Segmentation of multiple anatomical structures is of great importance in medical image analysis. In this study, we proposed a $mathcal{W}$-net to simultaneously segment both the optic disc (OD) and the exudates in retinal images based on the multi-ta
sk learning (MTL) scheme. We introduced a class-balanced loss and a multi-task weighted loss to alleviate the imbalanced problem and to improve the robustness and generalization property of the $mathcal{W}$-net. We demonstrated the effectiveness of our approach by applying five-fold cross-validation experiments on two public datasets e_ophtha_EX and DiaRetDb1. We achieved F1-score of 94.76% and 95.73% for OD segmentation, and 92.80% and 94.14% for exudates segmentation. To further prove the generalization property of the proposed method, we applied the trained model on the DRIONS-DB dataset for OD segmentation and on the MESSIDOR dataset for exudate segmentation. Our results demonstrated that by choosing the optimal weights of each task, the MTL based $mathcal{W}$-net outperformed separate models trained individually on each task. Code and pre-trained models will be available at: url{https://github.com/FundusResearch/MTL_for_OD_and_exudates.git}.
On the basis of the Keldysh method of non-equilibrium systems, we develop a theory of electron tunneling in normal-metal/superconductor junctions. By using the tunneling Hamiltonian model (being appropriate for the tight-binding systems), the tunneli
ng current can be exactly obtained in terms of the equilibrium Green functions of the normal metal and the superconductor. We calculate the conductance of various junctions. The discrepancy between the present treatment and the well-known scheme by Blonder, Tinkham, and Klapwijk is found for some junctions of low interfacial potential barrier.
