 2021 and May 2022. Thematic saturation was achieved after 22 interviews. Most participants were from either Western Australia (52%) or Victoria (16%) and were consultants (96%). More participants reported feeling optimistic (10/25) than neutral (6/25), pessimistic (2/25) or mixed (7/25) towards the use of AI in the ED. A minority expressed scepticism regarding the feasibility or value of implementing AI into the ED. Multiple potential risks and ethical issues were discussed by participants including skill loss from overreliance on AI, algorithmic bias, patient privacy and concerns over liability. Participants also discussed perceived inadequacies in existing information technology systems. Participants felt that AI technologies would be used as decision support tools and not replace the roles of emergency clinicians. Participants were not concerned about the impact of AI on their job security. Most (17/25) participants thought that AI would impact emergency medicine within the next 10 years. CONCLUSIONS: Emergency clinicians interviewed were generally optimistic about the use of AI in emergency medicine, so long as it is used as a decision support tool and they maintain the ability to override its recommendations.

Applications of natural language processing at emergency department triage: A narrative review.

INTRODUCTION: Natural language processing (NLP) uses various computational methods to analyse and understand human language, and has been applied to data acquired at Emergency Department (ED) triage to predict various outcomes. The objective of this scoping review is to evaluate how NLP has been applied to data acquired at ED triage, assess if NLP based models outperform humans or current risk stratification techniques when predicting outcomes, and assess if incorporating free-text improve predictive performance of models when compared to predictive models that use only structured data. METHODS: All English language peer-reviewed research that applied an NLP technique to free-text obtained at ED triage was eligible for inclusion. We excluded studies focusing solely on disease surveillance, and studies that used information obtained after triage. We searched the electronic databases MEDLINE, Embase, Cochrane Database of Systematic Reviews, Web of Science, and Scopus for medical subject headings and text keywords related to NLP and triage. Databases were last searched on 01/01/2022. Risk of bias in studies was assessed using the Prediction model Risk of Bias Assessment Tool (PROBAST). Due to the high level of heterogeneity between studies and high risk of bias, a metanalysis was not conducted. Instead, a narrative synthesis is provided. RESULTS: In total, 3730 studies were screened, and 20 studies were included. The population size varied greatly between studies ranging from 1.8 million patients to 598 triage notes. The most common outcomes assessed were prediction of triage score, prediction of admission, and prediction of critical illness. NLP models achieved high accuracy in predicting need for admission, triage score, critical illness, and mapping free-text chief complaints to structured fields. Incorporating both structured data and free-text data improved results when compared to models that used only structured data. However, the majority of studies (80%) were assessed to have a high risk of bias, and only one study reported the deployment of an NLP model into clinical practice. CONCLUSION: Unstructured free-text triage notes have been used by NLP models to predict clinically relevant outcomes. However, the majority of studies have a high risk of bias, most research is retrospective, and there are few examples of implementation into clinical practice. Future work is needed to prospectively assess if applying NLP to data acquired at ED triage improves ED outcomes when compared to usual clinical practice.

Enhancing Risk Stratification on Coronary Computed Tomography Angiography: The Role of Artificial Intelligence.

PURPOSE: To describe and outline the role of artificial intelligence (AI) in assisting coronary computed tomography angiography (CCTA) in enhancing risk stratification. METHODS: A comprehensive review of the literature was performed to identify published work investigating the utility of applying AI to CCTA. FINDINGS: CCTA is an excellent diagnostic tool for the detection of atherosclerotic cardiovascular disease. The noninvasive nature and high diagnostic accuracy have made CCTA a viable alternative to invasive coronary angiography to detect luminal stenosis. However, it is now understood that stenosis is just one factor that predicts cardiac risk and other factors need to be considered. CCTA-derived plaque biomarkers have since emerged as established predictors of cardiac events to improve risk stratification. Despite awareness of these biomarkers, they are still yet to be incorporated into routine clinical practice. The major barriers to implementation include the specialized skills required for image evaluation and the time intensive nature of analysis. With the many recent advancements in the technology, AI presents itself as a promising solution. AI is attractive because it has the potential to rapidly automate technically challenging tasks with exceptional accuracy. IMPLICATIONS: Developments in the field of AI are occurring at a rapid rate. There is already increasing evidence of the potential AI has to greatly improve the utility of CCTA by improving analysis time and extracting additional prognostic data from new plaque biomarkers. There are, however, technical and ethical challenges that need to be considered before implementing such technology into routine clinical practice.

The role of the host-microbiome and metabolomics in sarcoidosis.

Sarcoidosis is a complex inflammatory fibrotic disease that affects multiple organ systems. It is characterized by the infiltration of lymphocytes and mononuclear phagocytes, which form non-caseating granulomas in affected organs. The lungs and intrathoracic lymph nodes are the most commonly affected organs. The underlying cause of sarcoidosis is unknown, but it is believed to occur in genetically predisposed individuals who are exposed to pathogenic organisms, environmental contaminants, or self and non-self-antigens. Recent research has suggested that the microbiome may play a role in the development of respiratory conditions, including sarcoidosis. Additionally, metabolomic studies have identified potential biomarkers for monitoring sarcoidosis progression. This review will focus on recent microbiome and metabolomic findings in sarcoidosis, with the goal of shedding light on the pathogenesis and possible diagnostic and therapeutic approaches.

Insights into the associations between the gut microbiome, its metabolites, and heart failure.

Heart failure (HF) is the end stage of most cardiovascular diseases and remains a significant health problem globally. We aimed to assess whether patients with left ventricular ejection fraction ≤45% had alterations in both the gut microbiome profile and production of associated metabolites when compared with a healthy cohort. We also examined the associated inflammatory, metabolomic, and lipidomic profiles of patients with HF. This single center, observational study, recruited 73 patients with HF and 59 healthy volunteers. Blood and stool samples were collected at baseline and 6-mo follow-up, along with anthropometric and clinical data. When compared with healthy controls, patients with HF had reduced gut bacterial alpha diversity at follow-up (P = 0.004) but not at baseline. The stool microbiota of patients with HF was characterized by a depletion of operational taxonomic units representing commensal Clostridia at both baseline and follow-up. Patients with HF also had significantly elevated baseline plasma acetate (P = 0.007), plasma trimethylamine-N-oxide (TMAO) (P = 0.003), serum soluble CD14 (sCD14; P = 0.005), and soluble CD163 (sCD163; P = 0.004) levels compared with healthy controls. Furthermore, patients with HF had a distinct metabolomic and lipidomic profile at baseline when compared with healthy controls. Differences in the composition of the gut microbiome and the levels of associated metabolites were observed in patients with HF when compared with a healthy cohort. This was also associated with an altered metabolomic and lipidomic profile. Our study identifies microorganisms and metabolites that could represent new therapeutic targets and diagnostic tools in the pathogenesis of HF.NEW & NOTEWORTHY We found a reduction in gut bacterial alpha diversity in patients with heart failure (HF) and that the stool microbiota of patients with HF was characterized by depletion of operational taxonomic units representing commensal Clostridia at both baseline and follow-up. Patients with HF also had altered bacterial metabolites and increased inflammatory profiles compared with healthy controls. A distinct metabolomic and lipidomic profile was present in patients with HF at baseline when compared with healthy controls.

Artificial Intelligence in the Image-Guided Care of Atrial Fibrillation.

Atrial fibrillation arises mainly due to abnormalities in the cardiac conduction system and is associated with anatomical remodeling of the atria and the pulmonary veins. Cardiovascular imaging techniques, such as echocardiography, computed tomography, and magnetic resonance imaging, are crucial in the management of atrial fibrillation, as they not only provide anatomical context to evaluate structural alterations but also help in determining treatment strategies. However, interpreting these images requires significant human expertise. The potential of artificial intelligence in analyzing these images has been repeatedly suggested due to its ability to automate the process with precision comparable to human experts. This review summarizes the benefits of artificial intelligence in enhancing the clinical care of patients with atrial fibrillation through cardiovascular image analysis. It provides a detailed overview of the two most critical steps in image-guided AF management, namely, segmentation and classification. For segmentation, the state-of-the-art artificial intelligence methodologies and the factors influencing the segmentation performance are discussed. For classification, the applications of artificial intelligence in the diagnosis and prognosis of atrial fibrillation are provided. Finally, this review also scrutinizes the current challenges hindering the clinical applicability of these methods, with the aim of guiding future research toward more effective integration into clinical practice.