Ghena Hammour;Harry Davies;Giuseppe Atzori;Ciro Della Monica;Kiran K. G. Ravindran;Victoria Revell;Derk-Jan Dijk;Danilo P. Mandic
{"title":"From Scalp to Ear-EEG: A Generalizable Transfer Learning Model for Automatic Sleep Scoring in Older People","authors":"Ghena Hammour;Harry Davies;Giuseppe Atzori;Ciro Della Monica;Kiran K. G. Ravindran;Victoria Revell;Derk-Jan Dijk;Danilo P. Mandic","doi":"10.1109/JTEHM.2024.3388852","DOIUrl":null,"url":null,"abstract":"Objective: Sleep monitoring has extensively utilized electroencephalogram (EEG) data collected from the scalp, yielding very large data repositories and well-trained analysis models. Yet, this wealth of data is lacking for emerging, less intrusive modalities, such as ear-EEG.Methods and procedures: The current study seeks to harness the abundance of open-source scalp EEG datasets by applying models pre-trained on data, either directly or with minimal fine-tuning; this is achieved in the context of effective sleep analysis from ear-EEG data that was recorded using a single in-ear electrode, referenced to the ipsilateral mastoid, and developed in-house as described in our previous work. Unlike previous studies, our research uniquely focuses on an older cohort (17 subjects aged 65-83, mean age 71.8 years, some with health conditions), and employs LightGBM for transfer learning, diverging from previous deep learning approaches. Results: Results show that the initial accuracy of the pre-trained model on ear-EEG was 70.1%, but fine-tuning the model with ear-EEG data improved its classification accuracy to 73.7%. The fine-tuned model exhibited a statistically significant improvement (p < 0.05, dependent t-test) for 10 out of the 13 participants, as reflected by an enhanced average Cohen’s kappa score (a statistical measure of inter-rater agreement for categorical items) of 0.639, indicating a stronger agreement between automated and expert classifications of sleep stages. Comparative SHAP value analysis revealed a shift in feature importance for the N3 sleep stage, underscoring the effectiveness of the fine-tuning process.Conclusion: Our findings underscore the potential of fine-tuning pre-trained scalp EEG models on ear-EEG data to enhance classification accuracy, particularly within an older population and using feature-based methods for transfer learning. This approach presents a promising avenue for ear-EEG analysis in sleep studies, offering new insights into the applicability of transfer learning across different populations and computational techniques.Clinical impact: An enhanced ear-EEG method could be pivotal in remote monitoring settings, allowing for continuous, non-invasive sleep quality assessment in elderly patients with conditions like dementia or sleep apnea.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"12 ","pages":"448-456"},"PeriodicalIF":3.7000,"publicationDate":"2024-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10504255","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","FirstCategoryId":"5","ListUrlMain":"https://ieeexplore.ieee.org/document/10504255/","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 0
Abstract
Objective: Sleep monitoring has extensively utilized electroencephalogram (EEG) data collected from the scalp, yielding very large data repositories and well-trained analysis models. Yet, this wealth of data is lacking for emerging, less intrusive modalities, such as ear-EEG.Methods and procedures: The current study seeks to harness the abundance of open-source scalp EEG datasets by applying models pre-trained on data, either directly or with minimal fine-tuning; this is achieved in the context of effective sleep analysis from ear-EEG data that was recorded using a single in-ear electrode, referenced to the ipsilateral mastoid, and developed in-house as described in our previous work. Unlike previous studies, our research uniquely focuses on an older cohort (17 subjects aged 65-83, mean age 71.8 years, some with health conditions), and employs LightGBM for transfer learning, diverging from previous deep learning approaches. Results: Results show that the initial accuracy of the pre-trained model on ear-EEG was 70.1%, but fine-tuning the model with ear-EEG data improved its classification accuracy to 73.7%. The fine-tuned model exhibited a statistically significant improvement (p < 0.05, dependent t-test) for 10 out of the 13 participants, as reflected by an enhanced average Cohen’s kappa score (a statistical measure of inter-rater agreement for categorical items) of 0.639, indicating a stronger agreement between automated and expert classifications of sleep stages. Comparative SHAP value analysis revealed a shift in feature importance for the N3 sleep stage, underscoring the effectiveness of the fine-tuning process.Conclusion: Our findings underscore the potential of fine-tuning pre-trained scalp EEG models on ear-EEG data to enhance classification accuracy, particularly within an older population and using feature-based methods for transfer learning. This approach presents a promising avenue for ear-EEG analysis in sleep studies, offering new insights into the applicability of transfer learning across different populations and computational techniques.Clinical impact: An enhanced ear-EEG method could be pivotal in remote monitoring settings, allowing for continuous, non-invasive sleep quality assessment in elderly patients with conditions like dementia or sleep apnea.
期刊介绍:
The IEEE Journal of Translational Engineering in Health and Medicine is an open access product that bridges the engineering and clinical worlds, focusing on detailed descriptions of advanced technical solutions to a clinical need along with clinical results and healthcare relevance. The journal provides a platform for state-of-the-art technology directions in the interdisciplinary field of biomedical engineering, embracing engineering, life sciences and medicine. A unique aspect of the journal is its ability to foster a collaboration between physicians and engineers for presenting broad and compelling real world technological and engineering solutions that can be implemented in the interest of improving quality of patient care and treatment outcomes, thereby reducing costs and improving efficiency. The journal provides an active forum for clinical research and relevant state-of the-art technology for members of all the IEEE societies that have an interest in biomedical engineering as well as reaching out directly to physicians and the medical community through the American Medical Association (AMA) and other clinical societies. The scope of the journal includes, but is not limited, to topics on: Medical devices, healthcare delivery systems, global healthcare initiatives, and ICT based services; Technological relevance to healthcare cost reduction; Technology affecting healthcare management, decision-making, and policy; Advanced technical work that is applied to solving specific clinical needs.