Electroencephalogram (EEG) based on Brain-Computer Interfacing (BCI) is a biomedical signal that is the electrical activity from brain nerve cells. In addition, the medical images are a biomedical signal. The CT images, MRI images, B-ultrasound images, X-ray images and other medical images are generated as secondary or tertiary byproduct within physiological activities. These medical images are obtained by applying specific input to the life system, followed by receiving and measuring its output signal, to calculate the static or dynamic parameters of the system. These biological signals are used in clinical diagnosis, patient monitoring and biomedical research. Therefore, how to effectively detect, analyze, and study biomedical signals is of great significance for human beings to study life phenomena and medical science. Recently, machine learning technologies, especially deep learning, have significantly advanced biomedical signal analysis.
Various unsolved problems remain while the studies of biomedical signals using the advanced deep learning become increasingly popular. Firstly, the parameterization of deep architectures requires effort and expertise, whereas the accuracy obtained is often sensitive to small variations. Secondly, deep architectures often appear as ‘black boxes’ and their function cannot be intuitively interpreted by domain experts. Thirdly, due to the particularity of biomedical field, the datasets are substantially limited. Fourthly, mapping data from one modality to another to achieve cross-source data analysis or fusing information from two or more signal sources to perform a more accurate prediction are still a challenge. Furthermore, different medical institutions may have different scanners or protocols for the same radiological image. This may result into a variance of radiological images data distribution due to differences of medical institutions. How to construct a prediction model using the data from these different medical institutions to clinical decision making is also a challenge. (This Research Topic is set out to direct the focus on these challenges and advance knowledge of deep learning application in biomedical signal analysis.)
We welcome authors to submit Original Research, Review, and Mini-Review articles focusing on, but not limited to, the following subtopics:
- Theories, decoding models, and datasets for EEG based on BCI.
- Region of interest segmentation, detection and classification for medical images.
- Automated parameterization techniques to reduce the number or impact of hyper-parameters.
- Deep learning (CNN, RNN, GAN, etc.) for EEG based on BCI and medical images.
- Transfer learning, federal learning and Domain adaptation for EEG based on BCI and medical images.
- Explainable deep learning making the solution interpretable.
Reviews on related challenging topics or comparative studies are also welcome.
Wu Wei is affiliated with the Data Science Officer, Alto Neuroscience Inc., Los Altos, CA 94022. All other Topic Editors declare no competing interests with regards to the Research Topic subject.
Electroencephalogram (EEG) based on Brain-Computer Interfacing (BCI) is a biomedical signal that is the electrical activity from brain nerve cells. In addition, the medical images are a biomedical signal. The CT images, MRI images, B-ultrasound images, X-ray images and other medical images are generated as secondary or tertiary byproduct within physiological activities. These medical images are obtained by applying specific input to the life system, followed by receiving and measuring its output signal, to calculate the static or dynamic parameters of the system. These biological signals are used in clinical diagnosis, patient monitoring and biomedical research. Therefore, how to effectively detect, analyze, and study biomedical signals is of great significance for human beings to study life phenomena and medical science. Recently, machine learning technologies, especially deep learning, have significantly advanced biomedical signal analysis.
Various unsolved problems remain while the studies of biomedical signals using the advanced deep learning become increasingly popular. Firstly, the parameterization of deep architectures requires effort and expertise, whereas the accuracy obtained is often sensitive to small variations. Secondly, deep architectures often appear as ‘black boxes’ and their function cannot be intuitively interpreted by domain experts. Thirdly, due to the particularity of biomedical field, the datasets are substantially limited. Fourthly, mapping data from one modality to another to achieve cross-source data analysis or fusing information from two or more signal sources to perform a more accurate prediction are still a challenge. Furthermore, different medical institutions may have different scanners or protocols for the same radiological image. This may result into a variance of radiological images data distribution due to differences of medical institutions. How to construct a prediction model using the data from these different medical institutions to clinical decision making is also a challenge. (This Research Topic is set out to direct the focus on these challenges and advance knowledge of deep learning application in biomedical signal analysis.)
We welcome authors to submit Original Research, Review, and Mini-Review articles focusing on, but not limited to, the following subtopics:
- Theories, decoding models, and datasets for EEG based on BCI.
- Region of interest segmentation, detection and classification for medical images.
- Automated parameterization techniques to reduce the number or impact of hyper-parameters.
- Deep learning (CNN, RNN, GAN, etc.) for EEG based on BCI and medical images.
- Transfer learning, federal learning and Domain adaptation for EEG based on BCI and medical images.
- Explainable deep learning making the solution interpretable.
Reviews on related challenging topics or comparative studies are also welcome.
Wu Wei is affiliated with the Data Science Officer, Alto Neuroscience Inc., Los Altos, CA 94022. All other Topic Editors declare no competing interests with regards to the Research Topic subject.