Background: Electrocardiographic (ECG) monitors have been widely used for diagnosing cardiac arrhythmias for decades. However, accurate analysis of ECG signals is difficult and time-consuming work because large amounts of beats need to be inspected. In order to enhance ECG beat classification, machine learning and deep learning methods have been studied. However, existing studies have limitations in model rigidity, model complexity, and inference speed. Objective: To classify ECG beats effectively and efficiently, we propose a baseline model with recurrent neural networks (RNNs). Furthermore, we also propose a lightweight model with fused RNN for speeding up the prediction time on central processing units (CPUs). Methods: We used 48 ECGs from the MIT-BIH (Massachusetts Institute of Technology-Beth Israel Hospital) Arrhythmia Database, and 76 ECGs were collected with S-Patch devices developed by Samsung SDS. We developed both baseline and lightweight models on the MXNet framework. We trained both models on graphics processing units and measured both models’ inference times on CPUs. Results: Our models achieved overall beat classification accuracies of 99.72% for the baseline model with RNN and 99.80% for the lightweight model with fused RNN. Moreover, our lightweight model reduced the inference time on CPUs without any loss of accuracy. The inference time for the lightweight model for 24-hour ECGs was 3 minutes, which is 5 times faster than the baseline model. Conclusions: Both our baseline and lightweight models achieved cardiologist-level accuracies. Furthermore, our lightweight model is competitive on CPU-based wearable hardware.
Expanding waste-to-energy (WtE) facilities is difficult, and with tightening incineration regulations, improvements in WtE facility operations are required to dispose of waste that is increasing by an average of 4.8% annually. To achieve this, an intelligent combustion control (ICC) system was studied using digital technologies such as the Internet of Things and artificial intelligence to improve the operation of WtE facilities. The ICC system in this study is composed of three modules: perception, decision, and control. Perception: collecting and visualizing digital data on the operating status of WtE facilities; Decision: using AI to propose optimal operation methods; Control: automatically controlling the WtE facility according to the AI-suggested optimization methods. The ICC system was applied to the “G” WtE facility, a solid waste WtE facility operating in Gyeonggi province, Republic of Korea, and the digital data collected over six months showed high quality, with low delay and a data loss rate of only 0.12%. Additionally, in January 2024, the ICC system was used to automatically control the second forced draft fan and induced draft fan over a four-day period. As a result, the incinerator flue gas temperature decreased by 0.66%, steam flow rate improved by 2.41%, power generation increased by 3.09%, CO emissions were reduced by 60.72%, and NOx emissions decreased by 7.33%. Future research will expand the ICC system to include the automatic control of the first forced draft fan and the operation time of the stoker.
Many existing studies on knowledge distillation have focused on methods in which a student model mimics a teacher model well. Simply imitating the teacher's knowledge, however, is not sufficient for the student to surpass that of the teacher. We explore a method to harness the knowledge of other students to complement the knowledge of the teacher. We propose deep collective knowledge distillation for model compression, called DCKD, which is a method for training student models with rich information to acquire knowledge from not only their teacher model but also other student models. The knowledge collected from several student models consists of a wealth of information about the correlation between classes. Our DCKD considers how to increase the correlation knowledge of classes during training. Our novel method enables us to create better performing student models for collecting knowledge. This simple yet powerful method achieves state-of-the-art performances in many experiments. For example, for ImageNet, ResNet18 trained with DCKD achieves 72.27\%, which outperforms the pretrained ResNet18 by 2.52\%. For CIFAR-100, the student model of ShuffleNetV1 with DCKD achieves 6.55\% higher top-1 accuracy than the pretrained ShuffleNetV1.