Personalized respiratory guidance strategy via millimeter-wave radar-based breathing consistency score

Key respiratory parameters such as inhalation-to-exhalation ratio and phase durations are known to influence physiological outcomes. However, most existing respiratory guidance systems rely on frequency-driven instructions, do not evaluate whether the user's breathing follows the intended waveform, and depend on contact sensors that can cause discomfort during prolonged use. To address these limitations, we propose a millimeter-wave radar-based personalized guidance strategy driven by real-time respiratory consistency feedback. Resting and 4-7-8 breathing signals from 20 participants were used to construct individualized progressive guidance curves via a sinusoidal-based segmented motion model fitted by the Grey Wolf Optimizer. A segmented dynamic time warping algorithm with duration-deviation penalty was developed to compute the Breathing Consistency Score (BCS) as a real-time feedback signal. A controlled crossover experiment compared BCS-driven guidance against frequency-driven guidance in terms of guidance efficiency, waveform-following consistency, and stress reduction, with subjective comfort additionally assessed through open-ended feedback. BCS-driven guidance significantly improved waveform-following consistency, reduced target acquisition time, and achieved greater stress reduction. Most participants also reported greater comfort. This real-time consistency feedback solution can help individuals with stress-related conditions such as anxiety disorders or post‑traumatic stress disorder (PTSD) better utilize respiratory guidance for stress regulation.