Skin lesion classification is a pivotal process in dermatology, enabling the early detection and precise diagnosis of skin diseases, leading to improved patient outcomes. Deep learning has shown great potential for this task by leveraging its ability to learn complex patterns from images. However, diagnostic accuracy is compromised by exclusive reliance on single-modality images. This research work proposes an innovative framework that unifies a Vision Transformer model with transfer learning, channel attention mechanism, and ROI for the accurate detection of skin conditions, including skin cancer. The proposed approach blends computer vision and machine-learning techniques, leveraging a comprehensive dataset comprised of macroscopic dermoscopic images, appended with patient metadata. Compared with conventional techniques, the proposed methodology exhibits significant improvements in various parameters, including sensitivity, specificity, and precision. Moreover, it demonstrates outstanding performance in real-world datasets, reinforcing its potential for clinical implementation. With a remarkable accuracy of 99%, the method outperforms existing approaches. Overall, this investigation underscores the transformative impact of deep learning and multimodal data analysis in the dermoscopic domain, projecting substantial headway into the field of skin lesion analytic diagnosis.

An increasing number of older adults integrate the Internet in their daily life. Although undertaking a greater range of online activities offers enhanced opportunities to live a self-determined and socially connected life, understanding of the breadth of Internet use is still limited. We address this gap by examining how individual factors representing inequalities between groups of older adults in society are associated with the breadth of their Internet use. The study analyzed survey data collected in 2017 from older adults living in a city in Germany (N = 1,136, age 65–90 years). The group of Internet users comprised of 69.0% (784) of participants, exhibited high levels of autonomy in outdoor activities (87.8%), and reported, on average, 4.92 online activities (SD = 2.31, range 1–8). Linear regression analysis showed positive associations for perceiving higher levels of behavioral control (PBC) in using digital technologies, being male and younger, and holding an academic degree. No associations were found for having medium level of education, living together with someone, and reporting better health. These findings highlight that despite the increased percentage of users, digital inequality regarding the range of online activities prevails. This inequality impedes subgroups of older adults to prepare for future situations in which online activities could substitute outdoor activities that might not be possible anymore. Interventions for these subgroups should emphasize digital skills that facilitate engagement in diverse online activities covering various purposes and life domains.