Rifqi Abdillah, R. Sarno, T. Amri, Faris Atoil Haq, K. R. Sungkono, Dwi Sunaryono
{"title":"用于开发噪声滤波器的光体积脉搏波数据信号聚类","authors":"Rifqi Abdillah, R. Sarno, T. Amri, Faris Atoil Haq, K. R. Sungkono, Dwi Sunaryono","doi":"10.1109/ICAIIC57133.2023.10066966","DOIUrl":null,"url":null,"abstract":"This paper aims to evaluate Photoplethysmography signals taken using fingertip pulse waves on human fingers which are generally not all in good condition. In common devices, the data obtained does not only contain photoplethysmography signals, but some noise also contaminates it. Noise is an unwanted ripple-shaped signal existing in signal transmission. Noise will interfere the desired quality of the received signal and ultimately change the information contained in the signal. This situation requires improvements to the photoplethysmography signal to make the signals are in the best condition so machine learning produces a more optimal output. Noise filters cannot be done with the same treatment because noise level in each data is different and must have different filter weights. This paper proposes a method to filter noise based on the level of noise in the signal. The approach taken in this study uses two stages, clustering and noise filtering. The first approach is clustering using the K-means clustering method by utilizing the coefficient of variation and slope features to group signals based on their noise level. The second approach uses exponential filtering, which performs by weighting the filter based on the cluster so that the data have different adjustments ratio of the level of smoothing. The result of the signal-to-noise ratio on Non-filtered Data is 181.49. Signal to noise ratio on the Constant Weighted Filter is 183.79 and increases to 187.48 after using the Clustered and Weighted Filter method.","PeriodicalId":105769,"journal":{"name":"2023 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Clustering of Photoplethysmography Data Signals for Developing Noise Filters\",\"authors\":\"Rifqi Abdillah, R. Sarno, T. Amri, Faris Atoil Haq, K. R. Sungkono, Dwi Sunaryono\",\"doi\":\"10.1109/ICAIIC57133.2023.10066966\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper aims to evaluate Photoplethysmography signals taken using fingertip pulse waves on human fingers which are generally not all in good condition. In common devices, the data obtained does not only contain photoplethysmography signals, but some noise also contaminates it. Noise is an unwanted ripple-shaped signal existing in signal transmission. Noise will interfere the desired quality of the received signal and ultimately change the information contained in the signal. This situation requires improvements to the photoplethysmography signal to make the signals are in the best condition so machine learning produces a more optimal output. Noise filters cannot be done with the same treatment because noise level in each data is different and must have different filter weights. This paper proposes a method to filter noise based on the level of noise in the signal. The approach taken in this study uses two stages, clustering and noise filtering. The first approach is clustering using the K-means clustering method by utilizing the coefficient of variation and slope features to group signals based on their noise level. The second approach uses exponential filtering, which performs by weighting the filter based on the cluster so that the data have different adjustments ratio of the level of smoothing. The result of the signal-to-noise ratio on Non-filtered Data is 181.49. Signal to noise ratio on the Constant Weighted Filter is 183.79 and increases to 187.48 after using the Clustered and Weighted Filter method.\",\"PeriodicalId\":105769,\"journal\":{\"name\":\"2023 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-02-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2023 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICAIIC57133.2023.10066966\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICAIIC57133.2023.10066966","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Clustering of Photoplethysmography Data Signals for Developing Noise Filters
This paper aims to evaluate Photoplethysmography signals taken using fingertip pulse waves on human fingers which are generally not all in good condition. In common devices, the data obtained does not only contain photoplethysmography signals, but some noise also contaminates it. Noise is an unwanted ripple-shaped signal existing in signal transmission. Noise will interfere the desired quality of the received signal and ultimately change the information contained in the signal. This situation requires improvements to the photoplethysmography signal to make the signals are in the best condition so machine learning produces a more optimal output. Noise filters cannot be done with the same treatment because noise level in each data is different and must have different filter weights. This paper proposes a method to filter noise based on the level of noise in the signal. The approach taken in this study uses two stages, clustering and noise filtering. The first approach is clustering using the K-means clustering method by utilizing the coefficient of variation and slope features to group signals based on their noise level. The second approach uses exponential filtering, which performs by weighting the filter based on the cluster so that the data have different adjustments ratio of the level of smoothing. The result of the signal-to-noise ratio on Non-filtered Data is 181.49. Signal to noise ratio on the Constant Weighted Filter is 183.79 and increases to 187.48 after using the Clustered and Weighted Filter method.