With the development of artificial intelligence technology, numerous research works that focus on AI-assisted healthcare, including clinical diagnosis, medical signal, and image processing, etc. In many cases, however, medical imaging may acquire data with few annotations, low signal-to-noise ratio, or small experimental samples, resulting in serious performance degradation during data processing. Therefore, data analysis of these data has become a serious problem.
Although many artificial intelligence and machine learning algorithms have been applied to deal with insufficient data, recent techniques are mainly applied to natural images such as photographs. Considering that the quality of many medical images is not sufficient for the application of conventional tools, techniques that can process the insufficient data in medical imaging are required.
To solve this problem, many works emerge over the years, such as semi-supervised learning, weakly supervised learning, unsupervised learning, transfer learning, and so on. Therefore, in this Research Topic, we would like to provide more solutions for medical neuroimaging with insufficient data. The list of possible topics includes but is not limited to:
1) Processing insufficient data for diagnosis of brain diseases;
2) Processing insufficient data for brain information management;
3) Processing brain data, signal, and image using few/no annotation;
4) Brain data augmentation methods;
5) Methods to improve the brain data representation;
With the development of artificial intelligence technology, numerous research works that focus on AI-assisted healthcare, including clinical diagnosis, medical signal, and image processing, etc. In many cases, however, medical imaging may acquire data with few annotations, low signal-to-noise ratio, or small experimental samples, resulting in serious performance degradation during data processing. Therefore, data analysis of these data has become a serious problem.
Although many artificial intelligence and machine learning algorithms have been applied to deal with insufficient data, recent techniques are mainly applied to natural images such as photographs. Considering that the quality of many medical images is not sufficient for the application of conventional tools, techniques that can process the insufficient data in medical imaging are required.
To solve this problem, many works emerge over the years, such as semi-supervised learning, weakly supervised learning, unsupervised learning, transfer learning, and so on. Therefore, in this Research Topic, we would like to provide more solutions for medical neuroimaging with insufficient data. The list of possible topics includes but is not limited to:
1) Processing insufficient data for diagnosis of brain diseases;
2) Processing insufficient data for brain information management;
3) Processing brain data, signal, and image using few/no annotation;
4) Brain data augmentation methods;
5) Methods to improve the brain data representation;