The Chinese University of Hong Kong

The General Hospital Corporation

Hong Centre for Cerebro-cardiovascular Health Engineering Limited

Yuanting Zhang

Ni Zhao

Nan Ji

Quan Zhang

Jing Liu

Peng Su

Shanshan Yuan

Ting Xiang

Xiaorong Ding

Zijun Liu

Wenxuan Dai

Yao Li

The present invention relates to method and apparatus for heart-level blood pressure estimation based on ensemble learning. The invention relates to the technical field of biomedical technology. The method comprises: acquiring radial arterial blood pressure related information and a hydrostatic pressure of the radial artery of a subject; performing blood pressure conversion based on the radial arterial blood pressure related information and the hydrostatic pressure of the radial artery to obtain at least two brachial arterial blood pressure values; inputting the at least two brachial arterial blood pressure values into a trained correction model to obtain a target brachial arterial blood pressure value. The present invention realizes the conversion from a peripheral blood pressure, such as a radial arterial blood pressure measured by a wearable device, to a heart-level blood pressure, i.e. a brachial arterial blood pressure for clinical use, and performs feature fusion and a feature conversion on the feature information of each brachial arterial blood pressure value, so that the determined target brachial arterial blood pressure value becomes more accurate, and improves the accuracy and stability of the blood pressure estimation.

Method and apparatus for heart-level blood pressure estimation based on ensemble learning

The present invention relates to a multi-modal sensor module, comprising an electrode base, an insulating film sheet, a functional layer and a protective film; the protective film is used for isolating external environment with a side away from the functional layer for use in contacting surface of biological tissues; the functional layer comprises a substrate and at least one sensor module provided in isolation in the substrate; a side of the electrode base away from the insulating film sheet is used for contacting surface of the biological tissues; the multi-modal sensor module is in a state of collecting electrocardiogram (ECG) signal, a side of the protective film away from the functional layer and a side of the electrode base away from the insulating film sheet are both for contacting surface of the biological tissues, so that the protective film, the biological tissues and the electrode base form an ECG signal collection path. The present invention integrates each sensor module in the same functional layer for measuring various physiological signals or estimating arterial blood pressure, and the use of the electrode base and the protective film coordinates collection of the ECG signal of the biological tissues. The design is clever and the structure of the multi-modal sensor module is simplified.

Multi-modal sensor module and a multi-modal physiological signal monitoring system

A system and method for measuring and monitoring blood pressure is provided. The system includes a wearable device and a tonometry device coupled to the wearable device. The Tonometry device is configured to compress a superficial temporal artery (STA] of a user. A sensor pad is attached to the wearable device adjacent the tonometry device. A blood pressure sensor is integrated within the sensor pad for continuous, unobtrusive blood pressure monitoring.

System for measuring and monitoring blood pressure

Methods and devices for long term, cuffless, and continuous arterial blood pressure estimation using an end-to-end deep learning approach are provided. A deep learning system comprises three modules: a deep convolutional neural network (CNN) module for learning powerful features; a one- or multi-layer recurrent neural network (RNN) module for modeling the temporal dependencies in blood pressure dynamics; and a mixture density network (MDN) module for predicting final blood pressure value. This system takes raw physiological signals, such as photoplethysmogram and/or electrocardiography signals, as inputs and yields arterial blood pressure readings in real time.

Deep learning approach for long term, cuffless, and continuous arterial blood pressure estimation

Accurate and effective methods for measuring cardiovascular and respiratory parameters are provided. The method for deriving a depth-specific photoplethysmography (PPG) signal from multi-wavelength PPG signals includes choosing light wavelength combinations, calibrating a multi-layer light-tissue interaction model referring to a physiological signal, and generating the depth-specific PPG signal from the multi-wavelength PPG signals based on the calibrated light-tissue interaction model. The disclosed method for cuff-less blood pressure measurement includes recording a physiological signal and multi-wavelength PPG signals of a predetermined body part, deriving the depth-specific PPG signal reflecting the arterial blood volume with the physiological signal as a reference, calculating the pulse transit time (PTT) from the physiological signal and the derived arterial blood PPG signal, and calculating the blood pressure from the calibrated PTT and blood pressure relationship.

Method for measuring cardiovascular and respiratory parameters based on multi-wavelength photoplethysmography

Growing community of inventors

Hong Kong, China

Yuanting Zhang