AFFD-Net: A Dual-Decoder Network Based on Attention-Enhancing and Feature Fusion for Retinal Vessel Segmentation

Abstract

mHealth interventions hold promise for supporting the self- management of chronic diseases, yet their limited utilisation remains a problem. Given the significant variability among indi- viduals with chronic diseases, tailored approaches are imperative. Adaptive User Interfaces (AUIs) may help to address the diverse and evolving needs of this demographic. To investigate this approach, we developed an AUI prototype informed by existing literature and used it as the basis for a focus group and interview study involving 22 participants. Concurrently, a quantitative survey was carried out to extract preferences for AUIs in chronic disease related applications with 90 participants. Our findings reveal that user engagement with AUIs is influenced by individual capabilities and disease severity. Additionally, we explore user preferences for AUIs, expanding the adaptation literature by uncovering usage challenges, proposing practical strategies for enhanced AUI design, and acknowledging potential trade-offs between usability and adaptation. Lastly, we present design considerations for AUIs in chronic disease applications, aiming to prevent user overload and maintain critical software functionality and usability aspects.

Citation

Zijian, Xiang, Ning Chunyu, Li Mingye, Ma Kaizheng, Shi Lemin, Wang Wei, and Ye Guanshi. "AFFD-Net: A Dual-Decoder Network Based on Attention-Enhancing and Feature Fusion for Retinal Vessel Segmentation." IEEE Access 11 (2023): 45871-45887.