Parkinson’s Disease Prediction using PD-Net: A Transformed based Approach

Parkinson’s disease is a long-term condition that affects a person’s ability to move and perform daily tasks. Detecting it early is very important because it helps doctors provide better treatment and support before symptoms become severe. However, early signs are often small and easy to miss. In this work, we introduce a method called PD-Net that helps in recognizing these early signs by observing how a person walks, talks, and moves over time. The system looks for patterns or changes in these activities that are commonly linked to Parkinson’s. It carefully watches for movements that stand out or repeat in a specific way and uses that information to suggest whether a person might be at risk. The method was tested using real examples and compared with other existing tools. PD-Net gave more accurate and reliable results, correctly identifying warning signs more often. It also helps explain which actions were important in making that decision, making it easier for doctors to understand and trust the system. This approach has the potential to be used in real healthcare settings and may support doctors in providing earlier and more accurate diagnoses. In the future, it can be improved further by including more health-related information and working with wearable devices