WO2022021664A1 - 一种快速检测新型冠状病毒肺炎的方法及系统 - Google Patents

一种快速检测新型冠状病毒肺炎的方法及系统 Download PDF

Info

Publication number
WO2022021664A1
WO2022021664A1 PCT/CN2020/127489 CN2020127489W WO2022021664A1 WO 2022021664 A1 WO2022021664 A1 WO 2022021664A1 CN 2020127489 W CN2020127489 W CN 2020127489W WO 2022021664 A1 WO2022021664 A1 WO 2022021664A1
Authority
WO
WIPO (PCT)
Prior art keywords
subject
pneumonia
algorithm
novel coronavirus
human
Prior art date
Application number
PCT/CN2020/127489
Other languages
English (en)
French (fr)
Inventor
邓伟侨
杨丽
周威
孙磊
Original Assignee
山东大学
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 山东大学 filed Critical 山东大学
Publication of WO2022021664A1 publication Critical patent/WO2022021664A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning

Definitions

  • the invention belongs to the technical field of biomedicine and data processing, and in particular relates to a method and system for rapidly detecting novel coronavirus pneumonia.
  • coronavirus A new type of respiratory system disease, the main clinical symptoms are fever, fatigue, and dry cough. After isolation and identification, the pathogen was confirmed to be a new type of coronavirus, which was named "SARS-CoV-2" by the International Committee on Taxonomy of Viruses. This is the seventh coronavirus currently found in humans. On February 11, 2020, the global research and innovation forum held by the World Health Organization in Geneva officially named the disease caused by the new coronavirus as COVID-19.
  • the suspected case also has one of the following etiological or serological evidence: real-time fluorescent RT-PCR detection of new coronavirus nucleic acid positive, viral gene sequencing, and has been Knowing that the novel coronavirus is highly homologous, the serum novel coronavirus-specific IgM and IgG antibodies are positive, the serum novel coronavirus-specific antibody IgG antibody has changed from negative to positive, or the recovery period is 4 times or more higher than the acute phase, the diagnosis can be confirmed. patients with novel coronavirus pneumonia.
  • the present invention provides a method and system for rapid detection of novel coronavirus pneumonia.
  • the present invention takes the two types of people with COVID-19 pneumonia and healthy people as classification targets, and compares the expiratory NO concentration with the
  • the basic features of the human body are also used as feature quantities to classify patients with COVID-19 pneumonia and healthy people, and the diagnostic accuracy rate is as high as 90%.
  • the detection cost of the invention is low, the speed is high, and the accuracy rate is high, so it has good practical application value.
  • a first aspect of the present invention provides a method for rapid detection of novel coronavirus pneumonia, the detection method comprising:
  • the probability of the subject's disease is determined.
  • the construction method of the new coronary pneumonia human detection model includes: training the collected human body characteristics based on the algorithm to obtain the new coronary pneumonia human detection model, so as to realize the classification of healthy people and new coronary pneumonia patients.
  • the sources of human feature collection include healthy people, patients with new coronavirus pneumonia, or suspected patients with new coronavirus pneumonia.
  • the subject can be a healthy person, a patient with novel coronavirus pneumonia, or a suspected patient with novel coronavirus pneumonia; preferably, a suspected patient with novel coronavirus pneumonia.
  • the processing object in the present invention is the discontinuous classification label value
  • the preferred algorithm is a machine learning classification algorithm
  • the machine learning classification algorithm includes a random forest method, a support vector machine algorithm, K-proximity algorithm, etc., preferably random forest algorithm.
  • the feature quantity includes basic body sign information and exhaled NO concentration
  • the basic physical sign information of the human body includes but is not limited to height, weight, age, gender, body mass index, body surface area, blood type, whether there is a smoking history, whether there is any other lung or respiratory disease, etc.;
  • the specific method for judging the probability of a subject's disease is as follows: importing the subject's basic physical sign information and expiratory NO concentration data into the above-mentioned new coronary pneumonia human body detection model, and judging the subject's disease probability based on the new coronary pneumonia human body detection model. .
  • a second aspect of the present invention provides a system for detecting novel coronavirus pneumonia, the detection system comprising at least:
  • a data processing module configured to process the medical data of the subject to obtain the characteristic quantity of the subject, wherein the characteristic quantity includes the exhaled NO concentration data of the subject and the basic physical sign information of the human body;
  • a data analysis module used to process the feature quantity through a preset analysis model to determine the test result of the new coronary pneumonia of the subject
  • the preset analysis model is obtained by training the collected human body characteristics based on an algorithm, thereby realizing the classification of healthy people and patients with new coronary pneumonia.
  • NO concentration data can be obtained by exhaled NO detector
  • the basic physical sign information of the human body includes but is not limited to height, weight, age, gender, body mass index, body surface area, blood type, whether there is a smoking history, whether there is any other lung or respiratory disease, etc.;
  • the processing object in the present invention is the discontinuous classification label value
  • the preferred algorithm is a machine learning classification algorithm
  • the machine learning classification algorithm includes a random forest method, a support vector machine algorithm, K-proximity algorithm, etc., preferably random forest algorithm.
  • the sources of human characteristic data collection include healthy people, patients with novel coronavirus pneumonia or suspected patients with novel coronavirus pneumonia.
  • the human body characteristic quantity includes human body exhaled NO concentration data and human body basic physical sign information.
  • a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned method for rapidly detecting novel coronavirus pneumonia is implemented.
  • a fourth aspect of the present invention provides an electronic device, comprising:
  • the objects selected when the model is established include both COVID-19 patients and healthy people, and the objects predicted by the subjects also include healthy people and COVID-19 patients. 19 patients, and the model was constructed based on the random forest method, so that the accuracy of the detection results was greatly improved. After detection, the accuracy reached more than 90%.
  • the above-mentioned technical solution has the advantages of low cost, rapidity, non-destructiveness, high detection accuracy, pain-free subjects, good compliance, and good practical application value.
  • FIG. 1 is a schematic flowchart of the detection of novel coronavirus pneumonia of the present invention.
  • Fig. 2 is the characteristic quantity statistical distribution diagram of COVID-19 patients and healthy people in Example 1 of the present invention, wherein, (a)-(f) are the height and weight of COVID-19 patients and healthy people in Example 1 of the present invention, respectively , BMI, body surface area, age and statistical distribution of FeNO.
  • Example 3 is a machine learning classification model based on exhaled NO concentration, body mass index, body surface area, and age established in Example 1 of the present invention; wherein, (a) is the subject characteristic curve, and (b) is the model The training effect map, (c) is the prediction result of the model in the test set.
  • Example 4 is a machine learning classification model based on exhaled NO concentration, age, height and weight established in Example 2 of the present invention; wherein, (a) is the subject characteristic curve, and (b) is the training effect diagram of the model , (c) is the prediction result of the model in the test set.
  • Fig. 5 is a machine learning classification model based on expiratory NO concentration, height, weight, body mass index, body surface area, and age established in Case 3 of the implementation of the present invention; wherein, (a) is the subject characteristic curve, (b) is the training effect of the model, (c) is the prediction result of the model in the test set.
  • a method for rapid detection of novel coronavirus pneumonia includes:
  • the probability of the subject's disease is determined.
  • the sources of human feature collection include healthy people, patients with new coronavirus pneumonia, or suspected patients with new coronavirus pneumonia.
  • the processing object in the present invention is the discontinuous classification label value
  • the preferred algorithm is a machine learning classification algorithm
  • the machine learning classification algorithm includes a random forest method, a support vector machine algorithm, K-proximity algorithm, etc., preferably random forest algorithm.
  • a random forest is a classifier that contains multiple decision trees, and its output classes are determined by the mode of the classes output by the individual trees.
  • the feature quantity includes basic body sign information and exhaled NO concentration
  • the basic physical sign information of the human body includes but is not limited to height, weight, age, gender, BMI, body surface area, blood type, whether there is a smoking history, whether there is other lung or respiratory diseases, etc.;
  • the construction method of the new coronary pneumonia human detection model includes: training the collected human body characteristics based on the algorithm to obtain the new coronary pneumonia human detection model, so as to realize the classification of healthy people and new coronary pneumonia patients.
  • the construction method of the new coronary pneumonia human detection model 25% of the entire data is extracted as the test set, and the remaining 75% is used as the training set, and the 10-fold cross-validation method is used for training. Thereby selecting the best random forest classification parameters.
  • the classification accuracy rate can reach more than 90%
  • the positive detection rate can reach more than 90%
  • the negative judgment correct rate can also reach more than 90%.
  • the area under the subject characteristic curve is more than 90%.
  • the COVID-19 human detection model of this application will output the probability of the subject being sick when it is used for diagnosis, and determine whether the subject is sick with COVID-19 according to the size of the probability value.
  • the subject can be a healthy person, a novel coronavirus pneumonia patient, or a novel coronavirus pneumonia suspected patient.
  • the specific method for detecting the probability of a subject's disease is as follows: importing the subject's basic physical sign information and expiratory NO concentration data into the above-mentioned new coronary pneumonia human body detection model, and judging the subject's disease probability based on the new coronary pneumonia human body detection model.
  • test subjects are required not to eat or drink for the first 3 hours; environment: test subjects are required to avoid inhaling air with NO>10ppb during the test.
  • the expiratory pressure, time and flow rate need to be monitored during the test.
  • the detection limit of the exhaled NO detector should be lower than 5ppb, and the analytical accuracy/repeatability should be less than or equal to 5 (10) ppb or 10% (20%).
  • the instrument must be calibrated regularly and timely to ensure the accuracy and stability of the instrument.
  • a system for detecting novel coronavirus pneumonia at least includes:
  • a data processing module configured to process the medical data of the subject to obtain the characteristic quantity of the subject, wherein the medical data includes the exhaled NO concentration data of the subject and the basic physical sign information of the human body;
  • a data analysis module used to process the feature quantity through a preset analysis model to determine the test result of the new coronary pneumonia of the subject
  • the preset analysis model is obtained by training the collected human body characteristics based on an algorithm, thereby realizing the classification of healthy people and patients with new coronary pneumonia.
  • NO concentration data may be obtained by an exhaled breath NO detector
  • the basic physical sign information of the human body includes but is not limited to height, weight, age, gender, body mass index, body surface area, blood type, whether there is a history of smoking, whether there is other lung or respiratory diseases, etc.;
  • the processing object in the present invention is the discontinuous classification label value
  • the preferred algorithm is a machine learning classification algorithm
  • the machine learning classification algorithm includes a random forest method, a support vector machine algorithm, K-proximity algorithm, etc., preferably random forest algorithm.
  • the source of human characteristic quantity collection includes healthy people, patients with novel coronavirus pneumonia, or suspected patients with novel coronavirus pneumonia.
  • a computer-readable storage medium is provided on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned method for rapidly detecting novel coronavirus pneumonia is implemented.
  • an electronic device comprising:
  • the detection limit of the exhaled NO detector should be lower than 5ppb, and the analytical accuracy/repeatability should be less than or equal to 5 (10) ppb or 10% (20%).
  • the instrument must be calibrated regularly and timely to ensure the accuracy and stability of the instrument.
  • the basic physical information of the subjects was collected and the expiratory NO concentration was measured.
  • the NO concentration test parameter setting and process are as follows:
  • the NO concentration is measured by the exhaled NO detector at room temperature and the relative humidity range is not more than 80%.
  • the testers recorded the basic physical information such as height, weight, age, gender, and disease history of the non-COVID-19 patients, and also measured the NO concentration in the exhaled breath.
  • Figure 3(b) is a schematic diagram of the model training
  • Figure 3(c) is the prediction result of the model in the test set. According to the simulation results, the positive detection rate of the model can reach 91.7%, and the correct rate of negative judgment is 91.3%.
  • NO concentration test parameter setting and process are as follows:
  • the NO concentration is measured by the exhaled NO detector at room temperature and the relative humidity range is not more than 80%.
  • the testers recorded the basic physical information such as height, weight, age, gender, and disease history of the non-COVID-19 patients, and also measured the NO concentration in the exhaled breath.
  • Figure 4(b) is a schematic diagram of the model training
  • Figure 4(c) is the prediction result of the model in the test set. According to the simulation results, the positive detection rate of the model can reach 83.3%, and the negative judgment correct rate is 87.0%.
  • NO concentration test parameter setting and process are as follows:
  • the NO concentration is measured by the exhaled NO detector at room temperature and the relative humidity range is not more than 80%.
  • the testers recorded the basic physical information such as height, weight, age, gender, and disease history of the non-COVID-19 patients, and also measured the NO concentration in the exhaled breath.
  • Figure 5(b) is a schematic diagram of the model training
  • Figure 5(c) is the prediction result of the model in the test set. According to the simulation results, the positive detection rate of this model can reach 83.3%, and the correct rate of negative judgment is 87.0%.
  • a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, implements the steps in the method for rapid detection of novel coronavirus pneumonia as shown in FIG. 1 .
  • a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the program shown in FIG. 1 when the processor executes the program The steps in the rapid detection method of novel coronavirus pneumonia.
  • embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied therein, including but not limited to disk storage, optical storage, and the like.
  • These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions
  • the apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), or a random access memory (Random Access Memory, RAM) or the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Pathology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Mathematical Physics (AREA)
  • Evolutionary Biology (AREA)
  • Biomedical Technology (AREA)
  • Databases & Information Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Investigating Or Analysing Biological Materials (AREA)
  • Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
  • Apparatus Associated With Microorganisms And Enzymes (AREA)

Abstract

一种快速检测新型冠状病毒肺炎的方法及系统，属于生物医学和数据处理技术领域。该方法和系统通过以COVID-19肺炎患者和健康人这两类人作为分类目标，将呼气NO浓度与人体基本特征同时作为特征量来对COVID-19肺炎患者和健康人进行分类，诊断正确率达90％以上。该方法和系统检测成本低，速度快，准确率高，因此具有良好的实际应用之价值。

Description

一种快速检测新型冠状病毒肺炎的方法及系统 技术领域
本发明属于生物医学和数据处理技术领域，具体涉及一种快速检测新型冠状病毒肺炎的方法及系统。
背景技术
公开该背景技术部分的信息仅仅旨在增加对本发明的总体背景的理解，而不必然被视为承认或以任何形式暗示该信息构成已经成为本领域一般技术人员所公知的现有技术。
一种新型呼吸系统疾病，主要临床症状表现为发热、乏力、干咳。经分离鉴定，确认其病原为一种新型冠状病毒，国际病毒分类学委员会将其命名为“SARS-CoV-2”。这是目前在人体中发现的第七种冠状病毒。2020年2月11日，世界卫生组织在日内瓦召开的全球研究与创新论坛将新型冠状病毒导致的疾病正式命名为COVID-19。
根据当前新型冠状病毒肺炎诊疗方案(试行第七版)的规定，疑似病例同时具备以下病原学或血清学证据者之一：实时荧光RT-PCR检测新型冠状病毒核酸阳性、病毒基因测序，与已知新型冠状病毒高度同源、血清新型冠状病毒特异性IgM和IgG抗体阳性、血清新型冠状病毒特异性抗体IgG抗体由阴性转为阳性或恢复期较急性期4倍及以上升高，即可确诊为新型冠状病毒肺炎患者。但是发明人发现，采用实时荧光RT-PCR或者病毒基因组测序对技术、设备，操作人员的要求较高，且耗时较长，同时采样过程繁琐极易引起人体不适；而特异性抗体检测则需要较长窗口期。
发明内容
针对上述现有技术的不足，本发明提供一种快速检测新型冠状病毒肺炎的方法及系统，本发明通过以COVID-19肺炎患者和健康人这两类人作为分类目标，将呼气NO浓度与人体基本特征同时作为特征量来对COVID-19肺 炎患者和健康人进行分类，诊断正确率高达90％以上。本发明检测成本低，速度快，准确率高，因此具有良好的实际应用之价值。
本发明是通过如下技术方案实现的：
本发明的第一个方面，提供一种快速检测新型冠状病毒肺炎的方法，所述检测方法包括：
获取受试者的人体基本体征信息和呼气NO浓度数据作为特征量；
根据预先基于特征量建立的新冠肺炎人体检测模型，判断受试者患病概率。
其中，新冠肺炎人体检测模型的构建方法包括：基于算法对于采集到的人体特征量进行训练，获得新冠肺炎人体检测模型，从而实现对健康人和新冠肺炎患者的分类。
需要说明的是，新冠肺炎人体检测模型的构建方法中，人体特征量采集来源包括健康人、新型冠状病毒肺炎患者或新型冠状病毒肺炎疑似患者。
所述受试者可以是健康人、新型冠状病毒肺炎患者或新型冠状病毒肺炎疑似患者；优选为新冠肺炎疑似患者。
所述算法是本领域技术人员公知的，由于本发明中处理对象为不连续的分类标签值，因此优选算法为机器学习分类算法，所述机器学习分类算法包括随机森林方法，支持向量机算法，K临近算法等，优选为随机森林算法。
所述特征量包括人体基本体征信息和呼气NO浓度；
所述人体基本体征信息包括但不限于身高、体重、年龄、性别、体重身高指数、体表面积、血型、有无吸烟史、有无其他肺部或呼吸道疾病等；
所述判断受试者患病概率的具体方法为：将受试者基本体征信息和呼气NO浓度数据导入上述新冠肺炎人体检测模型中，基于新冠肺炎人体检测模型判断该受试者患病概率。
本发明的第二个方面，提供一种用于检测新型冠状病毒肺炎的系统，所 述检测系统至少包括：
数据处理模块，用于对受试者的医疗数据进行处理，以获得所述受试者的特征量，其中，所述特征量包括受试者呼气NO浓度数据和人体基本体征信息；
数据分析模块，用于通过预设分析模型处理所述特征量，以确定所述受试者的新冠肺炎检测结果；
所述预设分析模型，其是基于算法对于采集到的人体特征量进行训练获得，从而实现对健康人和新冠肺炎患者的分类。
其中，NO浓度数据可通过呼出气NO检测仪获得；
所述人体基本体征信息包括但不限于身高、体重、年龄、性别、体重身高指数、体表面积、血型、有无吸烟史、有无其他肺部或呼吸道疾病等；
所述算法是本领域技术人员公知的，由于本发明中处理对象为不连续的分类标签值，因此优选算法为机器学习分类算法，所述机器学习分类算法包括随机森林方法，支持向量机算法，K临近算法等，优选为随机森林算法。
人体特征量采集来源包括健康人、新型冠状病毒肺炎患者或新型冠状病毒肺炎疑似患者。
人体特征量包括人体呼气NO浓度数据和人体基本体征信息。
本发明的第三个方面，提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现上述的用于快速检测新型冠状病毒肺炎的方法。
本发明的第四个方面，提供一种电子设备，包括：
存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现上述用于快速检测新型冠状病毒肺炎的方法。
以上一个或多个技术方案的有益技术效果：
上述技术方案提出的检测用于快速检测新型冠状病毒肺炎的方法及系统中，模型建立时选取的对象既包括COVID-19患者又包括健康人，受试者预测的对象也是包括健康人和COVID-19患者，同时基于随机森林方法进行模型的构建，从而使得检测结果准确度大大提高，经检测，准确度达到90％以上。
上述技术方案具有成本低、快速、无损、检测准确度高等优点，受试者无痛苦，顺从性好，具有良好的实际应用之价值。
附图说明
构成本发明的一部分的说明书附图用来提供对本发明的进一步理解，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。
图1为本发明新型冠状病毒肺炎检测的流程示意图。
图2为本发明实施例1中COVID-19患者和健康人的特征量统计分布图其中，(a)-(f)分别为本发明实施例1中COVID-19患者和健康人的身高、体重、BMI、体表面积、年龄和FeNO的统计分布图。
图3为本发明实施例1中建立的基于呼气NO浓度、身高体重指数、体表面积、年龄的机器学习分类模型；其中，(a)为受试者特征曲线，(b)为该模型的训练效果图，(c)是该模型在测试集中的预测结果。
图4为本发明实施例2中建立的基于呼气NO浓度、年龄、身高以及体重的机器学习分类模型；其中，(a)为受试者特征曲线，(b)为该模型的训练效果图，(c)为该模型在测试集中的预测结果。
图5为本发明实施案例3中建立的基于呼气NO浓度、身高、体重、身高体重指数、体表面积、年龄的机器学习分类模型；其中，(a)为受试者特征曲线，(b)为该模型的训练效果图，(c)是该模型在测试集中的预测结果。
具体实施方式
应该指出，以下详细说明都是例示性的，旨在对本发明提供进一步的说明。除非另有指明，本文使用的所有技术和科学术语具有与本发明所属技术领域的普通技术人员通常理解的相同含义。
需要注意的是，这里所使用的术语仅是为了描述具体实施方式，而非意图限制根据本发明的示例性实施方式。如在这里所使用的，除非上下文另外明确指出，否则单数形式也意图包括复数形式，此外，还应当理解的是，当在本说明书中使用术语“包含”和/或“包括”时，其指明存在特征、步骤、操作、器件、组件和/或它们的组合。应理解，本发明的保护范围不局限于下述特定的具体实施方案；还应当理解，本发明实施例中使用的术语是为了描述特定的具体实施方案，而不是为了限制本发明的保护范围。
本发明的一个具体实施方式中，提供一种快速检测新型冠状病毒肺炎的方法，所述检测方法包括：
获取受试者的人体基本体征信息和呼气NO浓度数据处理后作为特征量；
根据预先基于特征量建立的新冠肺炎人体检测模型，判断受试者患病概率。
需要说明的是，新冠肺炎人体检测模型的构建方法中，人体特征量采集来源包括健康人、新型冠状病毒肺炎患者或新型冠状病毒肺炎疑似患者。
所述算法是本领域技术人员公知的，由于本发明中处理对象为不连续的分类标签值，因此优选算法为机器学习分类算法，所述机器学习分类算法包括随机森林方法，支持向量机算法，K临近算法等，优选为随机森林算法。随机森林是一个包含多个决策树的分类器，并且其输出的类别是由个别树输出的类别的众数而定。
所述特征量包括人体基本体征信息和呼气NO浓度；
所述人体基本体征信息包括但不限于身高、体重、年龄、性别、体重身 高指数、体表面积、血型、有无吸烟史、有无其他肺部或呼吸道疾病等；
其中，新冠肺炎人体检测模型的构建方法包括：基于算法对于采集到的人体特征量进行训练，获得新冠肺炎人体检测模型，从而实现对健康人和新冠肺炎患者的分类。
具体的，新冠肺炎人体检测模型的构建方法中，从整个数据中抽取25％作为测试集，剩下的75％作为训练集，采用10折交叉验证(10-fold Cross-validation)方法进行训练，从而选择最佳的随机森林分类参数。
以训练好的模型计算测试集中样本的患病概率。分类准确率可以达到90％以上，阳性检出率90％以上，阴性判断正确率也可以达到90％以上。受试者特征曲线线下面积90％以上。本申请的新冠肺炎人体检测模型用于诊断时会输出受试者患病的概率，根据概率值的大小来判断受试者是否患COVID-19。
所述受试者可以是健康人、新型冠状病毒肺炎患者或新型冠状病毒肺炎疑似患者。
所述检测受试者患病概率具体方法为：将受试者基本体征信息和呼气NO浓度数据导入上述新冠肺炎人体检测模型中，基于新冠肺炎人体检测模型判断该受试者患病概率。
其中，呼气NO浓度的测定需要严格遵照美国胸科学会(ATS)与欧洲呼吸学会(ERS)2005年共同制定的技术标准。
饮食方面：要求测试者前3小时不得饮食；环境方面：要求测试者测试时避免吸入NO>10ppb的空气。
其他方面：测试前1小时不得吸烟、饮酒，避免运动或做其他肺部功能检测或其他检测，除此之外测试时还应避免漏气、换气、憋气以及喷口水等。
关于仪器方面，测试时需要监控呼气压力、时间以及流速，呼出气NO检测仪的检测限要低于5ppb，分析准确性/重复性小于等于5(10)ppb或 10％(20％)。仪器必须定期及时标定，保证仪器的准确性与稳定性。
本发明的又一具体实施方式中，提供一种用于检测新型冠状病毒肺炎的系统，所述检测系统至少包括：
数据处理模块，用于对受试者的医疗数据进行处理，以获得所述受试者的特征量，其中，所述医疗数据包括受试者呼气NO浓度数据和人体基本体征信息；
数据分析模块，用于通过预设分析模型处理所述特征量，以确定所述受试者的新冠肺炎检测结果；
所述预设分析模型，其是基于算法对于采集到的人体特征量进行训练获得，从而实现对健康人和新冠肺炎患者的分类。
在一些实施方式中，NO浓度数据可通过呼出气NO检测仪获得；
在一些实施方式中，所述人体基本体征信息包括但不限于身高、体重、年龄、性别、体重身高指数、体表面积、血型、有无吸烟史、有无其他肺部或呼吸道疾病等；
所述算法是本领域技术人员公知的，由于本发明中处理对象为不连续的分类标签值，因此优选算法为机器学习分类算法，所述机器学习分类算法包括随机森林方法，支持向量机算法，K临近算法等，优选为随机森林算法。
在一些实施方式中，人体特征量采集来源包括健康人、新型冠状病毒肺炎患者或新型冠状病毒肺炎疑似患者。
本发明的又一具体实施方式中，提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现上述的用于快速检测新型冠状病毒肺炎的方法。
本发明的又一具体实施方式中，提供一种电子设备，包括：
存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现上述用于快速检测新型冠状病毒肺炎的方 法。
以下通过实施例对本发明做进一步解释说明，但不构成对本发明的限制。应理解这些实施例仅用于说明本发明而不用于限制本发明的范围。另外，实施例中未详细说明的生物学方法均为本领域常规的方法，具体操作可参看生物指南或产品说明书。下述实施例在华中科技大学同济医学院附属梨园医院进行，于2020年3月7号开始采集，包括109个未患COVID-19的样本和46个罹患COVID-19的样本。记录了他们的身高、体重、年龄、性别、血型、患病时间、有无其他肺部疾病等信息。然后通过对样本进行处理得到了分析的结果。呼气NO浓度的测定需要严格遵照美国胸科学会(ATS)与欧洲呼吸学会(ERS)2005年共同制定的技术标准。饮食方面：要求测试者前3小时不得饮食；环境方面：要求测试者测试时避免吸入NO>10ppb的空气。其他方面：测试前1小时不得吸烟、饮酒，避免运动或做其他肺部功能检测或其他检测，除此之外测试时还应避免漏气、换气、憋气以及喷口水等。关于仪器方面，测试时需要监控呼气压力、时间以及流速，呼出气NO检测仪的检测限要低于5ppb，分析准确性/重复性小于等于5(10)ppb或10％(20％)。仪器必须定期及时标定，保证仪器的准确性与稳定性。
实施例1
采集受试者的基本体征信息并测得呼气NO浓度。NO浓度测试参数设置及流程为：
NO浓度由呼出气NO检测仪在室温，相对湿度范围不大于80％的环境下测试得到。测试人员分别记录了非新冠肺炎患者人群的身高、体重、年龄、性别、疾病史等基本体征信息，同时测量了呼出气中NO浓度等信息。
图3是本发明采用FeNO、年龄、身高体重指数、体表面积为描述符建立的机器学习分类模型的受试者曲线，由图可得，受试者曲线下面积达到 了AUC＝1.0。图3(b)是该模型训练示意图，图3(c)是该模型在测试集中的预测结果。由模拟结果可得该模型阳性检出率可达91.7％，阴性判断正确率为91.3％。
实施例2
采集非新冠肺炎患者的基本体征信息并测得呼气NO浓度。NO浓度测试参数设置及流程为：
NO浓度由呼出气NO检测仪在室温，相对湿度范围不大于80％的环境下测试得到。测试人员分别记录了非新冠肺炎患者人群的身高、体重、年龄、性别、疾病史等基本体征信息，同时测量了呼出气中NO浓度等信息。
图4(a)是本发明采用FeNO、年龄、身高、体重为描述符建立的机器学习分类模型的受试者曲线，由图可得，受试者曲线下面积达到了AUC＝1.0。图4(b)是该模型训练示意图，图4(c)是该模型在测试集中的预测结果。由模拟结果可得该模型阳性检出率可达83.3％，阴性判断正确率为87.0％。
实施例3
采集非新冠肺炎患者的基本体征信息并测得呼气NO浓度。NO浓度测试参数设置及流程为：
NO浓度由呼出气NO检测仪在室温，相对湿度范围不大于80％的环境下测试得到。测试人员分别记录了非新冠肺炎患者人群的身高、体重、年龄、性别、疾病史等基本体征信息，同时测量了呼出气中NO浓度等信息。
图5(a)是本发明采用FeNO、年龄、身高、体重、身高体重指数、体表面积为描述符建立的机器学习分类模型的受试者曲线，由图可得，受试者曲线下面积达到了AUC＝1.0。图5(b)是该模型训练示意图，图5(c)是该模型在测试集中的预测结果。由模拟结果可得该模型阳性检出率可达83.3％， 阴性判断正确率为87.0％。
在另一实施例中，提供了一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如图1所示的快速检测新型冠状病毒肺炎方法中的步骤。
在另一实施例中，提供了一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现如图1所示的快速检测新型冠状病毒肺炎方法中的步骤。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用硬件实施例、软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器和光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备 上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(RandomAccessMemory，RAM)等。
应注意的是，以上实例仅用于说明本发明的技术方案而非对其进行限制。尽管参照所给出的实例对本发明进行了详细说明，但是本领域的普通技术人员可根据需要对本发明的技术方案进行修改或者等同替换，而不脱离本发明技术方案的精神和范围。

Claims (10)

  1. 一种快速检测新型冠状病毒肺炎的方法，其特征在于，所述检测方法包括：
    获取受试者的人体基本体征信息和呼气NO浓度数据处理后作为特征量；
    根据预先基于特征量建立的新冠肺炎人体检测模型，判断受试者患病概率；
    其中，新冠肺炎人体检测模型的构建方法中，人体特征量采集来源包括健康人、新型冠状病毒肺炎患者或新型冠状病毒肺炎疑似患者；
    所述特征量包括人体基本体征信息和呼气NO浓度。
  2. 如权利要求1所述的方法，其特征在于，所述人体基本体征信息包括身高、体重、年龄、性别、体重身高指数、体表面积、血型、有无吸烟史、有无其他肺部或呼吸道疾病。
  3. 如权利要求1所述的方法，其特征在于，新冠肺炎人体检测模型的构建方法包括：基于算法对于采集到的人体特征量进行训练，获得新冠肺炎人体检测模型，从而实现对健康人和新冠肺炎患者的分类；
    优选的，新冠肺炎人体检测模型的构建方法中，从整个数据中抽取25％作为测试集，剩下的75％作为训练集，采用10折交叉验证方法进行训练。
  4. 如权利要求1所述的方法，其特征在于，所述受试者是健康人、新型冠状病毒肺炎患者或新型冠状病毒肺炎疑似患者；优选为新型冠状病毒肺炎疑似患者。
  5. 如权利要求1所述的方法，其特征在于，所述判断受试者患病概率具体方法为：将受试者基本体征信息和呼气NO浓度数据导入所述新冠肺炎人体检测模型中，基于新冠肺炎人体检测模型判断该受试者患病概率。
  6. 一种用于检测新型冠状病毒肺炎的系统，其特征在于，所述检测系统至少包括：
    数据处理模块，用于对受试者的医疗数据进行处理，以获得所述受试者的特征量，其中，所述医疗数据包括受试者呼气NO浓度数据和人体基本体征信息；
    数据分析模块，用于通过预设分析模型处理所述特征量，以确定所述受试者的新冠肺炎检测结果；
    所述预设分析模型，其是基于算法对于采集到的人体特征量进行训练获得，从而实现对健康人和新冠肺炎患者的分类。
  7. 如权利要求6所述的系统，其特征在于，所述人体基本体征信息包括身高、体重、年龄、性别、体重身高指数、体表面积、血型、有无吸烟史、有无其他肺部或呼吸道疾病。
  8. 如权利要求6所述的系统，其特征在于，所述算法为机器学习分类算法；优选的，所述机器学习分类算法包括随机森林方法，支持向量机算法，K临近算法；进一步优选为随机森林算法；
    人体特征量采集来源包括健康人、新型冠状病毒肺炎患者或新型冠状病毒肺炎疑似患者；
    人体特征量包括人体呼气NO浓度数据和人体基本体征信息。
  9. 一种计算机可读存储介质，其特征在于，其上存储有计算机程序，所述计算机程序被处理器执行时实现权利要求1-5任一项所述的用于快速检测新型冠状病毒肺炎的方法。
  10. 一种电子设备，其特征在于，包括：
    存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现权利要求1-5任一项所述用于快速检测新型冠状病毒肺炎的方法。
PCT/CN2020/127489 2020-07-30 2020-11-09 一种快速检测新型冠状病毒肺炎的方法及系统 WO2022021664A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010753816.7A CN111951964A (zh) 2020-07-30 2020-07-30 一种快速检测新型冠状病毒肺炎的方法及系统
CN202010753816.7 2020-07-30

Publications (1)

Publication Number Publication Date
WO2022021664A1 true WO2022021664A1 (zh) 2022-02-03

Family

ID=73338775

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/127489 WO2022021664A1 (zh) 2020-07-30 2020-11-09 一种快速检测新型冠状病毒肺炎的方法及系统

Country Status (2)

Country Link
CN (1) CN111951964A (zh)
WO (1) WO2022021664A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112525804B (zh) * 2020-11-21 2022-07-22 烟台智医医药科技有限公司 全血细胞计数在预测SARS-CoV-2感染中的应用
CN112766352B (zh) * 2021-01-13 2024-03-29 大连海事大学 一种基于极端梯度提升算法的新型冠状病毒分类方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200027558A1 (en) * 2016-09-26 2020-01-23 The University Of Queensland A method and apparatus for automatic disease state diagnosis
CN111261281A (zh) * 2020-01-18 2020-06-09 中国医学科学院北京协和医院 一种模型训练方法、机械通气表型识别方法及呼吸机
CN111297403A (zh) * 2020-02-25 2020-06-19 中国矿业大学 一种尘肺群体快速精准筛查及早期预警方法
CN111402347A (zh) * 2020-03-20 2020-07-10 吴刚 一种基于物联网的新冠肺炎疫情防控系统
CN111413492A (zh) * 2020-04-30 2020-07-14 山东大学 一种用于检测新型冠状病毒covid-2019肺炎的方法及系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018129414A1 (en) * 2017-01-08 2018-07-12 The Henry M. Jackson Foundation For The Advancement Of Military Medicine, Inc. Systems and methods for using supervised learning to predict subject-specific pneumonia outcomes
CN109817328A (zh) * 2018-12-29 2019-05-28 上海依智医疗技术有限公司 一种判别哮喘的方法及装置
CN111261282A (zh) * 2020-01-21 2020-06-09 南京航空航天大学 一种基于机器学习的脓毒症早期预测方法
CN111041089B (zh) * 2020-03-13 2020-06-19 广州微远基因科技有限公司 Covid-19感染的宿主标志物的应用

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200027558A1 (en) * 2016-09-26 2020-01-23 The University Of Queensland A method and apparatus for automatic disease state diagnosis
CN111261281A (zh) * 2020-01-18 2020-06-09 中国医学科学院北京协和医院 一种模型训练方法、机械通气表型识别方法及呼吸机
CN111297403A (zh) * 2020-02-25 2020-06-19 中国矿业大学 一种尘肺群体快速精准筛查及早期预警方法
CN111402347A (zh) * 2020-03-20 2020-07-10 吴刚 一种基于物联网的新冠肺炎疫情防控系统
CN111413492A (zh) * 2020-04-30 2020-07-14 山东大学 一种用于检测新型冠状病毒covid-2019肺炎的方法及系统

Also Published As

Publication number Publication date
CN111951964A (zh) 2020-11-17

Similar Documents

Publication Publication Date Title
Bansal et al. Cough Classification for COVID-19 based on audio mfcc features using Convolutional Neural Networks
Davis et al. Exhaled breath testing–a tool for the clinician and researcher
Singh et al. Effective overview of different ML models used for prediction of COVID-19 patients
WO2021217747A1 (zh) 一种用于检测新型冠状病毒covid-2019肺炎的方法及系统
Berna et al. Reproducible breath metabolite changes in children with SARS-CoV-2 infection
Baker et al. Failure of infant observation scales in detecting serious illness in febrile, 4-to 8-week-old infants
van Geffen et al. Diagnosing viral and bacterial respiratory infections in acute COPD exacerbations by an electronic nose: a pilot study
Furr et al. Development of a colic severity score for predicting the outcome of equine colic
Ingram et al. Eosinophil cationic protein in serum and nasal washes from wheezing infants and children
WO2022021664A1 (zh) 一种快速检测新型冠状病毒肺炎的方法及系统
CN104856679A (zh) 用于哮喘、肺结核及肺癌诊断及疾病管控的呼吸分析系统及方法
CN106456053A (zh) 用于气道疾病评估的呼出气的选择、分段及分析
CN111816321B (zh) 基于法定诊断标准智能识别传染病的系统、设备及存储介质
CN111944927A (zh) 新冠病毒核酸10合1混采检测技术
Infante et al. Classification of voluntary coughs applied to the screening of respiratory disease
Xue et al. Deep Convolutional Neural Network Approach for COVID-19 Detection.
Mijwil Deep Convolutional Neural Network Architecture to Detect COVID-19 from Chest X-Ray Images
Arnold et al. A comparison efficacy study of commercial nasopharyngeal swabs versus a novel 3D printed swab for the detection of SARS-CoV-2
Suh et al. Bronchial hyperresponsiveness to methacholine and adenosine 5′‐monophosphate, and the presence and degree of atopy in young children with asthma
Abeyratne et al. Cough Sound Analysis-A new tool for diagnosing Pneumonia
KR101785788B1 (ko) 호흡가스 분석에 기반하여 미병 분류를 보조하는 컴퓨팅 시스템 및 방법
Tripathi et al. Coronavirus: Diagnosis, detection, and analysis
CN113066584A (zh) 早发型败血症的预测方法及系统
Zamora-Mendoza et al. Determination of global chemical patterns in exhaled breath for the discrimination of lung damage in postCOVID patients using olfactory technology
JP2023145811A (ja) 学習モデルの生成方法、プログラム、演算装置

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: 20947506

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20947506

Country of ref document: EP

Kind code of ref document: A1