We propose a novel cross-domain transfer learning framework that leverages knowledge from multiple datasets to improve medical image segmentation on a target task. Our method employs a teacher-student learning paradigm, where a joint teacher model aggregates domain-invariant features from diverse datasets, and a dataset-specific student model is trained via knowledge distillation. We validate our approach on six medical imaging datasets —BrainMetShare, ISLES, BraTS (MRI-based) and Lung MSD, LiTS, KiTS (CT-based)— demonstrating its effectiveness in addressing distributional shifts and enhancing segmentation accuracy across heterogeneous tasks. The results show consistent improvements over baseline models. These findings underscore the potential of multi-dataset knowledge distillation for robust and generalizable segmentation in medical imaging applications.
In this paper, we proposed a scheduling method for semiconductor manufacturing systems by extending that for a job shop scheduling problem (JSSP) using deep reinforcement learning approach. The proposed method can handle the unique characteristics of the semiconductor manufacturing process, i.e., the reentrancy and the flexibility. The advantage of the proposed method is scalable and can effectively handle increasing problem sizes. From the results of experiments, we confirmed that even when small-sized tasks were used in the training, the scheduling performance for large-sized tasks was equal to or better than the case using large-sized tasks in the training.
In recent years, there has been an observable proliferation of travel services available to travelers. Consequently, a shift has occurred in the manner in which travelers plan their journeys. Historically, travelers depended on travel guides and literature as primary sources of information. In the contemporary context, however, travelers are increasingly turning to emotional and experiential data to inform their travel planning decisions. The objective of this study is to propose travel destinations that are congruent with travelers’ preferences and emotional states. In this study, we propose the EmoTABI that extends the TABIMAE Mind Search System introducing an atmosphere emotion extraction.This system utilizes image data to discern travelers’ emotions, identify objects, and assess the prevailing atmosphere of a given location. The findings from the demonstration of this system indicate its potential for predicting the success of proposed tourist destinations in terms of their emotional appeal to travelers.
The primary goal of the article is to analyse the Generative Artificial Intelligence (Gen AI tools) with respect to their potential utility level for legitimate purposes as well as malicious use in the cybersecurity area. With this end in view, the author summarises the benefits and drawbacks of the deepfake technology in an attempt to determine whether the overall evaluation of the phenomenon should be positive or negative. The paper presents potential areas for legitimate use of deepfakes confronted with examples of the negative impact of the manipulated content on the area of cybersecurity. In addition, the author also classifies deepfakes into two categories depending on their type and utility level. In the part devoted to cybersecurity threats risks for the organisational cybersecurity area have been summarised, social engineering techniques used in deepfake campaigns presented and remedial actions to curb the dissemination of deepfakes proposed. The paper’s main objective is to answer the two research questions formulated: RQ 1: Can online users benefit from the deepfake technology? and RQ 2: What actions are required to mitigate the harmful impact of deepfakes on the cybersecurity area? The paper finishes with the conclusion and indication of potential further research avenues.
