IEEE Consumer Technology Society の技術協賛を受けて、組込みシステムとIoTのトピックについての査読論文の発表を行います。
01:35 pm: Arrhythmia Classification Using EFFICIENTNET-V2 with 2-D Scalogram Image Representation
Muhammad Furqon (Sepuluh Nopember Institute of Technology Surabaya, Indonesia); Supeno Mardi Susiki Nugroho (Sepuluh Nopember Institute Of Technology, Indonesia); Reza Fuad Rachmadi (Institut Teknologi Sepuluh Nopember, Indonesia); Arief Kurniawan (Institut Teknologi Sepuluh Nopember, Indonesia); I Ketut Eddy Purnama (Institut Teknologi Sepuluh Nopember, Indonesia); Mpu Aji (Sepuluh Nopember Institute of Technology Surabaya, Indonesia)
Abstract: Cardiovascular disease is part of global death's main cause. It is the term for all types of diseases that affect the heart or blood vessels. Heart Disease is a type of cardiovascular disease. It's can detect early by examining the arrhythmia presence. Arrhythmia is an abnormal heart rhythm which commonly diagnoses and evaluate by analyzing an electrocardiogram (ECG) signal. In classical techniques, a cardiologist/ clinician used an ECG to monitor the patient's heart rate and rhythm and read the patient's activity journal to diagnose arrhythmias and to develop appropriate treatment plans. The use of an ECG, on the other hand, takes time and effort. The development of arrhythmias diagnoses, toward computational processes, such as detecting and classifying using machine learning and deep learning. A convolutional neural network (CNN) is a popular method used to classify arrhythmia. Dataset pre-processing was also considered to achieve the best performance models. Our study used the EfficientNet-V2 which is a type of convolutional neural network to perform the classification of five arrhythmias. In pre-processing, the ECG signal was cut each 1 second (360 data), augmentation is applied to balance the variation of the dataset, and Continues Wavelet Transform (CWT) is employed to transform the ECG signal into a scalogram. The dataset is then distributed with a modulus algorithm to get variety in each set of datasets. The color map is applied to convert scalograms into RGB images. By this scheme, our study achieved superior accuracy than the existing method, with an accuracy rate of 99.97%.
01:55 pm: Harnessing IoT Technology for the Development of Wearable Contact Tracing Solutions
Rex Acharya (Sam Houston State University, USA); Amar A Rasheed (Sam Houston State University, USA); Hacer Varol (Stephen F. Austin State University, USA); Mohamed Baza (College of Charleston, USA); Louanne Mozer Sallo (Sam Houston State University, USA); Rabi Mahapatr (Texas A&M University, USA)
Abstract: Advancements in mobile computing and embedded system technologies show superiority in controlling the transmission of a virus during the COVID-19 pandemic. They provide rapid contact data compared to manual contact tracing and medical monitoring methodologies. Data tracing capabilities were achieved by existing technologies via the utilization of smartphone-based exposure notifications systems and proximity sensing tools based on Bluetooth/GPS. Such systems lack user privacy and data anonymization capabilities, it also requires that users have continuous access to smartphones. This paper proposes the development of a lightweight wearable prototype for contact tracing. The proposed system is based on the deployment of an IoT development board, incorporated with a Bluetooth chipset for proximity sensing. To overcome the problem of user-to-smartphone accessibility, the processing of contact data by the exposure notification system has been migrated from mobile computing to a contact tracer management server that is managed by the community. Daily contact exposures to the virus are recorded by each user's contact tracing system and stored in the user's blockchain. Data integrity and immunity against data injection attacks were achieved via the implementation of lightweight block chaining and validation schemes. Data anonymization was also supported via the utilization of the onboard AES crypto engine. Meanwhile, support for user anonymity was incorporated into the proposed system through the implementation of an anonymous authentication protocol based on the Zero-Knowledge Proof (ZKP) approach. Finally, we have analyzed the performance of the system in terms of power consumption under various systems settings, such as encryption key size and the required number of ZKP validation attempts per authentication session.
How to participate: Those who wish to participate in the paper presentation session, please contact
with your name and e-mail address with the subject line that contains ‘attend’ without quotes.
The application will be accepted until 10:30 December 9th.
Details about teleconferencing software and URL will be sent to the e-mail address.
→Proceedings Available(one in
zipped archive, the other is
concatenated PDF)
Subject 行に attend という言葉を入れて
にメイルを送ってください。
折りかえしアクセス方法を本セッションの開催までにお送りします。
事前申込は12月9日10:30 (JST)まで受け付けます。