Intensive care data are valuable for improvement of health care, policy making and many other purposes. Vast amount of such data are stored in different locations, on many different devices and in different data silos. Sharing data among different sources is a big challenge due to regulatory, operational and security reasons. One potential solution is federated machine learning, which is a method that sends machine learning algorithms simultaneously to all data sources, trains models in each source and aggregates the learned models. This strategy allows utilization of valuable data without moving them. One challenge in applying federated machine learning is the possibly different distributions of data from diverse sources. To tackle this problem, we proposed an adaptive boosting method named LoAda-Boost that increases the efficiency of federated machine learning. Using intensive care unit data from hospitals, we investigated the performance of learning in IID and non-IID data distribution scenarios, and showed that the proposed LoAdaBoost method achieved higher predictive accuracy with lower computational complexity than the baseline method.

Abstract-Due to consider the gray level spatial distribution information, some image segmentation technologies based on entropy threshold can enhance the thresholding segmentation performance. However, they still cannot distinguish image edges and noise well. Even though GLGM(gray-level & gradientmagnitude) entropy can effectively solve the problem, but it cannot segment effectively multi-objective and complex image. In this paper, a GLGM entropy fast segmentation method based on GA is presented by combining Real-code-GA and GLGM entropy, and the single threshold segmentation of GLGM entropy is further extended to multilevel threshold segmentation. Our method compared with GLGM entropy multi-threshold exhaustive method, the segmentation result obtained by our method is basically the same as the result obtained by exhaustive method.

No abstract

When processing the background and target blurred image, 1D-Otsu and 2D-Otsu segmentation effect is not good. The proposed algorithm used the gray value of the pixels and their similarity with neighboring pixels in gray value to build a histogram which was called gray-level spatial correlation histogram. Then threshold value is obtained by calculating GLSC histogram maximum between-class variance. Integral figure was introduced in order to make the time complexity from original 2 2