11 October 2017 in International Journal of Multimedia Information Retrieval
2 November 2022 in IEEE Transactions on Computational Social Systems
Traditional classification algorithms struggle to categorize hypertensive retinopathy (HR) lesions correctly because they lack obvious characteristics. A regional IoT-enabled federated learning-based HR categorization approach (IoT-FHR) incorporating global and local attributes is suggested as a solution to this issue. The local feature arterial and venous nicking (AVN) classification model is fused with the overall IoT-FHR classification model to enhance the effect of the classification of IoT-FHR. The AVN classification model’s local lesion characteristics and the IoT-FHR classification model’s global lesion characteristics were combined using feature mean. After that, the results of the global IoT-FHR classification model are averaged with the results of the local AVN classification model. An easy neural network receives its input from the final outcome. The probability value of IoT-FHR in the fundus image is output by the sigmoid classifier after the neural network’s two fully connected and one dropout layer. The AVN classification makes a new kind of intersection detection algorithm suggestion. To determine the intersection points, the algorithm applies a logical and operation to the classified arteries and veins. It takes HR fundus pictures and extracts AVN image blocks using the region of interest extraction approach. The accuracy, sensitivity, and specificity of the suggested fusion model are 93.50%, 69.83%, and 98.33%, respectively, when tested on a private dataset. It is clear from the experiments and results that the suggested model leads the currently used methods when the single-stage classification model is compared with them.
1 January 2023 in BioMed Research International
Prediction of gene‐disease associations has grown in popularity in recent biomedical research. However, positive and unlabeled (PU) issues and limited gene‐disease association data are common concerns with present association prediction algorithms. A gene‐disease association prediction approach based on Katz‐enhanced inductive matrix completion is suggested in light of the abovementioned flaws. Preestimate based on the Katz technique and refined estimation based on the inductive matrix completion approach makes the model. The Katz technique is utilized to preestimate the gene‐disease association on the basis of gene‐disease heterogeneous network to mitigate the effects of association data‐sparse and PU issues. The Katz technique, however, necessarily introduces some noise when predicting gene‐disease connections due to the similarity network’s quality limitations. Therefore, the elastic net regularization approach is utilized to increase the resilience of the conventional inductive matrix completion model. As a result, the prediction effect of gene‐disease connections is increased using robustness and a better inductive matrix completion model. The experimental findings demonstrate that the proposed model has dramatically increased recall and precision compared to widely used gene‐disease association prediction approaches. It can also resolve the typical cold‐start issue in association prediction. The proposed KIMC method may consider integrating more diverse biological data sources in the future and also aid in the effective extraction of the feature data of genes and diseases with higher correlation from this biological data to improve the prediction effect.