About this Research Topic
Deep learning has shown potential advancement for nature images and has surpassed conventional machine learning methods in several tasks. Several applications of deep learning in medical imaging include screening for several diseases, such as analysis of retinal fundus images, and classification of brain cancer state and lung disease. The rapid development and application of deep learning in such diverse applications stem from the ability that deep learning can automatically extract the most relevant features for data interpretation and inference directly from the annotated data. However, this comes at the cost of requiring a huge amount of data with expert annotations to train the deep learning model.
In practice, acquiring a large amount of expert-annotated datasets is challenging in the medical image area. The reason is obtaining annotation from medical images requires experts such as physicians or radiologists who are busy with clinical duties to accurately label the medical images. Moreover, unlike natural image analysis, where the images could be easily captured using a standard camera, medical images include CT, MRI, PET, etc, which are difficult and expensive to obtain in general. Therefore, the requirement for deep learning in medical image applications to produce generalizable learning from small datasets is becoming necessary.
In this Research Topic, we aim to collect a variety of research papers on the topic of deep learning in medical imaging with limited datasets. We look forward to receiving reviews and research on state-of-the-art technology in the field. This research topic will include but is not limited to the following aspects:
1) One/Few-shot learning
2) Semi-supervised learning
3) Self-supervised learning
4) Medical image classification, segmentation, registration, diagnosis, prognostic, etc with limited data/label
5) Generative neural network
In practice, acquiring a large amount of expert-annotated datasets is challenging in the medical image area. The reason is obtaining annotation from medical images requires experts such as physicians or radiologists who are busy with clinical duties to accurately label the medical images. Moreover, unlike natural image analysis, where the images could be easily captured using a standard camera, medical images include CT, MRI, PET, etc, which are difficult and expensive to obtain in general. Therefore, the requirement for deep learning in medical image applications to produce generalizable learning from small datasets is becoming necessary.
In this Research Topic, we aim to collect a variety of research papers on the topic of deep learning in medical imaging with limited datasets. We look forward to receiving reviews and research on state-of-the-art technology in the field. This research topic will include but is not limited to the following aspects:
1) One/Few-shot learning
2) Semi-supervised learning
3) Self-supervised learning
4) Medical image classification, segmentation, registration, diagnosis, prognostic, etc with limited data/label
5) Generative neural network
Keywords: Deep learning, Medical Imaging, Limited Dataset, Training AI models, limited data training, Medical image processing, Medical image analysis, Medical image segmentation, Medical image classification
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
