Search

BackgroundThe detection of dyskalemias—hypokalemia and hyperkalemia—currently depends on laboratory tests. Since cardiac tissue is very sensitive to dyskalemia, electrocardiography (ECG) may be able to uncover clinically important dyskalemias before laboratory results. ObjectiveOur study aimed to develop a deep-learning model, ECG12Net, to detect dyskalemias based on ECG presentations and to evaluate the logic and performance of this model. MethodsSpanning from May 2011 to December 2016, 66,321 ECG records with corresponding serum potassium (K+) concentrations were obtained from 40,180 patients admitted to the emergency department. ECG12Net is an 82-layer convolutional neural network that estimates serum K+ concentration. Six clinicians—three emergency physicians and three cardiologists—participated in human-machine competition. Sensitivity, specificity, and balance accuracy were used to evaluate the performance of ECG12Net with that of these physicians. ResultsIn a human-machine competition including 300 ECGs of different serum K+ concentrations, the area under the curve for detecting hypokalemia and hyperkalemia with ECG12Net was 0.926 and 0.958, respectively, which was significantly better than that of our best clinicians. Moreover, in detecting hypokalemia and hyperkalemia, the sensitivities were 96.7% and 83.3%, respectively, and the specificities were 93.3% and 97.8%, respectively. In a test set including 13,222 ECGs, ECG12Net had a similar performance in terms of sensitivity for severe hypokalemia (95.6%) and severe hyperkalemia (84.5%), with a mean absolute error of 0.531. The specificities for detecting hypokalemia and hyperkalemia were 81.6% and 96.0%, respectively. ConclusionsA deep-learning model based on a 12-lead ECG may help physicians promptly recognize severe dyskalemias and thereby potentially reduce cardiac events.

Federated learning (FL) is a method used for training artificial intelligence models with data from multiple sources while maintaining data anonymity, thus removing many barriers to data sharing. Here we used data from 20 institutes across the globe to train a FL model, called EXAM (electronic medical record (EMR) chest X-ray AI model), that predicts the future oxygen requirements of symptomatic patients with COVID-19 using inputs of vital signs, laboratory data and chest X-rays. EXAM achieved an average area under the curve (AUC) >0.92 for predicting outcomes at 24 and 72 h from the time of initial presentation to the emergency room, and it provided 16% improvement in average AUC measured across all participating sites and an average increase in generalizability of 38% when compared with models trained at a single site using that site's data. For prediction of mechanical ventilation treatment or death at 24 h at the largest independent test site, EXAM achieved a sensitivity of 0.950 and specificity of 0.882. In this study, FL facilitated rapid data science collaboration without data exchange and generated a model that generalized across heterogeneous, unharmonized datasets for prediction of clinical outcomes in patients with COVID-19, setting the stage for the broader use of FL in healthcare.

Background: Demonstrating a positive impact of educational programs on clinical care outcomes is challenging. We assess students' knowledge, attitudes, and behaviors as short-term outcomes of integrating a structured Traditional Chinese Medicine course within a Western medical school. Methods: A prospective questionnaire survey was conducted among first-year and second-year undergraduate medical students who participated in the "Introduction to Chinese Medicine" course in 2020. The survey assessed their knowledge, attitudes, and behaviors before and after completing the 32-hour course. Results: In total, 89 participants who completed both pre- and post-course questionnaires exhibited significantly higher scores in Chinese medicine knowledge after the course (3.82 ± 0.67 vs. 4.05 ± 0.73, p = 0.002). The majority of medical students displayed positive attitudes towards traditional Chinese medicine (80.9% vs. 83.3%, p = 0.392), particularly regarding the statement, "Traditional Chinese medicine emphasizes holistic considerations and whole person health to a greater extent than Western medicine does" (75.3% vs. 88.8%, p = 0.017). Additionally, most medical students reported proactive behaviors related to integrative care with Chinese medicine, including inquiring about Chinese medicine therapy history (76.4%), the concurrent use of Western and Chinese medications (78.7%) and participating in academic exchanges with Chinese medicine physicians (78.7%) after the course. More medical students expressed an intention to proactively learn more about Chinese medicine after the course (40.4% vs. 51.7%, p = 0.031). Conclusions: The elective "Introduction to Chinese Medicine" course enhanced Taiwanese Western medical students' knowledge of Chinese medicine, fostered positive attitudes towards integrative healthcare involving Chinese medicine, and increased their willingness to learn more about Chinese medicine for future clinical practice. Trial registration: Registration prior to patient enrollment was not applicable because participants were not assigned to treatment groups in this study. [ABSTRACT FROM AUTHOR]

Primary cardiac angiosarcoma is a rare and aggressive malignancy with a poor prognosis. This case report highlights the challenges in diagnosing and managing such tumors, especially when complicated by cardiac tamponade. We present the case of a 52-year-old male with primary cardiac angiosarcoma, initially presenting with cardiac tamponade, which led to the subsequent diagnosis of an aggressive tumor. Prompt recognition and interdisciplinary collaboration facilitated timely intervention, resulting in successful surgical resection and favorable short-term outcomes. However, the overall prognosis remains poor due to the aggressive nature of the disease. This case underscores the importance of early recognition, comprehensive diagnostic evaluation, and aggressive multimodal treatment strategies. [ABSTRACT FROM AUTHOR]

Partial anomalous pulmonary venous connection (PAPVC) is a relatively rare congenital heart anomaly characterized by abnormal drainage of one or more pulmonary veins into a location other than the left atrium. In this case report, we present a 70-year-old male diagnosed with PAPVC with a vertical vein connected between the left upper pulmonary vein and the innominate vein. The transcatheter occlusion of the vertical vein was successfully achieved using a vascular plug. The report underscores transcatheter occlusion as a minimally invasive and effective alternative to surgical correction for PAPVC; it may provide advantages such as a shorter hospital stay, faster recovery time, and reduced complications. [ABSTRACT FROM AUTHOR]

We found that GroEL in Porphyromonas gingivalis accelerated tumor growth and increased mortality in tumor‐bearing mice; GroEL promoted proangiogenic function, which may be the reason for promoting tumor growth. To understand the regulatory mechanisms by which GroEL increases the proangiogenic function of endothelial progenitor cells (EPCs), we explored in this study. In EPCs, MTT assay, wound‐healing assay, and tube formation assay were performed to analyze its activity. Western blot and immunoprecipitation were used to study the protein expression along with next‐generation sequencing for miRNA expression. Finally, a murine tumorigenesis animal model was used to confirm the results of in vitro. The results indicated that thrombomodulin (TM) direct interacts with PI3K/Akt to inhibit the activation of signaling pathways. When the expression of TM is decreased by GroEL stimulation, molecules in the PI3K/Akt signaling axis are released and activated, resulting in increased migration and tube formation of EPCs. In addition, GroEL inhibits TM mRNA expression by activating miR‐1248, miR‐1291, and miR‐5701. Losing the functions of miR‐1248, miR‐1291, and miR‐5701 can effectively alleviate the GroEL‐induced decrease in TM protein levels and inhibit the proangiogenic abilities of EPCs. These results were also confirmed in animal experiments. In conclusion, the intracellular domain of the TM of EPCs plays a negative regulatory role in the proangiogenic capabilities of EPCs, mainly through direct interaction between TM and PI3K/Akt to inhibit the activation of signaling pathways. The effects of GroEL on tumor growth can be reduced by inhibiting the proangiogenic properties of EPCs through the inhibition of the expression of specific miRNAs. [ABSTRACT FROM AUTHOR]