Sleeveless blood pressure continuous monitoring method and system based on light guide film

By collecting multi-band reflected light signals from the fingertip skin through a photoconductive film and extracting dual-band pulse wave characteristics, combined with a personalized baseline model, the problem of cumbersome operation, large error, and ambient light interference in existing blood pressure measurement technologies is solved, achieving non-invasive, continuous, and accurate blood pressure monitoring.

CN122004813APending Publication Date: 2026-05-12常乐
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610350999.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-21
Publication Date
2026-05-12

AI Technical Summary

Technical Problem

Existing blood pressure measurement technologies suffer from problems such as cumbersome operation, inability to perform continuous monitoring, large errors, high costs, significant ambient light interference, and low signal-to-noise ratio, failing to meet the demand for non-invasive, continuous, and accurate blood pressure monitoring.

Method used

By using a photoconductive film to collect multi-band reflected light signals from the fingertip skin, and using a micro-nano grating structure to achieve directional light guidance and clutter filtering, the time difference between the peaks of the dual-band pulse wave and multi-dimensional pulse wave characteristics are extracted. Combined with a personalized baseline model, cuffless and imperceptible continuous blood pressure monitoring is achieved.

Benefits of technology

It achieves non-invasive, undetectable, and continuous blood pressure monitoring, reduces hardware costs, improves the signal-to-noise ratio, reduces ambient light interference, ensures measurement accuracy, and supports individual differences and privacy protection.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a cuff-free blood pressure continuous monitoring method and system based on a light guide film, multiband reflected light signals of fingertip skin are collected through the light guide film attached to a terminal screen, the light guide film realizes directional light guide and stray light filtering through a micro-nano grating array, the light guide film is aligned with screen pixel hardware, and the light guide film is connected with the terminal screen. And the signal to noise is obviously improved compared with a common camera scheme. According to the method, multi-dimensional features such as time difference, rise time and enhancement index between dual-band pulse wave crests are extracted, and systolic pressure and diastolic pressure are calculated in combination with a user personalized baseline model. The system supports dual modes of active measurement and night passive monitoring, motion artifacts are eliminated through an acceleration sensor, ambient light interference is inhibited through an ambient light sensor, and one-machine multi-person exclusive monitoring, local encrypted data storage and user autonomous management are realized through linkage with an applicant spectrum identity recognition patent. The blood pressure monitoring device does not need cuff inflation, does not need electrocardio reference, is free of oppression, can realize non-perception continuous blood pressure monitoring when a user uses a mobile phone daily, and has the advantages of being non-invasive, continuous, low in cost, high in precision and safe in privacy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the fields of optical acquisition, biometric monitoring, and health management technology, specifically to a cuffless continuous blood pressure monitoring method and system based on a photoconductive film. The method collects reflected light signals from the fingertip skin through a photoconductive film attached to the terminal screen, extracts the time difference between the peaks of the dual-band pulse wave and multi-dimensional pulse wave characteristic parameters, and combines them with the user's personal baseline data to achieve cuffless, imperceptible, and continuous blood pressure monitoring. It can be widely used in everyday terminal devices such as mobile phones and tablets. Citation application

[0002] This application is based on the applicant's previously undisclosed patent technology, specifically cited as follows: 1. Prior patent application 1: Application No. 2026102814971, Application Date 2026-03-10, Invention Title: A method for implementing permissionless AI interaction based on photoconductive tempered glass film; 2. Prior patent application 3: Application No. 2026103269949, Application Date 2026-03-17, Invention Title: Unauthorized Eye-tracking Touch Terminal Control Method and System Based on Optical Guide Interaction Layer; 3. Prior patent application: Application No. 2026103505829, Application Date 2026-03-20, Invention Title: A method and system for unique identification of human optical spectral features based on photoconductive film.

[0003] The aforementioned prior basic patents were all filed earlier than this application and had not been published before the filing date of this application. They do not constitute prior art for this application, and their core technical features are incorporated into this application by reference. Background Technology

[0004] Current blood pressure measurement technologies have the following drawbacks: Cuff-type blood pressure monitors: require wearing a cuff and inflating it, which is cumbersome to operate, cannot monitor continuously, and nighttime measurements can affect sleep. Furthermore, the pressure from the cuff can cause users to become nervous, leading to measurement errors. Wrist electronic blood pressure monitors still require inflation and pressurization, and have high requirements for measurement posture; the error is large when the wrist is not aligned with the heart. Wearable devices for blood pressure measurement rely on photoelectric sensors and electrocardiogram sensors, require dedicated hardware, and need regular calibration, resulting in high costs and difficulty in widespread adoption. Existing photoplethysmography technology can only measure heart rate and cannot accurately measure blood pressure, or it may require combining with electrocardiogram signals (dual sensors), which increases hardware complexity. Existing solutions for measuring blood pressure using mobile phone cameras rely on capturing video from the fingertip. These solutions are highly susceptible to ambient light interference, have a low signal-to-noise ratio, and cannot reliably extract weak pulse wave features. Consequently, their measurement accuracy fails to meet clinical reference requirements.

[0005] The applicant's previous patents have achieved passive optical acquisition technology using photoconductive films, which can acquire reflected light signals from the skin surface with high precision. Building on this, the present invention further utilizes the directional light guiding and clutter filtering characteristics of the photoconductive film to extract the time difference between the peaks of the dual-band pulse wave and multi-dimensional pulse wave features. Combined with a user-personalized baseline model, it achieves continuous blood pressure monitoring with a single sensor without the need for a cuff or ECG reference. Summary of the Invention

[0006] The purpose of this invention is to provide a cuffless continuous blood pressure monitoring method and system based on a photoconductive film. The method utilizes the photoconductive film to collect multi-band reflected light signals from the fingertip skin, extracts multi-dimensional features such as the time difference between peaks of the dual-band pulse wave, the pulse wave rise time, and the reflection wave enhancement index, and combines these features with a personalized baseline model constructed based on the user's age, height, weight, and historical blood pressure data to achieve non-invasive, unobtrusive, and continuous blood pressure monitoring. This can assist in the daily management of hypertensive patients and provide early warning of cardiovascular diseases.

[0007] S0: In response to user authorization, obtain the user's personal physiological parameters and construct or call a personalized blood pressure calculation model; the blood pressure monitoring supports user active measurement mode and nighttime passive monitoring mode; S1: Using a passive light guide film with a micro-nano grating structure, the system utilizes controllable multi-band light emitted from the terminal screen as a light source to collect spectral reflectance data of the user's fingertip skin in at least two different bands. The light guide film achieves directional light guidance and stray light filtering through a micro-nano grating array, resisting ambient light interference and significantly improving signal-to-noise ratio compared to ordinary camera acquisition schemes. The micro-nano grating array of the light guide film is hardware-level aligned with the screen pixels through photolithography, with an alignment error of less than 0.01 pixels. The acquisition location is the fingertip area of ​​the index or middle finger. S2: Extract pulse wave signals from the collected spectral reflectance data. Use a built-in accelerometer to detect motion artifacts. When the detected finger movement acceleration exceeds 0.1g, automatically discard the current data segment. Use an ambient light sensor to detect sudden changes in ambient light. When the ambient light intensity change exceeds a preset threshold, automatically trigger signal compensation or discard the interfered data segment. Calculate multi-dimensional pulse wave characteristic parameters, including at least one of the following: inter-peak time difference of dual-band pulse waves, pulse wave rise time, reflection wave enhancement index, and pulse wave area integral. The inter-peak time difference of dual-band pulse waves is defined as the time difference between the peak time of the main peak of the green band signal and the peak time of the main peak of the red band signal. This time difference is negatively correlated with systolic blood pressure, unlike the traditional ECG-pulse wave conduction time. S3: Input the extracted multi-dimensional pulse wave feature parameters into the personalized blood pressure calculation model. The personalized blood pressure calculation model takes the time difference between the peaks of the dual-band pulse wave, heart rate, age, and body mass index as mandatory input parameters, and the pulse wave rise time, reflection wave enhancement index, and ambient temperature as optional input parameters, and outputs estimated systolic and diastolic blood pressure. S4: Perform Kalman filtering fusion on multiple consecutive measurement results to output smooth blood pressure trend data. When multiple consecutive measurement results deviate from the individual baseline by more than a preset threshold, an abnormal warning is triggered, and the warning information is pushed to the user terminal and authorized contacts. S5: Blood pressure measurement results are stored locally on the terminal using AES-128 encryption, with the key stored in the terminal's secure area and not uploaded to any cloud server. Users can choose whether to synchronize to an authorized cloud server; a long-term blood pressure trend report is generated.

[0008] Photoconductive film: A passive optical film with a micro-nano grating structure, attached to the surface of the terminal screen. Its micro-nano grating array is aligned with the screen pixels at the hardware level through photolithography, with an alignment error of less than 0.01 pixels. It is used to collect multi-band reflected light signals from the user's fingertip skin through directional light guiding and stray light filtering. Signal acquisition module: It uses controllable multi-band light emitted from the terminal screen as a light source to continuously acquire reflected light signals at a sampling frequency of over 100Hz and extract raw pulse wave data. Feature extraction module: used to calculate multi-dimensional feature parameters from pulse wave signals, such as the time difference between peaks of the dual-band pulse wave, the rise time of the pulse wave, the reflection wave enhancement index, and the pulse wave area integral. Blood pressure calculation module: Built-in personalized blood pressure calculation model, with the time difference between the peaks of the dual-band pulse wave, heart rate, age, and body mass index as mandatory input parameters, and the pulse wave rise time and reflection wave enhancement index as optional input parameters, outputting estimated systolic and diastolic blood pressure; Calibration Management Module: Used to manage the user's initial calibration and periodic calibration processes, and maintain the user's personalized model coefficients; Data storage and display module: Used to store historical blood pressure data locally on the terminal using AES-128 encryption. The key is stored in the terminal's secure area and is not uploaded to any cloud server. Users can choose whether to synchronize to an authorized cloud server, generate trend reports, and support anomaly alerts and data sharing. Identity recognition module: It is used to collect the optical spectral features of the user's human body through the photoconductive film, compare them with the pre-stored spectral feature code to verify the user's identity. If the verification is successful, the blood pressure of the user is measured. If the verification fails, the measurement is refused and an unauthorized user is prompted, so as to realize one device for multiple people to have exclusive monitoring and privacy isolation. Detailed Implementation

[0009] The fundamental difference between the photoconductive film of this invention and ordinary camera acquisition schemes lies in the following: Ordinary cameras capture diffuse reflection stray light from the skin surface, which is subject to strong ambient light interference and low signal-to-noise ratio, making it impossible to stably extract weak pulse wave features. In contrast, this invention achieves directional light guidance, spatial filtering, and stray light suppression through hardware-level alignment of a micro / nano grating array with screen pixels (alignment error <0.01 pixels). It only acquires the effective signal modulated at the same frequency as the screen light source, achieving a significantly improved signal-to-noise ratio compared to ordinary camera schemes under screen light conditions, thereby enabling the extraction of weak pulse wave features required for blood pressure measurement. This technical effect is the core inventive aspect that distinguishes this invention from existing mobile phone camera blood pressure measurement schemes.

[0010] Upon first use, users enter basic information such as age, height, and weight through the terminal app and, following the instructions, simultaneously measure their blood pressure three times in a resting state using a standard cuff blood pressure monitor. The average value is then used as a baseline. Simultaneously, the system collects fingertip pulse wave signals via a photoconductive membrane. The baseline blood pressure value is then regressed and fitted with the synchronous pulse wave characteristics to establish a personalized blood pressure calculation model for each user. Users can check the model status at any time through the app. The system sends calibration reminders every 3-6 months, recommending recalibration to adapt to changes in physiological state. Blood pressure monitoring supports active measurement mode (finger actively placed in a designated area of ​​the screen) and passive nighttime monitoring mode (phone placed next to the pillow, automatically detecting fingertip contact and measuring).

[0011] The user places a smartphone with the photoconductor film applied flat on a table and gently places the tip of their index or middle finger on the designated area above the screen (the screen displays a guide light spot). The micro-nano grating array of the photoconductor film is hardware-level aligned with the screen pixels (alignment error <0.01 pixels), ensuring spatial resolution for signal acquisition. The system emits dual-band light (green, 540nm; red, 660nm) from the screen, which is uniformly conducted to the fingertip through the photoconductor film. The photoconductor film uses a micro-nano grating array to achieve directional light guidance and stray light filtering, collecting only the effective signal modulated at the same frequency as the screen light source, suppressing ambient light interference. The photoconductor film continuously collects the reflected light signal from the fingertip at a sampling frequency of 200Hz for 30 seconds.

[0012] During data acquisition, the system's built-in accelerometer detects finger movement in real time. When the acceleration exceeds 0.1g, the data segment is automatically discarded, and the user is prompted to remain stable. The ambient light sensor detects sudden changes in ambient light. When the light intensity change exceeds a preset threshold, signal compensation is automatically triggered or the interfered data segment is discarded. After the signal stabilizes, the feature extraction module identifies pulse wave feature points and calculates the peak time difference between the green and red bands as the inter-peak time difference of the dual-band pulse wave. Simultaneously, it calculates multi-dimensional feature parameters such as pulse wave rise time (the time interval from the pulse wave's inception to the peak value of the main peak), reflection enhancement index (the ratio of the peak value of the reflected wave after the main peak to the peak value of the main peak), and pulse wave area integral.

[0013] The blood pressure calculation module uses the user's personalized model and requires the time difference between the peaks of the dual-band pulse wave, heart rate, age, and body mass index as mandatory input parameters. It also combines optional parameters such as pulse wave rise time and reflection wave enhancement index to output estimated systolic and diastolic blood pressure.

[0014] The system performs Kalman filtering on multiple consecutive measurements to output smooth blood pressure trend data. When three consecutive measurements show a systolic blood pressure ≥140 mmHg or a diastolic blood pressure ≥90 mmHg, the system automatically sends an alert to the user's mobile phone and preset emergency contacts (such as family members or family doctors).

[0015] All measurement data is stored locally on the terminal using AES-128 encryption, with the key stored in a secure area on the terminal and not uploaded to any cloud server. Users can choose whether to synchronize encrypted data to an authorized cloud service (such as a family doctor or hospital health management platform). Long-term blood pressure trend reports are generated, allowing users to view historical data and export and share it.

[0016] Before each measurement, the system collects the optical spectral characteristics of the user's fingertip through a photoconductive film and compares them with pre-stored spectral feature codes to verify the user's identity. If verification is successful, the system uses the user's personalized blood pressure model for measurement; if verification fails, the system prompts "Unauthorized user, please use your own identity verification before measurement" and refuses the measurement. This mechanism ensures that blood pressure data for each family member is stored independently and without interference in scenarios where multiple people use the same device, and prevents data confusion caused by misuse by others.

[0017] Signal quality detection: The signal-to-noise ratio is calculated in real time during the acquisition process. When the signal-to-noise ratio is lower than the threshold, the user is prompted to adjust the finger pressure or clean the screen. Motion artifact removal: The system detects finger movement using an accelerometer. When the acceleration exceeds 0.1g, the data segment is automatically removed, prompting the user to remain stable. Ambient light interference suppression: By detecting sudden changes in light intensity through an ambient light sensor, signal compensation or removal of interfered data segments is automatically triggered; Calibration reminder: When more than 6 months have passed since the last calibration, a calibration reminder will be proactively pushed to guide the user to recalibrate using a standard cuff blood pressure monitor; Photoconductor film status detection: When the photoconductor film is not properly attached to the screen or is dirty, causing signal attenuation, it will automatically prompt the user to clean or reattach it; Nighttime contact detection: In the passive monitoring mode at night, the continuous contact status of the fingertip is determined by the change in the intensity of reflected light. Measurement is only started when there is stable contact and is paused when the contact is interrupted. Beneficial effects

[0018] Cuffless and non-invasive: No need to wear a cuff for inflation and pressure, no pressure, and high user acceptance; Continuous monitoring: Enables seamless and continuous monitoring during daily mobile phone use, capturing patterns of blood pressure fluctuations; Single-sensor solution: Only one sensor, the photoconductive film, is required, eliminating the need for ECG electrodes or other reference signals, resulting in low hardware costs; Personalized model: A user-specific model is established through the initial calibration, and subsequent measurements automatically adapt to individual differences; Multi-band anti-interference: Employs green and red dual-band acquisition to effectively correct for individual differences such as skin color and tissue thickness; Hardware-level precision: The photoconductor film is hardware aligned with the screen pixels (alignment error <0.01 pixels) to ensure signal acquisition accuracy; Directional light guiding and stray light filtering: Directional light guiding and stray light suppression are achieved through micro-nano grating arrays, which significantly improves the signal-to-noise ratio compared to ordinary camera solutions. This is the core innovation that distinguishes it from existing mobile phone camera blood pressure measurement solutions. Motion artifact removal and ambient light compensation: With the assistance of accelerometer and ambient light sensor, motion interference and sudden ambient light interference are effectively removed, improving measurement accuracy; Identity recognition linkage: Linked with the applicant's spectral identity recognition patent, it enables one device to monitor multiple users and isolate their privacy; Privacy and security: Measurement data is stored locally in encrypted form, managed independently by the user, and complies with medical and health data compliance requirements.

Claims

1. A cuffless continuous blood pressure monitoring method based on a photoconductive film, characterized in that, Includes the following steps: S0: In response to user authorization, obtain the user's personal physiological parameters and construct or call a personalized blood pressure calculation model; the blood pressure monitoring supports user active measurement mode and nighttime passive monitoring mode; S1: Using a passive light guide film with a micro-nano grating structure, the system utilizes controllable multi-band light emitted from the terminal screen as a light source to collect spectral reflectance data of the user's fingertip skin in at least two different bands. The light guide film achieves directional light guidance and stray light filtering through a micro-nano grating array, resisting ambient light interference and significantly improving signal-to-noise ratio compared to ordinary camera acquisition schemes. The micro-nano grating array of the light guide film is hardware-level aligned with the screen pixels through photolithography, with an alignment error of less than 0.01 pixels. The acquisition location is the fingertip area of ​​the index or middle finger. S2: Extract pulse wave signals from the collected spectral reflectance data. Use a built-in accelerometer to detect motion artifacts. When the detected finger movement acceleration exceeds 0.1g, automatically discard the current data segment. Use an ambient light sensor to detect sudden changes in ambient light. When the ambient light intensity change exceeds a preset threshold, automatically trigger signal compensation or discard the interfered data segment. Calculate multi-dimensional pulse wave characteristic parameters, including at least one of the following: inter-peak time difference of dual-band pulse waves, pulse wave rise time, reflection wave enhancement index, and pulse wave area integral. The inter-peak time difference of dual-band pulse waves is defined as the time difference between the peak time of the main peak of the green band signal and the peak time of the main peak of the red band signal. This time difference is negatively correlated with systolic blood pressure, unlike the traditional ECG-pulse wave conduction time. S3: Input the extracted multi-dimensional pulse wave feature parameters into the personalized blood pressure calculation model. The personalized blood pressure calculation model takes the time difference between the peaks of the dual-band pulse wave, heart rate, age, and body mass index as mandatory input parameters, and the pulse wave rise time, reflection wave enhancement index, and ambient temperature as optional input parameters, and outputs estimated systolic and diastolic blood pressure. S4: Perform Kalman filtering fusion on multiple consecutive measurement results to output smooth blood pressure trend data. When multiple consecutive measurement results deviate from the individual baseline by more than a preset threshold, an abnormal warning is triggered, and the warning information is pushed to the user terminal and authorized contacts. S5: Blood pressure measurement results are stored locally on the terminal using AES-128 encryption, with the key stored in the terminal's secure area and not uploaded to any cloud server. Users can choose whether to synchronize to an authorized cloud server; a long-term blood pressure trend report is generated.

2. The method according to claim 1, characterized in that, In step S1, the at least two different bands include a combination of a green band of 520-580nm and a red band of 640-700nm, or a combination of a green band of 520-580nm and a near-infrared band of 800-950nm.

3. The method according to claim 1, characterized in that, In the passive nighttime monitoring mode, the continuous contact status of the fingertip is determined by the change in reflected light intensity. When the reflected light intensity remains stable and higher than the preset threshold for 5 consecutive seconds, it is determined to be a stable contact and the measurement is started. The measurement is paused when the contact is interrupted.

4. The method according to claim 1, characterized in that, The personalized blood pressure calculation model is established by regression fitting of the user's standard blood pressure value during the initial calibration and the pulse wave characteristics collected simultaneously. The initial calibration uses a cuff blood pressure monitor to measure three times simultaneously in a resting state and take the average value as the baseline value. It is recommended that users recalibrate every 3-6 months thereafter.

5. A cuffless continuous blood pressure monitoring system based on a photoconductive film, characterized in that, include: Photoconductive film: A passive optical film with a micro-nano grating structure, attached to the surface of the terminal screen. Its micro-nano grating array is aligned with the screen pixels at the hardware level through photolithography, with an alignment error of less than 0.01 pixels. It is used to collect multi-band reflected light signals from the user's fingertip skin through directional light guiding and stray light filtering. Signal acquisition module: It uses controllable multi-band light emitted from the terminal screen as a light source to continuously acquire reflected light signals at a sampling frequency of over 100Hz and extract raw pulse wave data. Feature extraction module: used to calculate multi-dimensional feature parameters from pulse wave signals, such as the time difference between peaks of the dual-band pulse wave, the rise time of the pulse wave, the reflection wave enhancement index, and the pulse wave area integral. Blood pressure calculation module: Built-in personalized blood pressure calculation model, with the time difference between the peaks of the dual-band pulse wave, heart rate, age, and body mass index as mandatory input parameters, and the pulse wave rise time and reflection wave enhancement index as optional input parameters, outputting estimated systolic and diastolic blood pressure; Calibration Management Module: Used to manage the user's initial calibration and periodic calibration processes, and maintain the user's personalized model coefficients; Data storage and display module: Used to store historical blood pressure data locally on the terminal using AES-128 encryption. The key is stored in the terminal's secure area and is not uploaded to any cloud server. Users can choose whether to synchronize to an authorized cloud server, generate trend reports, and support anomaly alerts and data sharing. Identity recognition module: It is used to collect the optical spectral features of the user's human body through the photoconductive film, compare them with the pre-stored spectral feature code to verify the user's identity. If the verification is successful, the blood pressure of the user is measured. If the verification fails, the measurement is refused and an unauthorized user is prompted, so as to realize one device for multiple people to have exclusive monitoring and privacy isolation.