Yang Yi Poh;Ethan Grooby;Kenneth Tan;Lindsay Zhou;Arrabella King;Ashwin Ramanathan;Atul Malhotra;Mehrtash Harandi;Faezeh Marzbanrad
{"title":"NeoSSNet:利用深度学习实时分离新生儿胸音","authors":"Yang Yi Poh;Ethan Grooby;Kenneth Tan;Lindsay Zhou;Arrabella King;Ashwin Ramanathan;Atul Malhotra;Mehrtash Harandi;Faezeh Marzbanrad","doi":"10.1109/OJEMB.2024.3401571","DOIUrl":null,"url":null,"abstract":"<italic>Goal:</i>\n Auscultation for neonates is a simple and non-invasive method of diagnosing cardiovascular and respiratory disease. However, obtaining high-quality chest sounds containing only heart or lung sounds is non-trivial. Hence, this study introduces a new deep-learning model named NeoSSNet and evaluates its performance in neonatal chest sound separation with previous methods. \n<italic>Methods:</i>\n We propose a masked-based architecture similar to Conv-TasNet. The encoder and decoder consist of 1D convolution and 1D transposed convolution, while the mask generator consists of a convolution and transformer architecture. The input chest sounds were first encoded as a sequence of tokens using 1D convolution. The tokens were then passed to the mask generator to generate two masks, one for heart sounds and one for lung sounds. Each mask is then applied to the input token sequence. Lastly, the tokens are converted back to waveforms using 1D transposed convolution. \n<italic>Results:</i>\n Our proposed model showed superior results compared to the previous methods based on objective distortion measures, ranging from a 2.01 dB improvement to a 5.06 dB improvement. The proposed model is also significantly faster than the previous methods, with at least a 17-time improvement. \n<italic>Conclusions:</i>\n The proposed model could be a suitable preprocessing step for any health monitoring system where only the heart sound or lung sound is desired.","PeriodicalId":33825,"journal":{"name":"IEEE Open Journal of Engineering in Medicine and Biology","volume":"5 ","pages":"345-352"},"PeriodicalIF":2.7000,"publicationDate":"2024-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10531026","citationCount":"0","resultStr":"{\"title\":\"NeoSSNet: Real-Time Neonatal Chest Sound Separation Using Deep Learning\",\"authors\":\"Yang Yi Poh;Ethan Grooby;Kenneth Tan;Lindsay Zhou;Arrabella King;Ashwin Ramanathan;Atul Malhotra;Mehrtash Harandi;Faezeh Marzbanrad\",\"doi\":\"10.1109/OJEMB.2024.3401571\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<italic>Goal:</i>\\n Auscultation for neonates is a simple and non-invasive method of diagnosing cardiovascular and respiratory disease. However, obtaining high-quality chest sounds containing only heart or lung sounds is non-trivial. Hence, this study introduces a new deep-learning model named NeoSSNet and evaluates its performance in neonatal chest sound separation with previous methods. \\n<italic>Methods:</i>\\n We propose a masked-based architecture similar to Conv-TasNet. The encoder and decoder consist of 1D convolution and 1D transposed convolution, while the mask generator consists of a convolution and transformer architecture. The input chest sounds were first encoded as a sequence of tokens using 1D convolution. The tokens were then passed to the mask generator to generate two masks, one for heart sounds and one for lung sounds. Each mask is then applied to the input token sequence. Lastly, the tokens are converted back to waveforms using 1D transposed convolution. \\n<italic>Results:</i>\\n Our proposed model showed superior results compared to the previous methods based on objective distortion measures, ranging from a 2.01 dB improvement to a 5.06 dB improvement. The proposed model is also significantly faster than the previous methods, with at least a 17-time improvement. \\n<italic>Conclusions:</i>\\n The proposed model could be a suitable preprocessing step for any health monitoring system where only the heart sound or lung sound is desired.\",\"PeriodicalId\":33825,\"journal\":{\"name\":\"IEEE Open Journal of Engineering in Medicine and Biology\",\"volume\":\"5 \",\"pages\":\"345-352\"},\"PeriodicalIF\":2.7000,\"publicationDate\":\"2024-03-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10531026\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE Open Journal of Engineering in Medicine and Biology\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/10531026/\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENGINEERING, BIOMEDICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Open Journal of Engineering in Medicine and Biology","FirstCategoryId":"1085","ListUrlMain":"https://ieeexplore.ieee.org/document/10531026/","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 0
摘要
目标:新生儿听诊是诊断心血管和呼吸系统疾病的一种简单而无创的方法。然而,获得仅包含心音或肺音的高质量胸音并非易事。因此,本研究引入了一种名为 NeoSSNet 的新型深度学习模型,并评估了它与之前的方法在新生儿胸音分离方面的性能。方法:我们提出了一种与 Conv-TasNet 类似的基于掩码的架构。编码器和解码器由一维卷积和一维转置卷积组成,而掩码生成器则由卷积和变换器架构组成。首先使用一维卷积将输入的胸腔声音编码为标记序列。然后将标记传递给掩码生成器,生成两个掩码,一个用于心音,另一个用于肺音。然后将每个掩码应用于输入的标记序列。最后,使用一维转置卷积将标记转换回波形。结果根据客观失真测量结果,我们提出的模型与之前的方法相比显示出更优越的效果,改善幅度从 2.01 dB 到 5.06 dB 不等。此外,所提出的模型也明显快于之前的方法,至少提高了 17 倍。结论:对于任何只需要心音或肺音的健康监测系统来说,所提出的模型都是一个合适的预处理步骤。
NeoSSNet: Real-Time Neonatal Chest Sound Separation Using Deep Learning
Goal:
Auscultation for neonates is a simple and non-invasive method of diagnosing cardiovascular and respiratory disease. However, obtaining high-quality chest sounds containing only heart or lung sounds is non-trivial. Hence, this study introduces a new deep-learning model named NeoSSNet and evaluates its performance in neonatal chest sound separation with previous methods.
Methods:
We propose a masked-based architecture similar to Conv-TasNet. The encoder and decoder consist of 1D convolution and 1D transposed convolution, while the mask generator consists of a convolution and transformer architecture. The input chest sounds were first encoded as a sequence of tokens using 1D convolution. The tokens were then passed to the mask generator to generate two masks, one for heart sounds and one for lung sounds. Each mask is then applied to the input token sequence. Lastly, the tokens are converted back to waveforms using 1D transposed convolution.
Results:
Our proposed model showed superior results compared to the previous methods based on objective distortion measures, ranging from a 2.01 dB improvement to a 5.06 dB improvement. The proposed model is also significantly faster than the previous methods, with at least a 17-time improvement.
Conclusions:
The proposed model could be a suitable preprocessing step for any health monitoring system where only the heart sound or lung sound is desired.
期刊介绍:
The IEEE Open Journal of Engineering in Medicine and Biology (IEEE OJEMB) is dedicated to serving the community of innovators in medicine, technology, and the sciences, with the core goal of advancing the highest-quality interdisciplinary research between these disciplines. The journal firmly believes that the future of medicine depends on close collaboration between biology and technology, and that fostering interaction between these fields is an important way to advance key discoveries that can improve clinical care.IEEE OJEMB is a gold open access journal in which the authors retain the copyright to their papers and readers have free access to the full text and PDFs on the IEEE Xplore® Digital Library. However, authors are required to pay an article processing fee at the time their paper is accepted for publication, using to cover the cost of publication.