AI Accelerator with Ultralightweight Time-Period CNN-Based Model for Arrhythmia Classification.

This work proposes a classification system for arrhythmias, aiming to enhance the efficiency of the diagnostic process for cardiologists. The proposed algorithm includes a naive preprocessing procedure for electrocardiography (ECG) data applicable to various ECG databases. Additionally, this work proposes an ultralightweight model for arrhythmia classification based on a convolutional neural network and incorporating R-peak interval features to represent long-term rhythm information, thereby improving the model's classification performance. The proposed model is trained and tested by using the MIT-BIH and NCKU-CBIC databases in accordance with the classification standards of the Association for the Advancement of Medical Instrumentation (AAMI), achieving high accuracies of 98.32% and 97.1%. This work applies the arrhythmia classification algorithm to a web-based system, thus providing a graphical interface. The cloud-based execution of automated artificial intelligence (AI) classification allows cardiologists and patients to view ECG wave conditions instantly, thereby remarkably enhancing the quality of medical examination. This work also designs a customized integrated circuit for the hardware implementation of an AI accelerator. The accelerator utilizes a parallelized processing element array architecture to perform convolution and fully connected layer operations. It introduces proposed hybrid stationary techniques, combining input and weight stationary modes to increase data reuse drastically and reduce hardware execution cycles and power consumption, ultimately achieving high-performance computing. This accelerator is implemented in the form of a chip by using the TSMC 180 nm CMOS process. It exhibits a power consumption of 122 µW, a classification latency of 6.8 ms, and an energy efficiency of 0.83 µJ/classification.

Associations of urinary fetuin-A with histopathology and kidney events in biopsy-proven kidney disease.

Background: Fetuin-A is implicated in the pathogenesis of vascular calcification in chronic kidney disease (CKD); however, the relationship between fetuin-A, histopathologic lesions and long-term kidney outcomes in patients with various types of kidney disease remains unclear. Methods: We measured urinary fetuin-A levels in 335 individuals undergoing clinically indicated native kidney biopsy. The expressions of fetuin-A mRNA and protein in the kidney were assessed using RNA sequencing and immunohistochemistry. The association of urinary fetuin-A with histopathologic lesions and major adverse kidney events (MAKE), defined as a decline in estimated glomerular filtration rate (eGFR) of at least 40%, kidney failure or death, was analyzed. Results: Urinary fetuin-A levels showed a positive correlation with albuminuria (rs = 0.67, P < .001) and a negative correlation with eGFR (rs = -0.46, P < .001). After multivariate adjustment, higher urinary fetuin-A levels were associated with glomerular inflammation, mesangial expansion, interstitial fibrosis and tubular atrophy, and arteriolar sclerosis. Using a 1 transcript per million gene expression cutoff, we found kidney fetuin-A mRNA levels below the threshold in both individuals with normal kidney function and those with CKD. Additionally, immunohistochemistry revealed reduced fetuin-A staining in tubular cells of CKD patients compared with normal controls. During a median 21-month follow-up, 115 patients experienced MAKE, and Cox regression analysis confirmed a significant association between elevated urinary fetuin-A and MAKE. This association remained significant after adjusting for potential confounding factors. Conclusion: Urinary fetuin-A is associated with chronic histological damage and adverse clinical outcomes across a spectrum of biopsy-proven kidney diseases.

Practice Algorithm of Rotational Thromboelastometry-Guided (ROTEM-Guided) Bleeding Management in Liver Transplantation.

Liver transplantation (LT) is frequently complicated by coagulopathy associated with end-stage liver disease, which is often multifactorial and associated with hemostatic disturbances affecting both the procoagulant and anticoagulant systems. This rebalanced coagulation system may lead to bleeding diathesis or increased clot formation. Conventional coagulation tests cannot reflect these complex changes because they can only illustrate deficiencies in the procoagulant system. Viscoelastic tests such as rotational thromboelastometry (ROTEM) have been used in LT and have shown useful for detecting coagulopathy and guiding transfusions. Implementation of ROTEM-guided bleeding management algorithms has proven effectiveness in reducing bleeding, transfusion needs, complication rates, and healthcare costs in LT. This document is intended to provide a practice algorithm for the management of major bleeding and coagulopathy during LT and to encourage adaptation of the guidelines to individual institutional circumstances and resources.

Propofol-based total intravenous anesthesia is associated with less postoperative recurrence than desflurane anesthesia in thyroid cancer surgery.

BACKGROUND: The effects of anesthesia in patients undergoing thyroid cancer surgery are still not known. We investigated the relationship between the type of anesthesia and patient outcomes following elective thyroid cancer surgery. METHODS: This was a retrospective cohort study of patients who underwent elective surgical resection for papillary thyroid carcinoma between January 2009 and December 2019. Patients were grouped according to the type of anesthesia they received, desflurane or propofol. A Kaplan-Meier analysis was conducted, and survival/recurrence curves were presented from the date of surgery to death/recurrence. Univariable and multivariable Cox regression models were used to compare hazard ratios for recurrence after propensity matching. RESULTS: A total of 621 patients (22 deaths, 3.5%) under desflurane anesthesia and 588 patients (32 deaths, 5.4%) under propofol anesthesia were included. Five hundred and eighty-eight patients remained in each group after propensity matching. Propofol anesthesia was not associated with better survival compared to desflurane anesthesia in the matched analysis (P = 0.086). However, propofol anesthesia was associated with less recurrence (hazard ratio, 0.38; 95% confidence interval, 0.25-0.56; P < 0.001) in the matched analysis. CONCLUSIONS: Propofol anesthesia was associated with less recurrence, but not mortality, following surgery for papillary thyroid carcinoma than desflurane anesthesia. Further prospective investigation is needed to examine the influence of propofol anesthesia on patient outcomes following thyroid cancer surgery.

Plasma SCUBE2 as a novel biomarker associates with survival outcomes in patients with sepsis-associated acute kidney injury.

BACKGROUND: The adverse effects of sepsis-associated acute kidney injury (SA-AKI) highlight the need for new biomarkers. Signal Peptide-Complement C1r/C1s, Uegf, Bmp1-Epidermal Growth Factor-like Domain-Containing Protein 2 (SCUBE2), important for angiogenesis and endothelial integrity, has been linked to increased mortality in models of lipopolysaccharide-induced lung injury. This research aimed to assess the utility of plasma SCUBE2 levels as a prognostic indicator for SA-AKI in intensive care unit (ICU) patients. METHODS: Between September 2020 and December 2022, our study enrolled ICU patients diagnosed with stage 3 SA-AKI. We collected demographic information, illness severity indices, and laboratory data, including plasma SCUBE2 and sepsis-triggered cytokine levels. We employed receiver operating characteristic curves and DeLong tests to assess the predictive accuracy for survival, Kaplan-Meier curves to evaluate the relative risk of death, and multivariate logistic regression to identify independent mortality predictors. RESULTS: Among the total of 200 participants, the survivors had significantly higher plasma SCUBE2 levels (115.9 ng/mL) compared to those who died (35.6 ng/mL). SCUBE2 levels showed a positive correlation with the anti-inflammatory cytokine IL-10 and a negative correlation with the APACHE II score, SOFA score, C-reactive protein, and monocyte chemoattractant protein-1. Multivariate analysis revealed that elevated SCUBE2 and IL-10 levels were independently protective against mortality, and associated with the most favorable 30-day survival outcomes. CONCLUSIONS: In ICU patients with stage 3 SA-AKI, lower plasma levels of SCUBE2 were correlated with elevated pro-inflammatory factors, which impacted survival outcomes. This suggests that SCUBE2 could be a potential biomarker for predicting prognosis in patients with SA-AKI.