Data Augmentation of Wrist Pulse Signal for Traditional Chinese Medicine Using Wasserstein GAN

With the continuous development of human life and society, the medical field is constantly improving. However, modern medicine still faces many limitations, including challenging and previously unsolvable problems. In these cases, artificial intelligence (AI) can provide solutions. The research and application of generative adversarial networks (GAN) are a clear example. While most researchers focus on image augmentation, there are few one-dimensional data augmentation examples. The radiomics feature extracted from RT and CT images is one-dimensional data. As far as we know, we are the first to apply the WGAN-GP algorithm to generate radiomics data in the medical field. In this paper, we input a portion of the original real data samples into the model. The model learns the distribution of the input data samples and generates synthetic data samples with similar distribution to the original real data, which can solve the problem of obtaining annotated medical data samples. We have conducted experiments on the public dataset Heart Disease Cleveland and the private dataset. Compared with the traditional method of Synthetic Minority Oversampling Technique (SMOTE) and common GAN for data augmentation, our method has significantly improved the AUC and SEN values under different data proportions. At the same time, our method has also shown varying levels of improvement in ACC and SPE values. This demonstrates that our method is effective and feasible.

show abstract

GAN-based one dimensional medical data augmentation

Zhang

Wang

Zhang

et al. 2023

Soft Comput

View full text Add to dashboard Cite

show abstract

A review of traditional Chinese medicine diagnosis using machine learning: Inspection, auscultation-olfaction, inquiry, and palpation

Tian,

Chen,

et al. 2024

Computers in Biology and Medicine

View full text Add to dashboard Cite

Toward Multi-area Contactless Museum Visitor Counting with Commodity WiFi

Jiang

Xia

Feng

2023

J. Comput. Cult. Herit.

View full text Add to dashboard Cite

Multi-area visitor counting plays a critical role in museum management, which can help administrative staff better study visitor flows and hotspots, so that can ensure the quality and safety of visits. Internet of Things (IoT) techniques facilitate efficient recording and understanding of visitors’ spatial and temporal distribution in museums, and traditional visitor tracking applications use surveillance cameras or wireless connections with portable smart devices. However, these methods either involve privacy concerns or face the obstacle of getting natural behavioral data of all visitors. This paper explores an IoT monitoring methodology in the field of museum studies, proposing a commodity WiFi-based head counting framework that does not need the visitor to connect with any device. Our system analyzes the Channel State Information (CSI) amplitude fluctuations at the fixed receiver caused when visitors cross the line-of-sight (LoS) link. It enables multi-area visitor counting by achieving In-and-Out traffic detection at different sites with a convolutional neural network (CNN) algorithm. The method also allows for a rough classification of visitor types based on body size, and an extra transfer module is presented to reduce training time for increasing sensing scenarios. Over 2300 samples at 5 different sites were collected to test the usability. Experiment 1 implemented in 3 environments/deployments demonstrated that the proposed approach can be potentially implemented in variable sites of museums. It achieved high up to 95% and 99% accuracies for identifying the number and direction of people crossing, respectively. Experiment 2 sampled adults, children, and adult-child groups at a science museum and achieved approximately 89% classification accuracy of visitor types. Experiment 3 collected data for all cases in which up to 3 targets entered and exited simultaneously, and reached a recognition accuracy of around 88% for 9 different cases. The potential and limitations for the practical application of wireless contactless sensing to cultural spaces are discussed.

show abstract

scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.

Contact Info

customersupport@researchsolutions.com

10624 S. Eastern Ave., Ste. A-614

Henderson, NV 89052, USA

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Blog Terms and Conditions API Terms Privacy Policy Contact Cookie Preferences Do Not Sell or Share My Personal Information

Made with 💙 for researchers

Part of the Research Solutions Family.

Data Augmentation of Wrist Pulse Signal for Traditional Chinese Medicine Using Wasserstein GAN

Cited by 3 publications

References 7 publications

GAN-based one dimensional medical data augmentation

GAN-based one dimensional medical data augmentation

A review of traditional Chinese medicine diagnosis using machine learning: Inspection, auscultation-olfaction, inquiry, and palpation

Toward Multi-area Contactless Museum Visitor Counting with Commodity WiFi

Contact Info

Product

Resources

About