Severe mental disorders such as schizophrenia can cause autonomic nervous system dysfunction, leading to hemodynamic changes including increased heart rate, elevated blood pressure, and decreased heart rate variability. These physiological changes often appear before hallucinations, indicating the need for early detection technology based on physiological signals to prevent symptom escalation. This study developed the Hemodynamic Early Warning System (Alert Therapy Hemodynamic) to detect pre-hallucination hemodynamic patterns and to strengthen patient self-efficacy. The study used a simplified Borg & Gall Research and Development (R&D) model, consisting of five stages: needs identification, system design, prototype development, user testing, and revision. The prototype was created through multidisciplinary collaboration and includes multimodal sensors (HR, HRV, blood pressure), a pattern-detection algorithm, and a biofeedback module. Initial user testing evaluated function, comfort, and usability using the Functionality, Usability, Look-and-Feel, Evaluation (FULE) approach. The results showed that the pre-hallucination hemodynamic patterns were consistent with the literature, and the prototype functioned well and was accepted by users. This innovation is important for nurses, who play a key role in early detection of patient condition changes and in preventing symptom worsening. The system has the potential to improve recognition of early physical signs, enhance patient self-control, and increas clinical safety. Considering this urgency, future studies should include pre-clinical and limited clinical trials to assess the effectiveness, efficiency, and safety of this real-time HR/HRV-based pre-hallucination early detection tool in real patient populations.