Comprehending the regulatory mechanisms influencing blood pressure control is pivotal for continuous monitoring of this parameter. Implementing a personalized machine learning model, utilizing data-driven features, presents an opportunity to facilitate tracking blood pressure fluctuations in various conditions. In this work, data-driven photoplethysmograph features extracted from the brachial and digital arteries of 28 healthy subjects were used to feed a random forest classifier in an attempt to develop a system capable of tracking blood pressure. We evaluated the behavior of this latter classifier according to the different sizes of the training set and degrees of personalization used. Aggregated accuracy, precision, recall, and F1-score were equal to 95.1%, 95.2%, 95%, and 95.4% when 30% of a target subject’s pulse waveforms were combined with five randomly selected source subjects available in the dataset. Experimental findings illustrated that incorporating a pre-training stage with data from different subjects made it viable to discern morphological distinctions in beat-to-beat pulse waveforms under conditions of cognitive or physical workload.

This work introduces a novel wireless sensor for clinical Pulse Wave Velocity (PWV) assessment. Two pen-shaped acquisition probes have been designed to incorporate a custom printed circuit board (PCB), a rechargeable battery, and a commercial piezoresistive load cell as the sensing element. By using the Bluetooth Low Energy (BLE) 5.2 communication protocol, each probe transmits collected data to the USB dongle integrated into the receiving station realized for this application. A graphical user interface (GUI) has been designed to provide clinicians with real-time PWV values and feedback regarding the quality of the acquired signal. To assess the correct behavior of all system units, extensive validation of the developed firmware routines for data transfer and handling was conducted. The reliability of the PWV estimation was assessed in a pre-clinical study against the gold-standard device for noninvasive PWV (SphygmoCor), showing a robust linear correlation yielding a value of R ² =0.89. The Bland-Altman plot computed w.r.t the reference system reports a mean difference equal to -0.65 ± 0.41 m/s. Finally, electromagnetic compatibility and clinical usability tests were conducted to determine the device’s pre-compliance, according to the clinical safety rules and guidelines described in the IEC-60601 standard for medical devices.