Parkinson disease (PD) can significantly impair the quality of life of affected individuals. Especially, motor symptoms are challenging, as they worsen over time and increase the risk of falls and further injuries. This paper presents a wearable-based pro-adaptive assistive system (AS) for detecting PD-related symptoms and predicting disease progression. To this end, we present the initial structure and components of a pro-adaptive AS, which uses AI models to enable pro-adaptive decision-making based on the patient's condition. The developed wrist-worn wearable device uses an accelerometer to detect resting tremors and falls and can store sensor data locally on a microSD card. In the event of fall, this device can automatically emit a signal to alert caregivers. The proposed pro-adaptive AS enables adaptive assistance based on the patient's current condition.