The electrocardiogram (ECG) is a commonly used technique for detecting arrhythmias and many other cardiac diseases. Automatic ECG diagnosis has seen tremendous success in recent years, owing to the rapid development of the deep learning (DL) approach. Existing works on automatic ECG diagnosis can be divided roughly into two categories: prediction at the rhythm level from an ECG record, and prediction at the heartbeat level, although their relationship was seldom studied previously. In this paper, we address the following question: can we train an abnormal heartbeat detection model using solely data annotated at the rhythm level? We first used multiple instance learning (MIL) to model the relationship between an ECG record (whose label is given at the rhythm level and is provided as an input) and the heartbeats in the ECG (whose labels are to be predicted). Then, we sequentially trained two models, a rhythm model for detecting abnormal heartbeats in an ECG record labeled as arrhythmia, and a heartbeat model for classifying heartbeats as normal or various types of arrhythmias. We trained and tested our models using 61,853 ECG records with rhythm annotations. The experimental results demonstrate that the heartbeat model achieves a macro-average F1 score of 0.807 in classifying four types of arrhythmias as well as normal heartbeats. Our model significantly outperforms the model directly trained with 15,385 ECG heartbeats with heartbeat annotations, demonstrating the viability of our strategy for training a high-performing heartbeat-level automatic diagnostic model using only rhythm annotation.

Aims. Activins A and B, and their binding protein, follistatin, regulate glucose metabolism and inflammation. Consequently, their role in type 2 diabetes (T2D) was examined.Methods. Blood was taken from fasted participants (34 males; 58 females; 50–75 years) with diabetes or during an oral glucose tolerance test (OGTT). Clinical parameters were assessed, and blood assayed for activins, follistatin, and C-reactive protein.Results. Serum levels of activin A (93.3 ± 27.0 pg/mL, mean ± SD), B (81.8 ± 30.8 pg/mL), or follistatin (6.52 ± 3.15 ng/mL) were not different (P>0.05) between subjects with normal OGTT (n=39), impaired glucose tolerance and/or fasting glucose (n=17), or T2D (n=36). However, activin A and/or activin B were positively correlated with parameters of insulin resistance and T2D, including fasting glucose (P<0.001), fasting insulin (P=0.02), glycated hemoglobin (P=0.003), and homeostasis model assessment of insulin resistance (HOMA-IR;P<0.001). Follistatin was positively correlated with HOMA-IR alone (P=0.01).Conclusions. These data indicate that serum measurements of activin A, B, or follistatin cannot discriminate risk for T2D in individual patients, but the activins display a positive relationship with clinical parameters of the disease.

Acute myocardial infarction can be treated aggressively with intravenous thrombolysis, percutaneous coronary intervention, and coronary artery bypass grafting; however, recanalization can cause myocardial ischemia-reperfusion injury (MIRI). This is an important reason that restricts the treatment effect of patients. After the ischemic myocardium is restored to perfusion, an inflammatory response can occur within minutes and peak within a few days. Many pro-inflammatory cytokines can seriously damage cardiac function. Inflammation can regulate cardiomyocyte apoptosis, autophagy, pyroptosis, and necrosis, and is the main initiating factor leading to MIRI in cardiomyocytes. This article reviews the mechanism of inflammatory response in the ischemia-reperfusion period after acute myocardial infarction and the clinical value and application prospect of inhibiting inflammatory response in the treatment of acute myocardial infarction.