Although Human-Computer Interaction (HCI) has advanced in supporting collaborative technologies, there is still a gap in applying these innovations to address animal health challenges, particularly for cattle disease management. Lumpy Skin Disease (LSD), Foot and Mouth Disease (FMD), and Infectious Bovine Keratoconjunctivitis (IBK) are prevalent non-invasive diseases affecting cattle in Bangladesh and are considered highly contagious. To promote sustainable cities and societies in Bangladesh, timely detection and intervention for these diseases are essential. To address this, we conducted in-person interviews with 26 cattle farms having around 1300 cattle to understand their needs and perspectives on managing non-invasive cattle diseases. This study introduces a novel detection system using a deep Convolutional Neural Network (CNN) with 99% accuracy for identifying these diseases. Additionally, we designed a mobile application, ‘CattleSavior’, which integrates the detection system and six other useful features to enable farmers to detect diseases instantly by capturing an image of the affected area. A follow-up interview with 11 farms was conducted to assess the app’s usability based on farmer feedback. Our work adds value to the HCI field by providing a practical, collaborative solution for cattle farmers, fostering better animal health management within agricultural communities.