http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : ( Prabhu Sethuramalingam ) (검색결과 3 건)
- 무료
- 기관 내 무료
- 유료
-
Virtual health monitoring - A robot based approach
( Vigneswari Gowri ),( Prabhu Sethuramalingam ) 한국감성과학회 2021 한국감성과학회 국제학술대회(ICES) Vol.2021 No.-
Health monitoring and care is been considered as the major field in medical technological advancements. Sensors and wireless communication technologies has been applied with learning techniques to design low-cost, and low power integrated circuits with intelligent systems. This system detects, measures and analyses the health parameters such as Sp02, Heart rate, ECG and Body temperature. This system is capable of analyzing, processing and communicating the sensor data in real-time using Wi-Fi to achieve a seamless data transfer. These systems become an inseparable part of the medical environment both to the patient and the doctor. It enables the information transfer at a faster and accurate manner. Integration of these systems with robots has a major advantage of food and medicine delivery, UV sanitization and haptic video calling. Applied Machine learning algorithms with experimentation such as Min-max algorithm, Feature selection and SVM gives refines the data with most accurate values and predicts the medicine or further treatment to be provided. Along with this, robot integrated with this system serves as an emotional support with the sensor values. This can predict the patient condition on emotional basis and plays songs or movies of their preference and calls their paired friends or relatives which boosts their energy and normalizes the body parameters. The result of the combined algorithms gives 96.2% accuracy which can be improved on further classification of obtained data.
-
Virtual health monitoring - A robot based approach
( Vigneswari Gowri ),( Prabhu Sethuramalingam ) 한국감성과학회 2021 추계학술대회 Vol.2021 No.0
Health monitoring and care is been considered as the major field in medical technological advancements. Sensors and wireless communication technologies has been applied with learning techniques to design low-cost, and low power integrated circuits with intelligent systems. This system detects, measures and analyses the health parameters such as Sp02, Heart rate, ECG and Body temperature. This system is capable of analyzing, processing and communicating the sensor data in real-time using Wi-Fi to achieve a seamless data transfer. These systems become an inseparable part of the medical environment both to the patient and the doctor. It enables the information transfer at a faster and accurate manner. Integration of these systems with robots has a major advantage of food and medicine delivery, UV sanitization and haptic video calling. Applied Machine learning algorithms with experimentation such as Min-max algorithm, Feature selection and SVM gives refines the data with most accurate values and predicts the medicine or further treatment to be provided. Along with this, robot integrated with this system serves as an emotional support with the sensor values. This can predict the patient condition on emotional basis and plays songs or movies of their preference and calls their paired friends or relatives which boosts their energy and normalizes the body parameters. The result of the combined algorithms gives 96.2% accuracy which can be improved on further classification of obtained data.
-
EEGNet Classification for Enhancing Two-Class EEG-Based Motor Imagery- Brain Computer Interface
( Senthil Vadivelan. D ),( Prabhu Sethuramalingam ) 한국감성과학회 2023 한국감성과학회 국제학술대회(ICES) Vol.2023 No.-
Effective signal classification of motor imagery (MI) plays a pivotal role in the development of brain-computer interfaces (BCI). Paradigms of braincomputer interfaces (BCI) empower individuals to establish communication with the external world exclusively through their brain's intentions. While convolutional neural networks have seen a gradual adoption in the task of classifying motor imagery (MI) and have achieved impressive performance, several challenges persist, making the effective decoding of raw EEG signals a demanding task. These challenges include: 1) non-linearity, non-stationarity, and low signal-to-noise ratio inherent in EEG signals. 2) Many existing end-to-end MI models employ a single-scale convolution, which constrains the classification results as the optimal convolution scale varies among different subjects, a phenomenon known as subject variability. In this study, we address the aforementioned challenges by employing EEGnet to classify the lefthand and righthand motor imagery movements, a highly efficient and streamlined deep learning framework. The methodology presented in this study is assessed using MI datasets from BCI Competition IV 2a, achieving classification accuracies of 89%. These classification outcomes establish the proposed methodology as an effective approach for future BCI system design.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
