Singapore Health Services (SingHealth) is Singapore's largest group of healthcare institutions. The group was formed in 2000 and consists of four public hospitals, three community hospitals, five national specialty centres and a network of eight polyclinics. The Singapore General Hospital is the largest hospital in the group and serves as the flagship hospital for the cluster.
Importance:Heart failure (HF) is a common complication of type 2 diabetes (T2D). Oral semaglutide reduced the risk of major adverse cardiovascular (CV) events (MACE; comprising CV death, nonfatal myocardial infarction, or nonfatal stroke) in people with T2D in the SOUL trial, but the impact on HF outcomes in these participants is unknown. Objective:To evaluate the effect of oral semaglutide on HF events, MACE, and safety among participants with or without HF at baseline. Design, Setting, and Participants:This is a secondary analysis of the double-blind, placebo-controlled, event-driven, phase 3b SOUL randomized clinical trial, which was conducted at 444 centers in 33 countries. Participants were enrolled from June 17, 2019, to March 24, 2021, and had T2D and atherosclerotic CV disease and/or chronic kidney disease, stratified according to the presence or absence of HF history at baseline. Data were analyzed from December 2024 to August 2025. Intervention:Once-daily oral semaglutide or placebo in addition to standard of care. Main Outcomes and Measures:Prespecified composite HF outcome (time to first occurrence of HF hospitalization, urgent HF visit, or CV death). Results:Overall, 9650 participants (median [IQR] age, 66.0 [61.0-72.0] years; 2790 [28.9%] female) were randomized, with a mean (SD) follow-up of 47.5 (10.9) months. Of these participants, 2229 (23.1%) had HF history (991 [10.3%] with preserved ejection fraction, 592 [6.1%] with reduced ejection fraction, and 646 [6.7%] with unknown subtype). For participants with HF at baseline, the hazard ratio (HR) for risk of the composite HF outcome with oral semaglutide vs placebo was 0.78 (95% CI, 0.63-0.96) and was 1.01 (95% CI, 0.84-1.20) in those without HF at baseline (P for interaction = .06). Among participants with HF, the HR was 0.59 (95% CI, 0.39-0.86) in those with preserved ejection fraction and 0.98 (95% CI, 0.70-1.38) in those with reduced ejection fraction. There was no heterogeneity in the risk reduction of MACE with oral semaglutide in participants with HF history (HR, 0.83; 95% CI, 0.68-1.01) or without HF history (HR, 0.86; 95% CI, 0.75-0.98) (P for interaction = .77). Serious adverse event occurrence among participants with HF was similar with oral semaglutide (594 [53.8%]) and placebo (642 [57.1%]). Conclusions and Relevance:In this secondary analysis of the SOUL randomized clinical trial, among individuals with T2D, atherosclerotic CV disease, and/or chronic kidney disease, a reduction of HF events was observed with use of oral semaglutide compared with placebo in those with a history of HF, without increasing the risk of serious adverse events. These data support the potential benefit of oral semaglutide in reducing HF events in people with T2D and HF. Trial Registration:ClinicalTrials.gov Identifier: NCT03914326.
Introduction Ambulatory Blood Pressure Monitoring (ABPM) offers advantages over conventional methods in hypertension diagnosis and monitoring; however, it is underutilised in primary care. This study explores the feasibility and acceptability of ABPM among primary care providers (PCPs) in Singapore. Methods The study used a qualitative research design. Between August 2022 to January 2024, eleven primary care physicians (PCPs) from two public primary care clinics in the Eastern region of Singapore were interviewed using a semi-structured topic guide through focus group discussions and in-depth interviews. The audio-recorded interviews were transcribed verbatim before thematic analysis. Results Three main themes emerged: utility of ABPM, challenges to implementing ABPM, and utility and usability of the ABPM reports. PCPs recognised the value of ABPM in optimising hypertension management in primary care, particularly in addressing diagnostic dilemmas, guiding medication titration, and facilitating patient counselling. They also emphasised the importance of a user-friendly ABPM report to support clinical decision-making. However, implementation challenges were identified, including challenges in explaining ABPM to patients due to limited time and resources, unfamiliarity with its workflow and uncertainty in follow-up action for ABPM reported hypertension phenotypes. Additionally, some PCPs stressed that ABPM should be applied selectively and is not appropriate for all patients with hypertension. Conclusion While PCPs recognised the benefits of ABPM for accurate hypertension diagnosis and clinical decision-making, significant barriers remain. Addressing concerns about limited time, resources, and workflow integration is crucial for the broader adoption of ABPM in primary care settings. While user-friendly reports are preferred, careful attention to their design and presentation is essential.
Human brain neuron activities are incredibly significant nowadays. Neuronal behavior is assessed by analyzing signal data such as extracellular recording, which can offer scientists valuable information about diseases and neuron activities. One of the difficulties researchers confront while evaluating these signals is the existence of large volumes of spike data. Spikes are significant components of signal data that can happen as a consequence of vital biomarkers or physical issues such as electrode movements. Hence, distinguishing types of spikes is essential. From this spot, the spike classification concept commences. Previously, researchers classified spikes manually. The manual classification was not precise enough, as it involved extensive analysis. Consequently, Artificial Intelligence (AI) was introduced into neuroscience to assist clinicians in classifying spikes correctly. Recognizing noises from spikes produced by neural activity causes the spike classification task to bear a significant demand. Classifying spikes accurately and quickly reveals the role of AI in the scope of spike classification. This review provides an in-depth discussion of the importance and use of AI in spike classification. This work organizes materials in the spike classification field for future studies and fully describes how spikes are recognized. Therefore, the existing datasets are described first. The topic of spike classification is then separated into three major components: preprocessing, classification, and evaluation. Each of these sections introduces existing methods and determines their importance. Having been summarized and compared, more efficient algorithms are highlighted. The primary goal of this work is to provide a perspective on spike classification for future research, as well as a thorough grasp of the methodologies and issues involved. In this work, numerous studies were extracted from various databases. The PRISMA-related research guidelines were then used to choose papers. Then, research studies based on spike classification using machine learning and deep learning approaches with effective preprocessing were selected. Although there are research papers on spike sorting using the keyword spike, the primary focus of this study is on spike classification. Finally, 47 papers were selected for in-depth review. First, useful information on the datasets for these papers is supplied. In addition, preprocessing approaches, classification methods, and ultimate performance are investigated in each of these studies. The material is then summarized. Furthermore, the fundamental concerns regarding spike classification raised in the opening of this paper are thoroughly addressed throughout the review. Our reviewing outcomes illustrate that support vector machine and clustering-based algorithms drastically influence machine learning methods in terms of high accuracy and many uses. Moreover, convolutional neural networks, spiky neural networks, and attention-based techniques can classify spikes with considerable functionality among deep learning methods. Various preprocessing and classification techniques have been used practically to classify extracted signal data from patients in medical institutions. Our review emphasizes the importance of classifying spikes in neuroscience applications with machine learning and deep learning models. This can provide precious insights and hands-on solutions for using AI to classify real-world medical data. This article is categorized under:
Background Accurate assessment of left ventricular ejection fraction (LVEF) is crucial for heart failure (HF) diagnosis but requires skilled sonographers. Artificial intelligence-enabled point-of-care (AI-POC) devices may enable novices to assess LVEF, potentially reducing healthcare costs. We conducted a cost-minimization analysis comparing conventional sonographer-performed echocardiography versus novice-operated AI-POC devices. Methods Using a decision tree model, we compared the costs of diagnosing LVEF <50% in patients with suspected heart failure across two pathways: novice-operated AI-POC devices versus standard transthoracic echocardiogram (TTE) performed by sonographers. The model incorporated LVEF <50% prevalence, diagnostic accuracy metrics, and comprehensive cost data for both approaches. We conducted a probabilistic sensitivity analysis to test the robustness of our findings under varying assumptions. Results The AI-POC pathway demonstrated substantial cost savings, averaging S$1185 [US$1422] per patient compared to S$1403 [US$1684] for conventional TTE. In a single tertiary referral centre in Singapore, implementing AI-POC devices for LVEF assessment in 100 patients resulted in savings of S$21 669 [US$26 013]. Probabilistic sensitivity analysis suggested a 99.9% probability that the AI-POC approach would be cost-saving compared to standard TTE. Conclusions This study provides economic evidence that task-shifting echocardiographic assessment of LVEF to novices using AI-POC devices is likely cost-saving compared to standard TTE. This task-shifting strategy offers a cost-saving alternative to conventional sonographer-led TTE.
