WO2023186050A1 - 体征信号的解析方法、装置以及存储介质 - Google Patents

体征信号的解析方法、装置以及存储介质 Download PDF

Info

Publication number
WO2023186050A1
WO2023186050A1 PCT/CN2023/085265 CN2023085265W WO2023186050A1 WO 2023186050 A1 WO2023186050 A1 WO 2023186050A1 CN 2023085265 W CN2023085265 W CN 2023085265W WO 2023186050 A1 WO2023186050 A1 WO 2023186050A1
Authority
WO
WIPO (PCT)
Prior art keywords
optical power
signal
power signal
optical
target
Prior art date
Application number
PCT/CN2023/085265
Other languages
English (en)
French (fr)
Inventor
王划
王涛
崔修涛
陶钧
刘露
李龙
Original Assignee
毕威泰克(浙江)医疗器械有限公司
毕威泰克(上海)医疗器材有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 毕威泰克(浙江)医疗器械有限公司, 毕威泰克(上海)医疗器材有限公司 filed Critical 毕威泰克(浙江)医疗器械有限公司
Publication of WO2023186050A1 publication Critical patent/WO2023186050A1/zh

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
    • A61B5/02416—Detecting, measuring or recording pulse rate or heart rate using photoplethysmograph signals, e.g. generated by infrared radiation
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
    • A61B5/0816—Measuring devices for examining respiratory frequency
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6887—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient mounted on external non-worn devices, e.g. non-medical devices
    • A61B5/6892—Mats
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7253—Details of waveform analysis characterised by using transforms
    • A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms

Definitions

  • the present application relates to the field of vital sign detection, specifically, to a method, device and storage medium for analyzing vital sign signals.
  • the detection method of physical signs uses Fourier transform to calculate the QRS wave of the electrocardiogram.
  • this method is easily interfered by noise signals, and the accuracy of this method based on prior knowledge is difficult to exceed that of neural network big data training.
  • the accuracy of the model That is, the physical sign signal detection equipment in the related art has technical problems such as cumbersome steps to use, high cost, susceptibility to electromagnetic interference, poor accuracy of detection results, and impact on user experience.
  • Embodiments of the present application provide a method, device, and storage medium for analyzing body sign signals to at least solve the problem that when body sign detection equipment in related technologies detects body sign signals, they are susceptible to electromagnetic interference, have poor detection result accuracy, and affect user use. Experience technical issues.
  • a method for analyzing physical signs includes: obtaining the optical power signal of the optical fiber corresponding to the collected target object; inputting the optical power signal into the neural network model to obtain the target of the target object Sign signal, the neural network model is obtained by training through multiple sets of training data.
  • Each set of data in the multiple sets of data includes: the optical power signal of the sample object, and the physical sign signal corresponding to the optical power signal used to identify the sample object; in the display The interface displays target physical signals.
  • acquiring the optical power signal of the optical fiber of the target object includes: receiving the optical power signal transmitted by a sensory acquisition device in contact with the limb of the target object.
  • obtaining the optical power signal of the optical fiber of the target object includes: receiving the optical power signal transmitted by a non-sensory collection device that does not have physical contact with the monitored target object.
  • inputting the optical power signal into the neural network model includes: preprocessing the optical power signal to obtain a processed optical power signal, where the preprocessing method includes: filtering and noise reduction.
  • the above method further includes: performing Fourier transform on the processed optical power signal; and obtaining the corresponding optical power signal according to the Fourier transform.
  • the target frequency domain signal; the target frequency domain signal and the unprocessed optical power signal are simultaneously input to the neural network model.
  • the neural network model is integrated into the embedded device, and the target physical signal is displayed on the display interface, including: displaying the target physical signal on the display interface corresponding to the embedded device; or sending the target physical signal to the device held by other objects. Terminal, where other objects are objects that are associated with the target object.
  • the physical sign signal includes: respiratory frequency and heart rate.
  • the neural network model uses a shallow convolution layer to extract features of the optical power signal of the sample object, and shares convolution parameters, and is divided into Two-way network structure, where the two-way network structure is used to predict respiratory frequency and heart rate respectively.
  • a physical sign signal monitoring device including: a mattress, wherein the mattress includes: a first cushion layer and a second cushion layer; A fiber optic wire structure is laid between them; a light generator is linked to the first end of the fiber optic wire structure, where the light generator is used to emit a first optical signal and input the first optical signal to the first end of the fiber optic wire structure; light The receiver is connected to the second end of the optical fiber wire structure, and the optical receiver is used to receive the second optical signal output from the second end of the optical fiber wire structure; the processing module is connected to the optical receiver, wherein the processing module integrates a neural network model, Used to convert the second optical signal into an optical power signal, and input the optical power signal into the neural network model to obtain the target physical sign signal of the target object.
  • the neural network model is trained through multiple sets of training data, and each of the multiple sets of data
  • the set of data includes: the optical power signal of the sample object, and the physical sign signal corresponding to the optical power signal used to identify the
  • a device for analyzing physical signs including: an acquisition module, configured to acquire the optical power signal of the optical fiber corresponding to the collected target object; and a determination module, configured to The optical power signal is input to the neural network model to obtain the target sign signal of the target object.
  • the neural network model is trained through multiple sets of training data. Each set of data in the multiple sets of data includes: the optical power signal of the sample object, and the identification information.
  • the physical sign signal corresponding to the optical power signal of the sample object; the display module is set to display the target physical signal on the display interface.
  • a non-volatile storage medium includes a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute any arbitrary A method for analyzing physical signs.
  • the non-volatile storage medium is also used to store characteristic data for a predetermined period of time to avoid data loss in the event of power outage or network disconnection.
  • a processor is also provided, and the processor is configured to run a program, wherein when the program is running, any method of analyzing physical signs is executed.
  • Figure 1 is a schematic flowchart of an optional body sign signal analysis method according to an embodiment of the present application
  • Figure 2 is an optional time-frequency diagram after denoising according to an embodiment of the present application
  • Figure 3 is a schematic structural diagram of an optional neural network model according to an embodiment of the present application.
  • Figure 4 is a schematic diagram of the appearance of an optional physical sign signal detection device according to an embodiment of the present application.
  • Figure 5 is a schematic diagram of the placement of an optional physical sign signal detection device according to an embodiment of the present application.
  • Figure 6 is a schematic structural diagram of an optional body sign signal analysis device according to an embodiment of the present application.
  • an embodiment of a method for analyzing physical signs is provided. It should be noted that the steps shown in the flow chart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and ,Although a logical sequence is shown in the flowcharts, in some cases, the steps shown or described may be performed in a sequence different from that herein.
  • Figure 1 is a method for analyzing physical signs according to an embodiment of the present application. As shown in Figure 1, the method includes the following steps:
  • Step S102 obtain the optical power signal of the optical fiber corresponding to the collected target object
  • Step S104 input the optical power signal into the neural network model to obtain the target physical sign signal of the target object.
  • the neural network model is obtained by training through multiple sets of training data. Each set of data in the multiple sets of data includes: the optical power signal of the sample object, and the physical sign signal corresponding to the optical power signal used to identify the sample object;
  • Step S106 Display the target physical sign signal on the display interface.
  • the target sign signal of the target object is obtained by acquiring the optical power signal of the optical fiber corresponding to the target object; inputting the optical power signal into the neural network model, the neural network
  • the model is trained through multiple sets of training data.
  • Each set of data in the multiple sets of data includes: the optical power signal of the sample object, and the physical sign signal corresponding to the optical power signal used to identify the sample object; in the display interface Displaying the target sign signal achieves the purpose of determining the vital sign signal based on the optical fiber signal, thereby achieving technical effects such as avoiding electromagnetic interference, improving the accuracy of the sign signal detection results, improving the user experience, and thus solving the problems caused by related technologies.
  • the physical sign detection equipment is susceptible to electromagnetic interference, poor accuracy of detection results, and technical problems that affect the user experience.
  • sample data in multiple sets of training data can be obtained as follows. Specifically, ordinary volunteers can be recruited within the company, and the equipment can be placed in cooperative hospitals, nursing homes and other institutions to collect original data. , the age of the collection population is mainly distributed between 18 and 80 years old. There are healthy people and people with different diseases. When a person lies on the mat, data is automatically collected and transmitted to the cloud server via wifi for storage.
  • obtaining the optical power signal of the optical fiber of the target object can be achieved in the following manner. Specifically, the optical power signal transmitted by the sensory acquisition device that contacts the limb of the target object can be received. For example, electrode pads.
  • obtaining the optical power signal of the optical fiber of the target object may also be to receive the optical power signal transmitted by a non-sensory collection device that does not have physical contact with the monitored target object.
  • a non-sensory collection device that does not have physical contact with the monitored target object.
  • non-sensory monitoring equipment for example, non-sensory monitoring equipment.
  • sample data for the measured population (sample subjects), some of them can use monitors with Class II medical device certificates to collect user data in a sensory way by attaching electrode pads to the human body. Breathing and heart rate data.
  • non-sensory monitoring devices with Class II medical device certificates can be equipped to collect heart rate and respiratory data. These data can be sent to a cloud server for storage through wireless transmission.
  • inputting the optical power signal into the neural network model includes preprocessing the optical power signal to obtain a processed optical power signal, where the preprocessing method includes filtering and noise reduction.
  • the processed optical power signal can be Fourier transformed; the target frequency corresponding to the optical power signal can be obtained according to the Fourier transform. domain signal, and then input the target frequency domain signal and unprocessed optical power signal to the neural network model at the same time. That is, the collected optical fiber signal data is preprocessed by filtering, noise reduction, etc., and then short-time Fourier transform is performed to convert it into a time-frequency signal.
  • Figure 2 is a time-frequency schematic diagram after denoising, as shown in Figure 2 It shows that the noise in the timing signal can be further reduced through the above denoising process.
  • the above-mentioned neural network model can be integrated in an embedded device. Therefore, displaying the target physical signal on the display interface can display the target physical signal on the corresponding display interface of the embedded device; it is understandable that it can also be displayed on the display interface of the embedded device.
  • the neural network model can be integrated into the embedded device in the following ways, mainly including: model pruning, that is, removing components that have a small impact on the results; model quantization, such as reducing float32 to int8 ;Knowledge distillation, distilling the teacher's ability into the students. Generally, the students will be smaller than the teachers. We distill a large and deep network into a small network; parameter sharing achieves the purpose of reducing network parameters by sharing parameters.
  • model quantification&Pruning can be used to reduce the memory consumption of the model without sacrificing the accuracy of the model as much as possible (even improving the accuracy in some scenarios), and reduce overfitting to a certain extent.
  • model pruning can be roughly divided into the following four steps:
  • quantization is also to reduce the space occupied by the weights of the neural network.
  • the former considers reducing redundant connections in CNN, while quantization attempts to solve problems from the storage form of CNN parameters (such as using fewer bits to record weights, sharing/clustering weights, etc. ).
  • model quantization is equivalent to us wanting to map the original float32 to a new set in some way.
  • the above-mentioned physical signs include but are not limited to: respiratory frequency and heart rate.
  • a shallow convolution layer can be used to extract features of the optical power signal of the sample object. And share the convolution parameters, and then branch into two deep networks (two-way network structure). It should be noted that the two branches are used to predict breathing frequency and heart rate respectively.
  • FIG 3 is a schematic structural diagram of an optional neural network model in this application.
  • CBL is called Conv+BN+LeakyRelu.
  • C in CBM represents the convolution layer
  • B represents the BatchNorm layer
  • M represents Maxpooling layer
  • FC stands for Fully Connected Layer.
  • the neural network model shares shallow features, and then branches into two sets of network structures to extract respiration and heart rate features, and regress to obtain respiration and heart rate values.
  • the MSE loss function can be used in model training and early stop can be used to avoid model overfitting.
  • a physical signal monitoring device including: a mattress, wherein the mattress includes: a first cushion layer and a second cushion layer, and an optical fiber is laid between the first cushion layer and the second cushion layer.
  • a wire structure a light generator connected to a first end of the optical fiber wire structure, wherein the light generator is used to emit a first optical signal and input the first optical signal to the first end of the optical fiber wire structure; an optical receiver and an optical fiber The second end of the wire structure is connected, and the optical receiver is used to receive the second optical signal output by the second end of the optical fiber wire structure; the processing module is connected to the optical receiver, wherein the processing module is integrated with a neural network model for converting the second optical signal The two-light signal is converted into an optical power signal, and the optical power signal is input into the neural network model to obtain the target physical sign signal of the target object.
  • the neural network model is trained through multiple sets of training data. Each set of data in the multiple sets of data includes: The optical power signal of the sample object, and the physical
  • the purpose of determining vital sign signals based on optical fiber signals is achieved, thereby avoiding electromagnetic interference, improving the accuracy of physical sign signal detection results, improving user experience and other technical effects, thereby solving the problems caused by related technologies.
  • physical sign detection equipment is susceptible to electromagnetic interference, poor accuracy of detection results, and technical problems that affect user experience.
  • Figure 4 is a schematic diagram of an optional physical sign signal monitoring device of the present application.
  • the device can be directly laid under the bed sheets. Under the bedding and mattress, and can be placed under the chest;
  • Figure 5 is a schematic diagram of the placement of the device. As can be seen from Figure 5, the product can be placed on/under the mattress and placed at the user's chest.
  • Figure 6 is a physical sign signal analysis device according to an embodiment of the present application. As shown in Figure 6, the physical sign signal analysis device includes:
  • the acquisition module 40 is configured to acquire the optical power signal of the optical fiber corresponding to the collected target object
  • the determination module 42 is configured to input the optical power signal into the neural network model to obtain the target physical sign signal of the target object.
  • the neural network model is obtained through training of multiple sets of training data. Each set of data in the multiple sets of data includes: the light of the sample object. The power signal, and the physical sign signal corresponding to the optical power signal used to identify the sample object;
  • the display module 44 is configured to display the target physical signal on the display interface.
  • the acquisition module 40 is configured to obtain the optical power signal of the optical fiber corresponding to the collected target object; the determination module 42 is configured to input the optical power signal into the neural network model to obtain the target physical signs of the target object.
  • the neural network model is obtained by training through multiple sets of training data.
  • Each set of data in the multiple sets of data includes: the optical power signal of the sample object, and the physical sign signal corresponding to the optical power signal used to identify the sample object; display module 44 , is set to display the target physical sign signal on the display interface, achieving the purpose of determining the vital sign signal based on the optical fiber signal, thus achieving technical effects such as avoiding electromagnetic interference, improving the accuracy of the physical sign signal detection results, improving the user experience, and thus solving the problem
  • the physical sign detection equipment in the related art is susceptible to electromagnetic interference when detecting physical sign signals, the accuracy of the detection results is poor, and there are technical problems that affect the user experience.
  • a non-volatile storage medium includes a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute any arbitrary A method of analyzing physical signs.
  • the non-volatile storage medium is also used to store characteristic data for a predetermined period of time to avoid data loss in the event of power outage or network disconnection.
  • a processor is also provided, and the processor is configured to run a program, wherein when the program is running, any method of analyzing physical signs is executed.
  • the above storage medium is used to store program instructions to perform the following functions and realize the following functions:
  • optical power signal of the optical fiber corresponding to the collected target object input the optical power signal to the neural network model to obtain the target physical sign signal of the target object, and the neural network model is obtained by training through multiple sets of training data,
  • Each set of data in the multiple sets of data includes: an optical power signal of the sample object, and a physical sign signal corresponding to the optical power signal used to identify the sample object; the target physical signal is displayed on the display interface.
  • the above-mentioned processor is used to call program instructions in the memory to implement the following functions:
  • optical power signal of the optical fiber corresponding to the collected target object input the optical power signal to the neural network model to obtain the target physical sign signal of the target object, and the neural network model is obtained by training through multiple sets of training data,
  • Each set of data in the multiple sets of data includes: an optical power signal of the sample object, and a physical sign signal corresponding to the optical power signal used to identify the sample object; the target physical signal is displayed on the display interface.
  • a neural network model is used to analyze the optical power signal to determine the way of determining the physical sign signal. Specifically, by obtaining the optical power signal of the optical fiber corresponding to the collected target object; converting the optical power signal of the optical fiber Input to the neural network model to obtain the target physical signal of the target object.
  • the neural network model is trained through multiple sets of training data. Each set of data in the multiple sets of data includes: the optical power signal of the sample object, and the optical power signal used to identify the sample object.
  • the physical sign signal corresponding to the optical power signal; displaying the target physical sign signal on the display interface achieves the purpose of determining the vital sign signal based on the optical fiber signal, thereby avoiding electromagnetic interference, improving the accuracy of the physical sign signal detection results, and improving the user experience, etc.
  • the technical effect further solves the technical problems that the physical sign detection equipment in related technologies is susceptible to electromagnetic interference when detecting physical sign signals, resulting in poor accuracy of detection results and affecting the user experience.
  • the disclosed technical content can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units may be a logical functional division. In actual implementation, there may be other division methods.
  • multiple units or components may be combined or may be Integrated into another system, or some features can be ignored, or not implemented.
  • the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the units or modules may be in electrical or other forms.
  • the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application can be integrated into one processing unit, each unit can exist physically alone, or two or more units can be integrated into one unit.
  • the above integrated units can be implemented in the form of hardware or software functional units.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium.
  • the technical solution of the present application is essentially or contributes to the existing technology, or all or part of the technical solution can be used as a software product.
  • the computer software product is stored in a storage medium and includes a number of instructions to cause a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the methods described in various embodiments of this application. step.
  • the aforementioned storage media include: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program code. .
  • the technical solutions provided by the embodiments of this application are suitable for the field of vital sign detection.
  • the optical power signal of the optical fiber corresponding to the target object can be collected; the optical power signal can be input to the nerve
  • the network model obtains the target sign signal of the target object and displays the target sign signal on the display interface, achieving the purpose of determining the vital sign signal based on the optical fiber signal, thereby avoiding electromagnetic interference and improving the accuracy of the sign signal detection results. It can improve the user experience and other technical effects, thereby solving the technical problems that the physical sign detection equipment in related technologies is susceptible to electromagnetic interference when detecting physical signs, resulting in poor accuracy of detection results, and affecting the user experience.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Physiology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Cardiology (AREA)
  • Signal Processing (AREA)
  • Pulmonology (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Psychiatry (AREA)
  • Mathematical Physics (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

本申请公开了一种体征信号的解析方法、装置以及存储介质。其中，该方法包括：获取采集到的目标对象对应的光纤的光功率信号；将所述光功率信号输入至神经网络模型，得到所述目标对象的目标体征信号，所述神经网络模型为通过多组训练数据训练得到，所述多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识所述样本对象的光功率信号所对应的体征信号；在显示界面展示所述目标体征信号。

Description

体征信号的解析方法、装置以及存储介质
交叉援引：
本申请要求于2022年3月30日提交中国专利局、优先权号为202210326024.0、申请名称为“体征信号的解析方法、装置以及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及生命体征检测领域，具体而言，涉及一种体征信号的解析方法、装置以及存储介质。
背景技术
随着互联网设备和数字医疗产业的到来，健康和健康监测已成为一个日益关注的领域。监控生命体征诸如心率、心冲击信号以及呼吸率，在医疗保健机构内外都是需要的。在医疗机构内，生命体征跟踪是必不可少的，生命体征跟踪对于确保患者安全、诊断疾病、监测病人的进展、规划病人的护理有着重要作用。在医疗机构外，跟踪生命体征及姿态使个人能将他们的健康状况量化和概念化，帮助个人保持关注他们的健康和健康需要、明确进度、保持实现健康和健身目标是一种趋势。
当前消费品市场上的生命体征追踪器大多是很不准确的，其他更精确的设备，结构复杂，需要连接到插座和用导线连接到患者，这些设备使用步骤繁琐，会引起病人的焦虑，而且设备昂贵，不易携带，容易产生电磁干扰(EMI)，不易部署。另外体征信号的检测方法采用傅里叶变换计算心电图的QRS波，然而这种方法极容易被噪声信号干扰，且这种基于先验知识的方法，准确率很难超过神经网络大数据训练出来的模型的准确率。即相关技术中的体征信号检测设备存在使用步骤繁琐，成本高，易受电磁干扰，检测结果准确性较差，影响用户体验的技术问题。
针对上述的问题，目前尚未提出有效的解决方案。
发明内容
本申请实施例提供了一种体征信号的解析方法、装置以及存储介质，以至少解决由于相关技术中的体征检测设备在检测体征信号时，易受电磁干扰，检测结果准确性差，以及影响用户使用体验的技术问题。
根据本申请实施例的一个方面，提供了一种体征信号的解析方法，包括：获取采集到的目标对象对应的光纤的光功率信号；将光功率信号输入至神经网络模型，得到目标对象的目标体征信号，神经网络模型为通过多组训练数据训练得到，多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识样本对象的光功率信号所对应的体征信号；在显示界面展示目标体征信号。
可选地，获取目标对象的光纤的光功率信号，包括：接收接触目标对象肢体的有感采集设备传输的光功率信号。
可选地，获取目标对象的光纤的光功率信号，包括：接收与监测目标对象不发生肢体接触的无感采集设备传输的光功率信号。
可选地，将光功率信号输入至神经网络模型，包括：对光功率信号进行预处理，得到处理后的光功率信号，其中，预处理的方式包括：滤波以及降噪。
可选地，在对光功率信号进行预处理，得到处理后的光功率信号之后，上述方法还包括：对处理后的光功率信号进行傅里叶变换；根据傅里叶变换得到光功率信号对应的目标频域信号；将目标频域信号以及未处理的光功率信号同时输入至神经网络模型。
可选地，神经网络模型集成在嵌入式设备中，在显示界面展示目标体征信号，包括：在嵌入式设备对应的显示界面展示目标体征信号；或者将目标体征信号发送至其他对象所持有的终端，其中，其他对象为与目标对象具有关联关系的对象。
可选地，其中，体征信号包括：呼吸频率与心率，神经网络模型在训练的过程中，共用浅层卷积层对样本对象的光功率信号进行特征提取，且共用卷积参数，并分为两路网络结构，其中，两路网络结构分别用于预测呼吸频率与心率。
根据本申请实施例的另一方面，还提供一种体征信号监测设备，包括：床垫，其中，床垫包括：第一垫层与第二垫层，第一垫层与第二垫层之间铺设有光纤导线结构；光发生器，与光纤导线结构的第一端链接，其中，光发生器用于发射第一光信号，并将第一光信号输入至光纤导线结构的第一端；光接收器与光纤导线结构的第二端连接，光接收器用于接收光纤导线结构的第二端输出的第二光信号；处理模块，与光接收器连接，其中，处理模块集成有神经网络模型，用于将第二光信号转换为光功率信号，并将光功率信号输入至神经网络模型，得到目标对象的目标体征信号，神经网络模型为通过多组训练数据训练得到，多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识样本对象的光功率信号所对应的体征信号。
根据本申请实施例的另一方面，还提供了一种体征信号的解析装置，包括：获取模块，设置为获取采集到的目标对象对应的光纤的光功率信号；确定模块，设置为将 光功率信号输入至神经网络模型，得到目标对象的目标体征信号，神经网络模型为通过多组训练数据训练得到，多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识样本对象的光功率信号所对应的体征信号；展示模块，设置为在显示界面展示目标体征信号。
根据本申请实施例的另一方面，还提供了一种非易失性存储介质，非易失性存储介质包括存储的程序，其中，在程序运行时控制非易失性存储介质所在设备执行任意一种体征信号的解析方法，所述非易失性存储介质还用于存储预定时段的特征数据，以免在掉电或断网的情况下造成数据丢失。
根据本申请实施例的另一方面，还提供了一种处理器，处理器用于运行程序，其中，程序运行时执行任意一种体征信号的解析方法。
附图说明
此处所说明的附图用来提供对本申请的进一步理解，构成本申请的一部分，本申请的示意性实施例及其说明用于解释本申请，并不构成对本申请的不当限定。在附图中：
图1是根据本申请实施例的一种可选的体征信号的解析方法的流程示意图；
图2是根据本申请实施例中一种可选的去噪后的时频示意图；
图3是根据本申请实施例中一种可选的神经网络模型的结构示意图；
图4是根据本申请实施例中一种可选的体征信号检测设备的外观示意图；
图5是根据本申请实施例中一种可选的体征信号检测设备的放置方式示意图；
图6是根据本申请实施例的一种可选的体征信号的解析装置的结构示意图。
具体实施方式
为了使本技术领域的人员更好地理解本申请方案，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分的实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本申请保护的范围。
需要说明的是，本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这 样使用的数据在适当情况下可以互换，以便这里描述的本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
根据本申请实施例，提供了一种体征信号的解析方法的实施例，需要说明的是，在附图的流程图示出的步骤可以在诸如一组计算机可执行指令的计算机系统中执行，并且，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
图1是根据本申请实施例的体征信号的解析方法，如图1所示，该方法包括如下步骤：
步骤S102，获取采集到的目标对象对应的光纤的光功率信号；
步骤S104，将光功率信号输入至神经网络模型，得到目标对象的目标体征信号，神经网络模型为通过多组训练数据训练得到，多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识样本对象的光功率信号所对应的体征信号；
步骤S106，在显示界面展示目标体征信号。
在该体征信号的解析方法中，通过获取采集到的目标对象对应的光纤的光功率信号；将所述光功率信号输入至神经网络模型，得到所述目标对象的目标体征信号，所述神经网络模型为通过多组训练数据训练得到，所述多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识所述样本对象的光功率信号所对应的体征信号；在显示界面展示所述目标体征信号，达到了基于光纤信号确定生命体征信号的目的，从而实现了避免电磁干扰，提高体征信号检测结果的准确性，提升用户使用体验等技术效果，进而解决了由于相关技术中的体征检测设备在检测体征信号时，易受电磁干扰，检测结果准确性差，以及影响用户使用体验的技术问题。
本申请一些可选的实施中，多组训练数据中的样本数据可以通过如下获取，具体地，可以为企业内部召集普通志愿者，同时将设备放置到合作医院、养老院等机构进行原始数据的采集，采集人群年龄主要分布在18～80岁之间，有健康人群，同时有具有不同病种的人群。当人躺在垫子上时，会自动收集数据，并将数据通过wifi传输到云端服务器进行存储。
本申请一些实施例中，获取目标对象的光纤的光功率信号，可以通过如下方式实现，具体地，可接收接触目标对象肢体的有感采集设备传输的光功率信号。例如，电极片。
本申请另一些可选的实施例中，获取目标对象的光纤的光功率信号，还可以为接收与监测目标对象不发生肢体接触的无感采集设备传输的光功率信号。例如，无感监测设备。
可以理解的，在样本数据的获取中，对于被测人群(样本对象)，其中部分对象可采用了具有II类医疗器械证书的监护仪，通过在人体上贴电极片，以有感方式采集用户的呼吸，心率数据。对于其他部分人群，可配备具有II类医疗器械证书的无感监测设备采集心率和呼吸数据，这些数据可以无线传输的方式发送到云端服务器进行存储。
本申请一些实施例中，将光功率信号输入至神经网络模型，包括：对光功率信号进行预处理，得到处理后的光功率信号，其中，预处理的方式包括：滤波以及降噪。需要说明的是，在对光功率信号进行预处理，得到处理后的光功率信号之后，可对处理后的光功率信号进行傅里叶变换；根据傅里叶变换得到光功率信号对应的目标频域信号，然后再将目标频域信号以及未处理的光功率信号同时输入至神经网络模型。即对于采集到的光纤信号数据进行滤波、降噪等预处理，然后再进行短时傅里叶变换，将其转换成时频信号，图2是去噪后的时频示意图，如图2所示，通过上述去噪过程可以进一步减少时序信号中的噪声。
本申请一些实施例中，上述神经网络模型可集成在嵌入式设备中，因此，在显示界面展示目标体征信号，可以为在嵌入式设备对应的显示界面展示目标体征信号；可以理解的，也可将目标体征信号发送至其他对象所持有的终端，其中，其他对象为与目标对象具有关联关系的对象。例如，目标对象为张三，则其他对象可以为张三的医生、护士、父亲、母亲、其他监护人以及护理人等。
在一示例性实施例中，可通过如下方式将神经网络模型集成在嵌入式设备中，主要包括：模型剪枝，即移除对结果作用较小的组件；模型量化，比如将float32降到int8；知识蒸馏，将teacher的能力蒸馏到student上，一般student会比teacher小。我们把一个大而深的网络蒸馏到一个小的网络；参数共享，通过共享参数，达到减少网络参数的目的。
容易注意到的是，神经网络模型内部大量有待训练的权重(weights)往往占用了大量的内存空间以及存储带宽(如AlexNet～200MB，VGG-16～500MB)，给CNN在一些嵌入式平台、移动端上的部署带来了很大的困难。此外，庞大的计算量也能耗巨大，计算成本比较昂贵。为此，可模型量化剪枝(Quantization&Pruning)的方式，在尽可能不牺牲模型精度(甚至在一些场景能提升精度)的前提下，减小模型的内存消耗，以及一定程度上减小过拟合，模型剪枝大致可以分成以下4步：
1.首先进行正常的模型训练，训至基本收敛；
2.简单粗暴地移除权重低于某个阈值的连接；
3.对剪枝后的网络再重新训练从而精度恢复(类似fine-tuning)直至收敛；
4.再将剪枝后的稀疏网络用特定格式存储(CSR或CSC)。
可以理解的，量化的目的也是为了减小神经网络的权重(weights)所占空间大小。不过与剪枝不同的是，前者考虑减少CNN中的冗余连接，而量化试图从CNN参数的存储形式上解决问题(如使用更少的比特数来记录权重，对权重进行共享/聚类等)。从直观上理解，模型量化相当于我们想把原先的float32通过某种方式映射到一个新的集合上。
本申请一些可选的实施例中，上述体征信号包括但不限于：呼吸频率与心率，神经网络模型在训练的过程中，可共用浅层卷积层对样本对象的光功率信号进行特征提取，且共用卷积参数，然后分支为两个深层网络(两路网络结构)，需要说明的是，两路分支分别用于对呼吸频率与心率进行预测。
图3是本申请一种可选的神经网络模型的结构示意图，如图3所示，其中，CBL全称为Conv+BN+LeakyRelu，CBM中的C代表卷积层，B代表BatchNorm层，M代表Maxpooling层，FC表示Fully Connected Layer全连接层。从图3可以看出，该神经网络模型通过共用浅层特征，然后分支为两组网络结构进行呼吸和心率特征的提取，回归得到呼吸和心率值。需要说明的是，模型训练中可采用MSE损失函数，使用early stop，避免模型过拟合。
本申请一些实例中，还提供一种体征信号监测设备，包括：床垫，其中，床垫包括：第一垫层与第二垫层，第一垫层与第二垫层之间铺设有光纤导线结构；光发生器，与光纤导线结构的第一端链接，其中，光发生器用于发射第一光信号，并将第一光信号输入至光纤导线结构的第一端；光接收器与光纤导线结构的第二端连接，光接收器用于接收光纤导线结构的第二端输出的第二光信号；处理模块，与光接收器连接，其中，处理模块集成有神经网络模型，用于将第二光信号转换为光功率信号，并将光功率信号输入至神经网络模型，得到目标对象的目标体征信号，神经网络模型为通过多组训练数据训练得到，多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识样本对象的光功率信号所对应的体征信号。
通过上述体征信号监测设备达到了基于光纤信号确定生命体征信号的目的，从而实现了避免电磁干扰，提高体征信号检测结果的准确性，提升用户使用体验等技术效果，进而解决了由于相关技术中的体征检测设备在检测体征信号时，易受电磁干扰，检测结果准确性差，以及影响用户使用体验的技术问题。
图4是本申请一种可选的体征信号监测设备的示意图，该设备可直接铺在床单下、 被褥下以及床垫下，并可放置在胸腔下面；图5为该设备的放置示意图，从图5可看出，该产品可放置在床垫上/下，并置于用户的胸腔位置处。
图6根据本申请实施例的一种体征信号的解析装置，如图6所示，该体征信号的解析装置包括：
获取模块40，设置为获取采集到的目标对象对应的光纤的光功率信号；
确定模块42，设置为将光功率信号输入至神经网络模型，得到目标对象的目标体征信号，神经网络模型为通过多组训练数据训练得到，多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识样本对象的光功率信号所对应的体征信号；
展示模块44，设置为在显示界面展示目标体征信号。
该体征信号的解析装置中，获取模块40，设置为获取采集到的目标对象对应的光纤的光功率信号；确定模块42，设置为将光功率信号输入至神经网络模型，得到目标对象的目标体征信号，神经网络模型为通过多组训练数据训练得到，多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识样本对象的光功率信号所对应的体征信号；展示模块44，设置为在显示界面展示目标体征信号，达到了基于光纤信号确定生命体征信号的目的，从而实现了避免电磁干扰，提高体征信号检测结果的准确性，提升用户使用体验等技术效果，进而解决了由于相关技术中的体征检测设备在检测体征信号时，易受电磁干扰，检测结果准确性差，以及影响用户使用体验的技术问题。
根据本申请实施例的另一方面，还提供了一种非易失性存储介质，非易失性存储介质包括存储的程序，其中，在程序运行时控制非易失性存储介质所在设备执行任意一种体征信号的解析方法，非易失性存储介质还用于存储预定时段的特征数据，以免在掉电或断网的情况下造成数据丢失。
根据本申请实施例的另一方面，还提供了一种处理器，处理器用于运行程序，其中，程序运行时执行任意一种体征信号的解析方法。
具体地，上述存储介质用于存储执行以下功能的程序指令，实现以下功能：
获取采集到的目标对象对应的光纤的光功率信号；将所述光功率信号输入至神经网络模型，得到所述目标对象的目标体征信号，所述神经网络模型为通过多组训练数据训练得到，所述多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识所述样本对象的光功率信号所对应的体征信号；在显示界面展示所述目标体征信号。
具体地，上述处理器用于调用存储器中的程序指令，实现以下功能：
获取采集到的目标对象对应的光纤的光功率信号；将所述光功率信号输入至神经网络模型，得到所述目标对象的目标体征信号，所述神经网络模型为通过多组训练数据训练得到，所述多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识所述样本对象的光功率信号所对应的体征信号；在显示界面展示所述目标体征信号。
在本申请相关实施例中，通过采用神经网络模型对光功率信号进行分析，确定体征信号的方式，具体地，通过获取采集到的目标对象对应的光纤的光功率信号；将光纤的光功率信号输入至神经网络模型，得到目标对象的目标体征信号，神经网络模型为通过多组训练数据训练得到，多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识样本对象的光功率信号所对应的体征信号；在显示界面展示目标体征信号，达到了基于光纤信号确定生命体征信号的目的，从而实现了避免电磁干扰，提高体征信号检测结果的准确性，提升用户使用体验等技术效果，进而解决了由于相关技术中的体征检测设备在检测体征信号时，易受电磁干扰，检测结果准确性差，以及影响用户使用体验的技术问题。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。
在本申请的上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
在本申请所提供的几个实施例中，应该理解到，所揭露的技术内容，可通过其它的方式实现。其中，以上所描述的装置实施例仅仅是示意性的，例如所述单元的划分，可以为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，单元或模块的间接耦合或通信连接，可以是电性或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的 形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可为个人计算机、服务器或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述仅是本申请的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本申请原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视为本申请的保护范围。
工业实用性
本申请实施例提供的技术方案适用于生命体征检测领域，采用本申请提供的体征信号的解析方法，可以通过采集到的目标对象对应的光纤的光功率信号；将所述光功率信号输入至神经网络模型，得到所述目标对象的目标体征信号，并在显示界面展示所述目标体征信号，达到了基于光纤信号确定生命体征信号的目的，从而实现了避免电磁干扰，提高体征信号检测结果的准确性，提升用户使用体验等技术效果，进而解决了由于相关技术中的体征检测设备在检测体征信号时，易受电磁干扰，检测结果准确性差，以及影响用户使用体验的技术问题。

Claims (11)

  1. 一种体征信号的解析方法，包括：
    获取采集到的目标对象对应的光纤的光功率信号；
    将所述光功率信号输入至神经网络模型，得到所述目标对象的目标体征信号，所述神经网络模型为通过多组训练数据训练得到，所述多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识所述样本对象的光功率信号所对应的体征信号；
    在显示界面展示所述目标体征信号。
  2. 根据权利要求1所述的方法，其中，获取目标对象的光纤的光功率信号，包括：
    接收接触所述目标对象肢体的有感采集设备传输的所述光功率信号。
  3. 根据权利要求1所述的方法，其中，获取目标对象的光纤的光功率信号，包括：
    接收与监测所述目标对象不发生肢体接触的无感采集设备传输的所述光功率信号。
  4. 根据权利要求1所述的方法，其中，将所述光功率信号输入至神经网络模型，包括：
    对所述光功率信号进行预处理，得到处理后的光功率信号，其中，所述预处理的方式包括：滤波以及降噪。
  5. 根据权利要求4所述的方法，其中，在对所述光功率信号进行预处理，得到处理后的光功率信号之后，所述方法还包括：
    对所述处理后的光功率信号进行傅里叶变换；
    根据所述傅里叶变换得到所述光功率信号对应的目标频域信号；
    将所述目标频域信号及未处理的光功率信号同时输入至所述神经网络模型。
  6. 根据权利要求1所述的方法，其中，所述神经网络模型集成在嵌入式设备中，在显示界面展示所述目标体征信号，包括：
    在所述嵌入式设备对应的显示界面展示所述目标体征信号；或者
    将所述目标体征信号发送至其他对象所持有的终端，其中，所述其他对象为与所述目标对象具有关联关系的对象。
  7. 根据权利要求1至6中任意一项所述的方法，其中，所述体征信号包括：呼吸频 率与心率，所述神经网络模型在训练的过程中，共用浅层卷积层对所述样本对象的光功率信号进行特征提取，且共用卷积参数，并分为两路网络结构，其中，所述两路网络结构分别用于预测所述呼吸频率与心率。
  8. 一种体征信号监测设备，包括：
    床垫，其中，所述床垫包括：第一垫层与第二垫层，所述第一垫层与第二垫层之间铺设有光纤导线结构；
    光发生器，与所述光纤导线结构的第一端链接，其中，所述光发生器用于发射第一光信号，并将所述第一光信号输入至所述光纤导线结构的第一端；
    光接收器与所述光纤导线结构的第二端连接，所述光接收器用于接收所述光纤导线结构的第二端输出的第二光信号；
    处理模块，与所述光接收器连接，其中，所述处理模块集成有神经网络模型，用于将所述第二光信号转换为光功率信号，并将所述光功率信号输入至神经网络模型，得到目标对象的目标体征信号，所述神经网络模型为通过多组训练数据训练得到，所述多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识所述样本对象的光功率信号所对应的体征信号。
  9. 一种体征信号的解析装置，包括：
    获取模块，设置为获取采集到的目标对象对应的光纤的光功率信号；
    确定模块，设置为将所述光功率信号输入至神经网络模型，得到所述目标对象的目标体征信号，所述神经网络模型为通过多组训练数据训练得到，所述多组数据中的每组数据包括：样本对象的光功率信号，以及用于标识所述样本对象的光功率信号所对应的体征信号；
    展示模块，设置为在显示界面展示所述目标体征信号。
  10. 一种非易失性存储介质，所述非易失性存储介质包括存储的程序，其中，在所述程序运行时控制所述非易失性存储介质所在设备执行权利要求1至7中任意一项所述体征信号的解析方法，所述非易失性存储介质还用于存储预定时段的体征数据，以免在掉电或断网的情况下造成数据丢失。
  11. 一种处理器，所述处理器用于运行程序，其中，所述程序运行时执行权利要求1至7中任意一项所述体征信号的解析方法。
PCT/CN2023/085265 2022-03-30 2023-03-30 体征信号的解析方法、装置以及存储介质 WO2023186050A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210326024.0 2022-03-30
CN202210326024.0A CN114587301A (zh) 2022-03-30 2022-03-30 体征信号的解析方法、装置以及存储介质

Publications (1)

Publication Number Publication Date
WO2023186050A1 true WO2023186050A1 (zh) 2023-10-05

Family

ID=81812585

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/085265 WO2023186050A1 (zh) 2022-03-30 2023-03-30 体征信号的解析方法、装置以及存储介质

Country Status (2)

Country Link
CN (1) CN114587301A (zh)
WO (1) WO2023186050A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114587301A (zh) * 2022-03-30 2022-06-07 毕威泰克(上海)医疗器材有限公司 体征信号的解析方法、装置以及存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6498652B1 (en) * 2000-02-08 2002-12-24 Deepak Varshneya Fiber optic monitor using interferometry for detecting vital signs of a patient
US20140155774A1 (en) * 2012-11-30 2014-06-05 The Regents Of The University Of California Non-invasively determining respiration rate using pressure sensors
CN108567433A (zh) * 2018-05-16 2018-09-25 苏州安莱光电科技有限公司 一种多功能全光纤非侵入式的状态监测薄垫
US20200107753A1 (en) * 2018-10-08 2020-04-09 UDP Labs, Inc. Systems and Methods for Utilizing Gravity to Determine Subject-Specific Information
US20210059539A1 (en) * 2019-08-27 2021-03-04 Turtle Shell Technologies Private Limited System and a Method for Determining Breathing Rate as a Biofeedback
CN112754443A (zh) * 2021-01-20 2021-05-07 浙江想能睡眠科技股份有限公司 睡眠质量检测方法、系统、可读存储介质及床垫
CN114587301A (zh) * 2022-03-30 2022-06-07 毕威泰克(上海)医疗器材有限公司 体征信号的解析方法、装置以及存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6498652B1 (en) * 2000-02-08 2002-12-24 Deepak Varshneya Fiber optic monitor using interferometry for detecting vital signs of a patient
US20140155774A1 (en) * 2012-11-30 2014-06-05 The Regents Of The University Of California Non-invasively determining respiration rate using pressure sensors
CN108567433A (zh) * 2018-05-16 2018-09-25 苏州安莱光电科技有限公司 一种多功能全光纤非侵入式的状态监测薄垫
US20200107753A1 (en) * 2018-10-08 2020-04-09 UDP Labs, Inc. Systems and Methods for Utilizing Gravity to Determine Subject-Specific Information
US20210059539A1 (en) * 2019-08-27 2021-03-04 Turtle Shell Technologies Private Limited System and a Method for Determining Breathing Rate as a Biofeedback
CN112754443A (zh) * 2021-01-20 2021-05-07 浙江想能睡眠科技股份有限公司 睡眠质量检测方法、系统、可读存储介质及床垫
CN114587301A (zh) * 2022-03-30 2022-06-07 毕威泰克(上海)医疗器材有限公司 体征信号的解析方法、装置以及存储介质

Also Published As

Publication number Publication date
CN114587301A (zh) 2022-06-07

Similar Documents

Publication Publication Date Title
Athavale et al. Biosignal monitoring using wearables: Observations and opportunities
CN102014742B (zh) 现场可部署的震荡检测器
US9183351B2 (en) Mobile system with network-distributed data processing for biomedical applications
CA2787170C (en) Multivariate residual-based health index for human health monitoring
WO2014151874A1 (en) Systems, methods and devices for assessing and treating pain, discomfort and anxiety
WO2023186050A1 (zh) 体征信号的解析方法、装置以及存储介质
WO2016168979A1 (zh) 一种生命体征分析方法与系统
AU2021250913A1 (en) Localized collection of biological signals, cursor control in speech-assistance interface based on biological electrical signals and arousal detection based on biological electrical signals
CN111481207A (zh) 基于心冲击信号的睡眠姿势识别装置及方法
WO2015027544A1 (zh) 排尿感知检测方法及装置
CN201926979U (zh) 一种脑机接口设备
CN109247935B (zh) 一种人体夜间睡眠异常状态监测系统
CN204336935U (zh) 一种基于脉搏、呼吸和皮肤电阻的多道生理心理测试仪
CN201886480U (zh) 一种远程监护系统
CN110680285A (zh) 一种基于神经网络的麻醉程度监测装置
WO2020133339A1 (zh) 一种监护系统、数据采集端、数据接收显示端及监护方法
US20240156389A1 (en) Health monitoring system and method
CN113647973B (zh) 一种基于卷积神经网络的二便无损检测装置
CA3173469A1 (en) Systems and methods for remote patient screening and triage
CN209123173U (zh) 意识障碍促醒系统
Susič et al. Penetration of the ICT technology to the health care primary sector—Ljubljana PILOT
CN2761143Y (zh) 多导心理生理测试装置
CN219763346U (zh) 脑卒中预警与病情实时监测装置
CN206612808U (zh) 一种心肺疾病实时监控系统
Banerjee et al. Clinical evaluation of generative model based monitoring and comparison with compressive sensing

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23778415

Country of ref document: EP

Kind code of ref document: A1