Optimizing sleep staging on multimodal time series: Leveraging borderline synthetic minority oversampling technique and supervised convolutional contrastive learning.

导出 LinkOut

翻译打印收藏纠错

作者： Xinyu,Huang ^[1] ; Franziska,Schmelter ^[2] ; Muhammad Tausif,Irshad ^[3] ; Artur,Piet ^[4] ; Muhammad Adeel,Nisar ^[5] ; Christian,Sina ^[6] ; Marcin,Grzegorzek ^[7]

作者单位： Institute of Medical Informatics, University of Lübeck, Germany. Electronic address: x.huang@uni-luebeck.de. ^[1] Institute of Nutritional Medicine, University of Lübeck and University Medical Center Schleswig-Holstein, Lübeck, Germany. Electronic address: Franziska.Schmelter@uksh.de. ^[2] Institute of Medical Informatics, University of Lübeck, Germany. Electronic address: m.irshad@uni-luebeck.de. ^[3] Institute of Medical Informatics, University of Lübeck, Germany. Electronic address: ar.piet@uni-luebeck.de. ^[4] Department of IT, University of the Punjab, Lahore, Pakistan. Electronic address: adeel.nisar@pucit.edu.pk. ^[5] Institute of Nutritional Medicine, University of Lübeck and University Medical Center Schleswig-Holstein, Lübeck, Germany; Fraunhofer Research Institution for Individualized and Cell-Based Medical Engineering (IMTE), Lübeck, Germany. Electronic address: Christian.Sina@uksh.de. ^[6] Institute of Medical Informatics, University of Lübeck, Germany; Fraunhofer Research Institution for Individualized and Cell-Based Medical Engineering (IMTE), Lübeck, Germany. Electronic address: marcin.grzegorzek@uni-luebeck.de. ^[7]

关键词 B-SMOTE Biosignal processing Cross-domain adaption Multimodal polysomnography Residual network Sleep stage classification Supervised contrastive learning Truncated cross-entropy