Cardiac Output Monitoring via an Automatic Arm Cuff Device: Potential in Surgical Patients
Mahdi Jazini, Hadi Daher, Ravinder Kumar, Vishaal Dhamotharan, Abhijeet M. Sathe, Yaroslava Longhitano, Ezeldeen Abuelkasem, Anand Chandrasekhar, Michael R. Pinsky, Kathirvel Subramaniam, Raymond M. Planinsic, Jin-Oh Hahn, Sanjeev G. Shroff, Kimberly Howard-Quijano, Aman Mahajan, Ramakrishna Mukkamala
medRxiv 2025.10.09.25337689; doi: https://doi.org/10.1101/2025.10.09.25337689.
In this study, we investigated whether standard automatic arm cuff devices—widely used for non-invasive blood pressure monitoring—can also estimate cardiac output (CO) by analyzing cuff pressure oscillations that reflect pulsatile blood volume. Using a custom-designed cuff capable of measuring both pressure and air-volume changes during deflation, we collected data from 34 surgical patients and applied a simple formula combining heart rate, oscillation amplitude, cuff-arm compliance, and a body-to-arm surface area ratio. Our cuff-based CO estimates demonstrated moderate accuracy (correlation 0.60; 83% concordance) compared with thermodilution, performing similarly to traditional pulse contour analysis of invasive arterial pressure (correlation 0.62; 81% concordance). These results suggest that automatic arm cuffs may provide a promising non-invasive method for CO monitoring in surgical patients.
A popular validated home monitor uses the maximum oscillogram amplitude to compute blood pressure
Dhamotharan, V., Jazini, M., Kumar, R. et al. A popular validated home monitor uses the maximum oscillogram amplitude to compute blood pressure. Sci Rep 15, 35095 (2025). https://doi.org/10.1038/s41598-025-18850-w
In this work, we examined how oscillometric cuff devices estimate blood pressure (BP), despite variations in cuff placement and pulse pressure that can distort the oscillogram. Using a validated home BP monitor with its Universal cuff on four mandrels, we generated oscillograms with a non-invasive BP simulator and an external pressure sensor. We then tested multiple algorithms to replicate the monitor’s BP outputs. A traditional fixed-ratio method produced systolic and diastolic BP errors of 5.8 and 1.5 mmHg, respectively. In contrast, a variable-ratio method—where ratios depend inversely on the maximum oscillogram amplitude—reduced errors to 1.5 and 0.8 mmHg. Our findings suggest that using oscillogram amplitude helps compensate for pulse-pressure variability and provide insight into how oscillometric devices likely operate, which is critical for improving their accuracy.
Md A. Momin, M. Jazini, M. J. Rahman, and T. Mieno, Analysis & Sensing, e202400062, 2024.
Our study explores innovative foot pressure sensors using carbon nanotube (CNT)-coated cotton fibers, designed to detect and analyze human activities. These sensors demonstrate sensitivity, repeatability, and durability, confirmed through Python simulations and rigorous testing. They respond nonlinearly to pressure, influenced by fiber contact area, and withstand long-term outdoor exposure while retaining conductivity. The sensors can monitor activities like walking, running, and even muscle movements, showing potential for health monitoring and artificial voice synthesis. Using the MRMR algorithm, we also achieve activity recognition, and a piezoelectric sensor within the system estimates harvested power, advancing self-powered wearable technology for healthcare and assistive applications.
A Novel Combination of Neural Networks and FFT for Frequency Estimation of SAW Resonators’ Responses
M. M. Jazini, M. Khoshakhlagh and N. Masoumi, 2019 5th Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS), Shahrood, Iran, 2019, pp. 1-5
In this work, we focused on improving the accuracy of resonant frequency estimation in surface acoustic wave (SAW) resonator sensors. These sensors transmit damped resonant responses to a reader, which must quickly and accurately detect the resonant frequency to determine the measurand. Although Discrete Fourier Transform (DFT) techniques—enhanced with zero-padding, parabolic interpolation, and windowing—can estimate this frequency, they are highly sensitive to noise and distortions from long channels or high-frequency blocks. To address this limitation, we developed a noise-resistant neural-network–based frequency estimation method capable of maintaining high accuracy in harsh or noisy environments. We validated the approach through measurements on an experimental setup and demonstrated reliable performance by implementing the method on a microcontroller.
A new frequency detection method based on FFT in the application of SAW resonator sensor
M. M. Jazini, M. Khoshakhlagh, and N. Masoumi, in Proc. Iranian Conference on Electrical Engineering (ICEE), pp. 232-237, May 2018.
Our research highlights the advantages of passive Surface Acoustic Wave (SAW) sensors in wireless, battery-free measurement systems, powered by pulse signals from an interrogation unit. The SAW sensor emits a noisy, damped sine wave in response, resonating at a frequency that varies based on sensing parameters. The challenge lies in accurately detecting this frequency to measure the parameter precisely. While traditional methods like FFT, zero crossing, and peak detection are commonly used, they suffer from limitations in accuracy and speed. To address this, we developed a new frequency detection approach based on FFT, which improves performance and reduces the need for high ADC sampling rates in the interrogation unit. Our method also accurately detects the initial phase and amplitude of the response, offering a robust solution for SAW sensor applications.
Limitations of Fourier Transform Analysis in the Wireless SAW Sensor Systems
M. M. Jazini and N. Masoumi, Mar. 2020.
This study explores the unique capabilities of Surface Acoustic Wave Resonator (SAWR) sensors, known for their wireless, battery-free operation in measuring temperature, pressure, and torque across various environments. We conduct a theoretical analysis of SAWR measurement using Fast Fourier Transform (FFT) and validate our findings experimentally. The paper also reviews key components of the measurement unit, focusing on the frequency response of SAWR signals processed by an ADC and FFT-capable processor. Our results provide guidance for selecting optimal local oscillator frequencies in RF systems, enhancing measurement accuracy while reducing costs. Unlike previous approaches that chose intermediate frequencies (IF) experimentally, we propose new formulae to define theoretical bounds for the IF frequency region.
