Ultra-low-cost mechanical smartphone attachment for no-calibration blood pressure measurement

Yinan Xuan, Colin Barry, Jéssica de Souza, Jessica H. Wen, Nick Antipa, Alison Moore, Edward Jay WangScientific 2023 · DOI

TL;DR

This paper introduces BPClip, an ultra-low-cost mechanical attachment for smartphones that enables cuffless blood pressure measurement without per-user calibration. By leveraging a smartphone's camera and flash, BPClip measures oscillometric signals at the fingertip, offering a potentially democratized approach to at-home hypertension management.

What Problem Does This Paper Address?

Hypertension is a widespread and serious health issue, yet access to affordable and convenient blood pressure monitoring remains a challenge, particularly for low-income and underserved communities. Existing cuff-based monitors can be expensive and cumbersome for daily use. While cuffless solutions exist, many rely on complex algorithms like pulse transit time (PTT) or pulse wave analysis (PWA), which necessitate frequent and personalized calibration against a traditional cuff. This calibration requirement creates a barrier to widespread adoption, as users still need access to cuff-based devices. The authors aim to bridge this gap by developing a truly low-cost, calibration-free solution that utilizes ubiquitous smartphone technology.

Methods

The core innovation of BPClip lies in its mechanical design and its reliance on basic smartphone hardware. The system uses a 3D-printed plastic clip that attaches to a smartphone. This clip houses a spring-loaded mechanism with a pinhole projection. When a user presses their fingertip against the clip, the spring compresses, and the size of the pinhole projection changes, directly encoding the applied pressure. Simultaneously, the smartphone's flash illuminates the fingertip, and the camera captures the reflected light. Fluctuations in the brightness of the projected pinhole image correlate with arterial pulse amplitude.

The system operates on the principle of oscillometry, a standard method for BP measurement. By capturing both the applied pressure (via pinhole size) and the arterial pulse amplitude (via brightness fluctuations) simultaneously, the smartphone app can calculate blood pressure. A key aspect is the "no-calibration" requirement. This is achieved through an initial brightness adjustment step where the user applies minimal and maximal force, allowing the system to normalize the signal regardless of skin tone or flashlight brightness. The app then guides the user through 20 discrete force levels, collecting data at each to build a subsampled oscillogram for BP estimation.

Datasets

The study utilized data from 29 participants for feasibility testing and 24 participants for accuracy validation. The reference blood pressure measurements were obtained using a standard arm cuff device (Omron, model: BP7350). No new benchmarks were introduced; rather, the study focused on validating the BPClip system against established cuff-based measurements.

Key Insights and Ideas

  • Democratizing BP Measurement: The most significant insight is the potential to make accurate BP monitoring accessible to virtually anyone with a smartphone, regardless of their socioeconomic status. The ultra-low material cost ($0.80 per unit at low production) is a game-changer.
  • Oscillometry Reimagined: The paper ingeniously adapts the established oscillometric principle to a lens-less optical system using only a smartphone camera and flash, bypassing the need for expensive pressure sensors found in traditional cuffs.
  • The Power of Mechanical Encoding: The clever use of a spring-loaded pinhole to mechanically encode applied pressure is a novel approach that simplifies the hardware requirements and eliminates the need for complex electronic pressure sensors.
  • True Calibration-Free Potential: The initial brightness adjustment step, combined with the mechanical pressure encoding, appears to effectively circumvent the need for per-user calibration, a major hurdle for many existing cuffless technologies.

Key Results

  • Accuracy: In a validation study with 24 participants, BPClip achieved a mean absolute error (MAE) of 8.72 mmHg for systolic blood pressure (SBP) and 5.49 mmHg for diastolic blood pressure (DBP).
  • Cost-Effectiveness: The estimated material cost for a single BPClip unit is $0.80 in small-batch manufacturing (1000 units), with projections for even lower costs at scale. This is significantly cheaper than existing cuff-based monitors.
  • Usability: Measurements took an average of 212-251 seconds, and the system was found to be usable by both left-handed and right-handed individuals.

Limitations and Future Work

The authors acknowledge several limitations and suggest future directions:

  • Perfusion Dependency: The system requires adequate blood circulation in the finger. Four participants were excluded due to insufficient perfusion, even after hand warming. Future work could explore methods to improve blood perfusion or adapt the system for individuals with poor circulation.
  • Dexterity Requirements: The current design requires a certain level of finger dexterity to maintain consistent pressure. Future iterations could explore designs that are easier to use for individuals with tremors or weakened grip, potentially by using a higher stiffness spring or allowing thumb operation.
  • Finger Surface Mechanics: Factors like finger calluses or variations in skin mechanics could affect measurement accuracy. Further studies are needed to characterize these effects and develop models to address them.
  • Light Guide Design: While the current design accommodates a range of flashlight-to-camera distances, future work could explore using the front-facing camera and screen illumination for even greater universality, though this might require more sophisticated optical designs to compensate for lower screen brightness.
  • Spring Consistency: Variations in spring manufacturing could introduce bias. The authors suggest implementing a QR-code based calibration system at the manufacturing level to address this.

Summary updated

Abstract

We propose an ultra-low-cost at-home blood pressure monitor that leverages a plastic clip with a spring-loaded mechanism to enable a smartphone with a flash LED and camera to measure blood pressure. Our system, called BPClip, is based on the scientific premise of measuring oscillometry at the fingertip to measure blood pressure. To enable a smartphone to measure the pressure applied to the digital artery, a moveable pinhole projection moves closer to the camera as the user presses down on the clip with increased force. As a user presses on the device with increased force, the spring-loaded mechanism compresses. The size of the pinhole thus encodes the pressure applied to the finger. In conjunction, the brightness fluctuation of the pinhole projection correlates to the arterial pulse amplitude. By capturing the size and brightness of the pinhole projection with the built-in camera, the smartphone can measure a user's blood pressure with only a low-cost, plastic clip and an app. Unlike prior approaches, this system does not require a blood pressure cuff measurement for a user-specific calibration compared to pulse transit time and pulse wave analysis based blood pressure monitoring solutions. Our solution also does not require specialized smartphone models with custom sensors. Our early feasibility finding demonstrates that in a validation study with N = 29 participants with systolic blood pressures ranging from 88 to 157 mmHg, the BPClip system can achieve a mean absolute error of 8.72 and 5.49 for systolic and diastolic blood pressure, respectively. In an estimated cost projection study, we demonstrate that in small-batch manufacturing of 1000 units, the material cost is an estimated $0.80, suggesting that at full-scale production, our proposed BPClip concept can be produced at very low cost compared to existing cuff-based monitors for at-home blood pressure management.
BibTeX
@inproceedings{xuan2023ultra,
  author = {Yinan Xuan and Colin Barry and Jéssica de Souza and Jessica H. Wen and Nick Antipa and Alison Moore and Edward Jay Wang},
  title = {{Ultra-low-cost mechanical smartphone attachment for no-calibration blood pressure measurement}},
  booktitle = {Scientific},
  year = {2023},
  doi = {10.1038/s41598-023-34431-1},
}
Figure