Cardiac Arrhythmia Classification Using Advanced Deep Learning Techniques on Digitized ECG Datasets.

ECG classification or heartbeat classification is an extremely valuable tool in cardiology. Deep learning-based techniques for the analysis of ECG signals assist human experts in the timely diagnosis of cardiac diseases and help save precious lives. This research aims at digitizing a dataset of images of ECG records into time series signals and then applying deep learning (DL) techniques on the digitized dataset. State-of-the-art DL techniques are proposed for the classification of the ECG signals into different cardiac classes. Multiple DL models, including a convolutional neural network (CNN), a long short-term memory (LSTM) network, and a self-supervised learning (SSL)-based model using autoencoders are explored and compared in this study. The models are trained on the dataset generated from ECG plots of patients from various healthcare institutes in Pakistan. First, the ECG images are digitized, segmenting the lead II heartbeats, and then the digitized signals are passed to the proposed deep learning models for classification. Among the different DL models used in this study, the proposed CNN model achieves the highest accuracy of ∼92%. The proposed model is highly accurate and provides fast inference for real-time and direct monitoring of ECG signals that are captured from the electrodes (sensors) placed on different parts of the body. Using the digitized form of ECG signals instead of images for the classification of cardiac arrhythmia allows cardiologists to utilize DL models directly on ECG signals from an ECG machine for the real-time and accurate monitoring of ECGs.

Risk Factors of Cardiovascular and Cerebrovascular Events in Patients With Uraemia Complicated With Hypertension During Maintenance Haemodialysis Treatment.

INTRODUCTION: This study aimed to investigate the risk factors associated with major adverse cardiovascular (group of events that affect heart and blood vessels) and cerebrovascular (events affecting blood vessels supplying the brain) events (MACCE) in patients with uraemia complicated with hypertension who required maintenance haemodialysis (MHD) treatment. METHODOLOGY: Clinical data and laboratory indicators of 156 uraemia patients complicated with hypertension were collected and retrospectively analysed. The patients were admitted to a tertiary care hospital (Abbas Institute of Medical Sciences AIMS) in Muzaffarabad, Pakistan, from February 2018 to February 2022. The data was collected through consecutive sampling and patients were recruited after following the inclusion and exclusion criteria. RESULTS: Eighty-one out of 156 patients were not complicated with MACCE, and 75 patients were complicated with MACCE during the MHD treatment cycle, with an incidence of 48.08%. Compared to the non-MACCE group, the MACCE group's diabetes, body mass growth rate, triglyceride (TG), NT-proBNP, standard deviation and coefficient of variance for systolic and diastolic blood pressure (SBP-SD, SBP-CV, DBP-SD, and DBP-CV) showed significant differences (P<0.05) between the groups. Diabetes, body mass growth rate, TG, NT-proBNP, SBP-SD, SBP-CV, DBP-SD, and DBP-CV with odds ratios of 3.074, 3.202, 2.188, 2.512, 2.357, 2.431, 2.299, and 2.062 respectively were risk factors for MACCE in uraemia patients with hypertension. CONCLUSION:  From the results of this study, we inferred that patients with uraemia and hypertension complicated by MACCE in the treatment cycle of MHD were related to diabetes, body mass growth rate, TG, NT-proBNP, SBP-SD, SBP-CV, DBP-SD, and DBP-CV.

Unveiling the Knowledge Gap Regarding Standard Antibiotic Prophylaxis for Infective Endocarditis Among Physicians, Dentists and Cardiologists in Pakistan: A Cross-Sectional Study.

INTRODUCTION: Infective endocarditis is a microbial infection of the endocardial surface of the native and prosthetic valves. This study aimed to evaluate knowledge regarding the current guidelines for antibiotic prophylaxis for infective endocarditis. METHODS: This cross-sectional study was conducted among physicians, cardiologists, and dentists in Rawalpindi and Islamabad. The questionnaire was distributed as Google Forms among the required population, and responses were collected on a Google Response Sheet. RESULTS: The participants viewed rheumatic heart disease (83.7%) and heart transplant (96.7%) as the most vulnerable conditions that warrant the need for antibiotic prophylaxis. The other questions yielded average responses. CONCLUSION: The findings of this study emphasize the importance of adhering to standard guidelines and highlight the need for knowledge of the current guidelines.

Brain natriuretic peptide in acute heart failure and its association with glomerular filtration rate: A systematic review and meta-analysis.

BACKGROUND: Acute heart failure (AHF) is one of the most common cardiovascular diseases. Early diagnosis and prognosis are essential, as they can eventually lead to a fatal condition. Recently, brain natriuretic peptide (BNP) has been recognized as one of the most popular biomarkers for AHF. Changes in glomerular filtration rate (GFR) are often observed in AHF. METHODS: We searched PubMed, Google Scholar, and ScienceDirect between March and June 2023. Original case control studies written in English that assessed levels oh BNP in AHF were included. Systematic reviews, letters to editor, correspondence, comprehensive reviews, and duplicated studies were excluded. Funnel plots were constructed to assess publication bias. RESULTS: A total of 9 studies were selected and we obtained the mean difference (MD) of BNP level to be 2.57 (95% CI: 1.35, 3.78), and GFR to be -15.52, (95% CI: -23.35, -7.70) in AHF patients. Sensitivity analyses supported the robustness of the outcome. CONCLUSION: Results indicated that BNP was a promising prognostic biomarker of AHF, whereas GFR was found to be negatively correlated with AHF.

Primary Hyperparathyroidism: To Evaluate Benefit of Ultrasound and Tc99m-SESTAMIBI Scan in Localizing Abnormal Parathyroid Gland Before Surgery, in a Secondary Care Hospital.

Introduction Primary hyperparathyroidism is a common endocrine condition requiring parathyroidectomy for curative management. Localization of parathyroid gland by ultrasound and Tc99m-SESTAMIBI is important to opt for less invasive and comparatively lower complication risk surgery minimal invasive parathyroidectomy (MIP) instead of four-gland exploration surgery. Aim To evaluate ultrasound and Tc99m-SESTAMIBI in localization of abnormal parathyroid gland before surgery. Method and materials All patients of primary hyperparathyroidism (PHPT) that presented to a secondary care hospital (endocrinology department) from 2015-2019 were recruited retrospectively from electronic fusion system of hospital. Results of ultrasound parathyroid and Tc99m-SESTAMIBI done for localization of abnormal parathyroid gland were analyzed. Results Total PHPT patients recruited were 59, mean age 64.2 years, male 11 (18.64%) and female 48 (81.3%). Ultrasound parathyroid was done in 44 patients, Tc 99m-SESTAMIBI was done in 31, both tests were done in 31 patients. Combined concordant adenoma in both tests was seen in 11 (35%) cases which can opt for minimal invasive parathyroidectomy (MIP) with confidence whereas 65% of cases would require either four-gland exploration or further testing like single-photon emission computed tomography-computed tomography (SPECT-CT) or intraoperative parathyroid hormone measurement to opt for MIP. Conclusion Combined ultrasound parathyroid and Tc 99m-SESTAMIBI scan was useful in localization of parathyroid adenoma in 11 (35%) patients that can opt for MIP which is a lower complication risk surgery whereas 20 (65%) patients would need further investigation with SPECT-CT or intraoperative parathyroid hormone measurement or four-gland exploration surgery. Recommendation Third modality of investigation such as SPECT-CT or intraoperative parathyroid hormone measurement needs evaluation so that more patients can benefit from MIP instead of four-gland exploration surgery.