Qihang Wang
ABSTRACT
Combining deep learning-based methods and key posture sensor data can improve the accuracy of automatically identifying basic acupuncture techniques. This study aims to propose a novel deep learning-based method to identify the time characteristics of the physical parameters of acupuncture techniques, including twisting-supplementing (TS), twisting-draining (TD), level-supplementing level-draining (LSLD), lifting-inserting-supplementing (LIS), and lifting-inserting-draining (LID) methods. During the acupuncture process, six-axis posture sensors collect parameters such as lifting and inserting speed, twisting frequency, and angular rotation rate, and analyze the mapping relationship between physical parameters and different techniques. By using multi-dimensional signal control decision algorithms, the performance of our method is enhanced, addressing the issue of poor recognition based on single-dimensional signals. Experimental results demonstrate that our method achieves high accuracy in identifying five techniques. Our proposed method can effectively identify acupuncture techniques, making it useful for the evaluation and teaching of acupuncture techniques, facilitating the inheritance of traditional Chinese acupuncture techniques.
- Lewei Tang, 2014. Discussion on evolution and theoretical value of needling manipulation. Journal of Traditional Chinese Medicine 55, 9, 728-731.Google Scholar
- Yuanhao Du. 2018. Essential characteristics and clinical treatment regularity of acupuncture therapy. Chinese Acupuncture & Moxibustion 38, 6, 650-654.Google Scholar
- Jing Li, 2013. Perceptual motor features of expert acupuncture lifting-thrusting skills. Acupuncture in Medicine 31, 2, 172-177.Google ScholarCross Ref
- Meiren Chen, 2010. Study on the feasibility for judging the lifting-thrusting manipulation of acupuncture by using "Quality Control Figure" Measurement, Acupuncture Research 35, 02, 138-141.Google Scholar
- Chiyu Feng, 2019. Deep learning framework for Alzheimer's disease diagnosis via 3D-CNN and FSBi-LSTM, IEEE Access, 7, 63605-63618.Google ScholarCross Ref
- Muhammad Naveed Iqbal Qureshi, Jooyoung Oh, and Boreom Lee. 2019. 3D-CNN based discrimination of schizophrenia using resting-state fMRI, Artificial Intelligence in Medicine, 2019, 98, 10-17.Google ScholarDigital Library
- Sidratul Montaha, 2022. Time Distributed-CNN-LSTM: a hybrid approach combining CNN and LSTM to classify brain tumor on 3D MRI scans performing ablation study, IEEE Access, 10, 60039-60059.Google ScholarCross Ref
- Yin'e Hu, Huayuan Yang and Tangyi Liu. 2011. Study on cluster analysis of acupuncture-manipulation parameters, World Science Technology 13, 1, 59-63.Google ScholarCross Ref
- Tao Tu, 2021. Development and application of computer vision-based acupuncture manipulation classification system. Acupuncture Research 46, 6, 469-473.Google Scholar
- Sheng-Yi Gou, 2021. Recognition system of acupuncture manipulations based on an array PVDF tactile sensor and machine learning. Acupuncture Research 46, 6, 474-479.Google Scholar
- Chong Su, 2022. An action recognition method for manual acupuncture techniques using a tactile array finger cot. Computers in Biology and Medicine, 148, 105827.Google ScholarDigital Library
- Yeji Han, 2015.Quantification of the parameters of twisting-rotating acupuncture manipulation using a needle force measurement system. Integrative Medicine Research 4, 2, 57-65.Google ScholarCross Ref
- Robert T Davis, 2012. A new method for quantifying the needling component of acupuncture treatments, Acupuncture in Medicine 30, 2, 113-119.Google ScholarCross Ref
- Sha Li. 2021. Research progress on quantification of acupuncture stimulus based on kinematics and dynamics. Chinese Medicine Modern Distance Education of China 19, 14, 204-208.Google Scholar
- Dong Wu, 2021. Discussion on current state and research strategies of inter-disciplines of acupuncture-moxibustion and artificial intelligence. Acupuncture Research, 46(6), 541-545.Google Scholar
Index Terms
- Research on Acupuncture Technique Recognition Technology Based on Deep Learning Algorithms
Recommendations
Image Recognition Technology Based on Deep Learning
In order to apply deep learning to the field of image recognition, the basic principle, training process and model structure of deep belief networks (DBNs) in deep learning are analysed. For small samples, samples are down-sampled at the pretraining ...
Utilizing Technology Based Learning for Disaster Preparedness
Preparing for disasters can be a grueling, although necessary, exercise for those involved in emergency response. The large scale nature of disaster response poses many obstacles to executing an effective disaster preparedness drill that incorporates ...
Deep Learning Based Sub-Nyquist Modulation Recognition
ICCMS '23: Proceedings of the 2023 15th International Conference on Computer Modeling and SimulationIn this paper, we design a Convolutional Neural Network (CNN) for Sub-Nyquist modulation recognition and compare the performance of Long Short-Term Memory (LSTM) network and Convolutional Long Short-term Deep Neural Network (CLDNN) respectively. Unlike ...
Comments