8 May 2023 in IEEE Journal of Biomedical and Health Informatics
Distributed big data and digital healthcare technologies have great potential to promote medical services, but challenges arise when it comes to learning predictive model from diverse and complex e-health datasets. Federated Learning (FL), as a collaborative machine learning technique, aims to address the challenges by learning a joint predictive model across multi-site clients, especially for distributed medical institutions or hospitals. However, most existing FL methods assume that clients possess fully labeled data for training, which is often not the case in e-health datasets due to high labeling costs or expertise requirement. Therefore, this work proposes a novel and feasible approach to learn a Federated Semi-Supervised Learning (FSSL) model from distributed medical image domains, where a federated pseudo-labeling strategy for unlabeled clients is developed based on the embedded knowledge learned from labeled clients. This greatly mitigates the annotation deficiency at unlabeled clients and leads to a cost-effective and efficient medical image analysis tool. We demonstrated the effectiveness of our method by achieving significant improvements compared to the state-of-the-art in both fundus image and prostate MRI segmentation tasks, resulting in the highest Dice scores of 89.23 and 91.95 respectively even with only a few labeled clients participating in model training. This reveals the superiority of our method for practical deployment, ultimately facilitating the wider use of FL in healthcare and leading to better patient outcomes.
27 May 2019 in IEEE/ACM Transactions on Computational Biology and Bioinformatics
Mitosis detection is one of the challenging steps in biomedical imaging research, which can be used to observe the cell behavior. Most of the already existing methods that are applied in detecting mitosis usually contain many nonmitotic events (normal cell and background) in the result (false positives, FPs). In order to address such problem, in this study, we propose to apply 2.5-dimensional (2.5D) networks called CasDetNet_CLSTM, which can accurately detect mitotic events in 4D microscopic images. This CasDetNet_CLSTM involves a 2.5D faster region-based convolutional neural network (Faster R-CNN) as the first network, and a convolutional long short-term memory (CLSTM) network as the second network. The first network is used to select candidate cells using the information from nearby slices, whereas the second network uses temporal information to eliminate FPs and refine the result of the first network. Our experiment shows that the precision and recall of our networks yield better results than those of other state-of-the-art methods.
1 March 2010 in IEEE Transactions on Biomedical Engineering
Cell enumeration and diagnosis using peripheral blood smears are routine tasks in many biological and pathological examinations. Not every area in the smear is appropriate for such tasks due to severe cell clumping or sparsity. Manual working-area selection is slow, subjective, inconsistent, and statistically biased. Automatic working-area classification can reproducibly identify appropriate working smear areas. However, very little research has been reported in the literature. With the aim of providing a preprocessing step for further detailed cell enumeration and diagnosis for high-throughput screening (HTS), we propose an integrated algorithm for area classification and quantify both cell spreading and cell clumping in terms of individual clumps and the occurrence probabilities of the group of clumps over the image. Comprehensive comparisons are presented to compare the effect of these quantifications and their combinations. Our experiments using images of Giemsa-stained blood smears show that the method is efficient, accurate (above 88.9% hit rates for all areas in the validation set of 140 images), and robust (above 78.1% hit rates for a test set of 4878 images). This lays a good foundation for fast working-area selection in HTS.