Evaluation device, evaluation system, evaluation method, evaluation program, and recording medium

Abstract

An evaluation device (4) for evaluating the degree of severity of pneumonia of a patient of interest is provided with: an acquisition unit (42) that acquires a respiratory waveform of the patient of interest; a calculation unit (43) that calculates an index value indicating the instability of a respiratory cycle or a respiration frequency from the respiratory waveform; and an evaluation unit (44) that evaluates the degree of severity on the basis of the calculated value.

Description

Technical Field

[0001] The present invention relates to a technique for evaluating the severity of pneumonia, and particularly to a technique for evaluating the severity of pneumonia caused by COVID-19.

Background Art

[0002] COVID-19 has spread to various countries around the world. By the end of January 2021, the cumulative number of patients had exceeded 100 million, and the death toll had reached 2.1 million. In Japan as well, on January 8, 2021, 7,882 new patients were reported in a single day, and it has become impossible to admit all patients to medical institutions. That is, most COVID-19 patients are required to stay at home or in accommodation for treatment, and in the process, it has become necessary to admit only those patients who have become severely ill and require advanced treatment.

[0003] In response to this, the Ministry of Health, Labour and Welfare has stipulated that regarding the health observation of COVID-19 patients staying at home or in accommodation, prefectural health centers and others should regularly grasp the health status of these patients using information and communication devices such as telephones, and establish a system to enable these patients to promptly visit an appropriate medical institution when their symptoms worsen (page 4 of the following Non-Patent Document 1). In addition, the delivery of pulse oximeters to patients receiving home treatment is also being promoted to conduct health observation of these patients by checking their oxygen saturation (the following Non-Patent Document 2).

Prior Art Documents

Non-Patent Documents

[0004]

Non-Patent Document 1

Non-Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] In addition to respiratory failure such as pneumonia, COVID-19 is characterized by causing complications such as cardiovascular diseases such as arrhythmia and heart disorders, and thromboembolism such as pulmonary embolism and acute stroke, and rapidly deteriorating to death. Currently, COVID-19 patients undergoing home or accommodation-based convalescence are triaged for the need for hospitalization based on their subjective symptoms, body temperature, and oxygen saturation checked by the patients themselves. However, subjective symptoms and body temperature have low specificity for pneumonia and are not easily used as a basis for triage judgment. Although oxygen saturation is important information for triage, it requires the patient to measure it themselves while in poor physical condition and report the data to administrators such as the health center, which places a burden on the patient. And despite these self-checks, there have been successive cases of patients undergoing home or accommodation-based convalescence dying.

[0006] Thus, in order to prevent the death of patients undergoing home or accommodation-based convalescence, there is a need for a system that does not require the action of the patients themselves, can remotely monitor the conditions of a large number of patients, and can easily determine the necessity of hospitalization from the information.

[0007] The present invention has been made to solve the above problems, and an object thereof is to simply evaluate the severity of pneumonia.

Means for Solving the Problems

[0008] In order to solve the above problems, the present invention includes the following aspects. Item 1. An evaluation device for evaluating the severity of pneumonia in a target patient, An acquisition unit that acquires the respiratory waveform of the target patient; A calculation unit that calculates a value of an index indicating instability of the respiratory cycle or respiratory frequency from the respiratory waveform; An evaluation unit that evaluates the severity based on the calculated value; An evaluation device comprising the above. Item 2. The evaluation device according to Item 1, wherein the calculation unit calculates an RST, which is the reciprocal of the standard deviation of the respiratory frequency, as the value of the index. Item 3. The evaluation device according to Item 2, wherein the evaluation unit evaluates that the severity is high when the RST is less than a threshold value. Item 4. The evaluation device according to any one of Items 1 to 3, wherein the pneumonia is pneumonia caused by COVID-19. Item 5. A detection device that detects a signal including the respiratory waveform of the target patient; The evaluation device according to any one of Items 1 to 4; An evaluation system comprising the above. Item 6. The evaluation system according to Item 5, wherein the detection device includes a seat sensor provided under the target patient in the bed, and detects a body pressure signal generated by the seat sensor as the signal. Item 7. An evaluation method for evaluating the severity of pneumonia in a target patient, comprising: An acquisition step of acquiring the respiratory waveform of the target patient; A calculation step of calculating a value of an index indicating instability of the respiratory cycle or respiratory frequency from the respiratory waveform; An evaluation step of evaluating the severity based on the calculated value; An evaluation method comprising the above. Item 8. An evaluation program for operating a computer as each part of the evaluation device according to any one of Items 1 to 4. Item 9. A computer-readable recording medium recording the evaluation program according to Item 8.

Advantages of the Invention

[0009] According to the present invention, the severity of pneumonia is evaluated based on the value of an index indicating the instability of the respiratory cycle or respiratory frequency such as RST. Since the index can be easily monitored from a remote target patient, the severity of pneumonia can be easily evaluated.

Brief Description of Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0011] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Note that the present invention is not limited to the following embodiments.

[0012] (System Configuration) FIG. 1 is a block diagram showing a schematic configuration of an evaluation system 1 according to an embodiment of the present invention. The evaluation system 1 is a system for evaluating the severity of pneumonia in a target patient, and includes a detection device 2, a relay terminal 3, an evaluation device 4, and a browsing terminal 5.

[0013] In this embodiment, the target patient is determined to be positive for COVID-19 and is a person who is recuperating at home or in a lodging facility. The detection device 2 and the relay terminal 3 are provided in the home or lodging facility where the target patient stays. The evaluation device 4 is provided on the cloud, and the browsing terminal 5 is provided in an institution (such as a health center or a hospital) that observes the health of the target patient. Note that the evaluation device 4 and the browsing terminal 5 may be configured as a single device.

[0014] FIG. 2 is a schematic diagram showing an installation example of the detection device 2 and the relay terminal 3. The detection device 2 is a device that detects a signal including the respiratory waveform of a target patient, and includes a sheet sensor 21 and a measurement unit 22.

[0015] The sheet sensor 21 is a piezoelectric type body movement sensor composed of a thin and soft strip-shaped piezoelectric rubber, and is provided, for example, on the bed used by the target patient. In FIG. 2, the illustration of the sheet and the comforter laid on the sheet sensor 21 is omitted. While the target patient is in bed, pressure is applied to the sheet sensor 21, and the sheet sensor 21 generates an analog body pressure signal by the piezoelectric effect. The body pressure signal includes, in addition to the respiratory waveform, waveforms due to body movements other than breathing and noise.

[0016] The measurement unit 22 is connected to the sheet sensor 21, and performs AD conversion on the body pressure signal generated by the sheet sensor 21. The measurement unit 22 also has a function of communicating with the relay terminal 3 using Bluetooth (registered trademark), and transmits the body pressure signal to the relay terminal 3.

[0017] The relay terminal 3 is composed of a smartphone and transmits the body pressure signal received from the measurement unit 22 to the evaluation device 4 on the cloud via the Internet N. If the measurement unit 22 can be connected to the Internet N, the relay terminal 3 may be omitted and the body pressure signal may be directly transmitted from the measurement unit 22 to the evaluation device 4.

[0018] The evaluation device 4 shown in FIG. 1 can be configured as a server device. The evaluation device 4 includes a storage unit 41, an acquisition unit 42, a calculation unit 43, and an evaluation unit 44.

[0019] The storage unit 41 can be configured by, for example, an HDD or an SSD. Various data such as the evaluation program D1 are stored in the storage unit 41.

[0020] Each of the acquisition unit 42, the calculation unit 43, and the evaluation unit 44 may be realized hardware-wise by a logic circuit or the like, or may be realized software-wise using a CPU or the like. When each of the above units is realized software-wise, the evaluation program D1 can be read out and executed by the CPU or the like of the evaluation device 4 in the main storage device to realize each of the above units. The evaluation program D1 may be downloaded to the evaluation device 4 via a communication network such as the Internet N, or may be installed in the evaluation device 4 via a computer-readable non-transitory recording medium such as a CD-R on which the evaluation program D1 is recorded.

[0021] The acquisition unit 42 has a function of acquiring the respiratory waveform of the target patient. In the present embodiment, the acquisition unit 42 extracts the respiratory waveform by removing waveforms due to body movements other than breathing and noise from the body pressure signal received from the detection device 2 via the relay terminal 3.

[0022] The calculation unit 43 has a function of calculating a value of an index indicating the instability of the respiratory cycle or respiratory frequency from the respiratory waveform acquired by the acquisition unit 42. In the present embodiment, the calculation unit 43 calculates, by performing frequency analysis (maximum entropy method) on the respiratory waveform, RST (Respiratory Stability Time), which is the reciprocal of the standard deviation of the respiratory frequency, as the value of the index. Specifically, RST is calculated by extracting the frequency band of the respiratory cycle from the respiratory waveform, obtaining the standard deviation of frequencies of 5% or more of the maximum value of the respiratory frequency component, and taking the reciprocal of the standard deviation. RST = 1 / A A: Standard deviation of frequencies of 5% or more of the maximum value of the respiratory frequency component

[0023] The evaluation unit 44 has a function of evaluating the severity of pneumonia of the target patient based on the value calculated by the calculation unit 43. In the present embodiment, the evaluation unit 44 evaluates that the severity is high when RST is less than the threshold value. When the target patient is a positive case of COVID-19, the threshold value is preferably 20 to 30 seconds, and more preferably 26 seconds.

[0024] The evaluation result by the evaluation unit 44 is stored in the storage unit 41 and transmitted to the browsing terminal 5 via the Internet N. In the browsing terminal 5, the RST of the target patient is displayed in a list, and for the target patient evaluated to have a high severity, the numerical value is displayed in a different manner (such as different colors, highlighting, etc.) from other patients. Thereby, the user (such as the administrator of the health center) of the browsing terminal 5 can easily identify the target patient who needs hospitalization.

[0025] (Processing procedure) FIG. 3 is a flowchart showing a processing procedure of an evaluation method for evaluating the severity of pneumonia of a target patient using the evaluation system 1 according to the present embodiment.

[0026] In step S1, the seat sensor 21 of the detection device 2 detects the body pressure signal of the target patient who is in bed. In step S2 (acquisition step), the acquisition unit 42 of the evaluation device 4 acquires the respiratory waveform of the target patient from the body pressure signal received from the detection device 2. In step S3 (calculation step), the calculation unit 43 of the evaluation device 4 calculates the RST from the respiratory waveform acquired by the acquisition unit 42. In step S4 (evaluation step), the evaluation unit 44 of the evaluation device 4 evaluates the severity of pneumonia of the target patient based on the value calculated by the calculation unit 43. In step S5, the browsing terminal 5 displays the evaluation result by the evaluation unit 44.

[0027] Note that the operating entity of each step is not limited to the above.

[0028] (Effect) As described above, in this embodiment, the severity of pneumonia is evaluated based on the RST. The RST is obtained by automatically extracting and analyzing the respiratory waveform from the body pressure signal generated by the seat sensor 21 provided under the target patient who is in bed. That is, the RST does not require a special action of the patient himself like a pulse oximeter, and can be obtained in a non-invasive and unconstrained environment for the patient. In addition, since it is not necessary for an administrator or the like to face the target patient, the RST can be easily, inexpensively and continuously monitored every day from a remote location even for patients isolated at home or in a lodging and convalescent facility due to infectious diseases such as COVID-19. Furthermore, if the RST is less than a certain threshold value, it can be determined that the condition has deteriorated such as the onset of pneumonia.

[0029] Thus, by using the evaluation system according to this embodiment, it becomes possible to detect patients who need treatment in the hospital based on the RST. As a result, patients who need hospitalization can be hospitalized in a timely manner, so that not only can the death of patients during home or lodging and convalescence be reduced, but also a sense of security of being watched remotely can be given to the patients.

[0030] In addition, since a small number of employees at a health center or the like can manage a large number of patients who are receiving home or inpatient medical treatment in a centralized manner, it also leads to the effective utilization of medical resources. Also, for patients, the burden on the mind and body and the economic burden can be reduced by detecting early deterioration.

[0031] (Principle) RST is an index created by the inventor of the present application to quantitatively evaluate changes in breathing patterns.

[0032] There are four stretch receptors in the bronchi and interstitium of the lungs. An increase in pulmonary artery pressure or interstitial edema due to heart failure stimulates these biosensors and destabilizes breathing. Also, circulatory delay and enhanced central chemoreflex due to heart failure also destabilize breathing and induce periodic breathing. RST, which can be calculated from the frequency analysis of the breathing waveform, is excellent in that it can quantify these unstable breaths with a single index.

[0033] In the case of pneumonia patients such as pneumonia caused by COVID-19 or aspiration pneumonia, it is considered that the lung stretch receptors are stimulated by the inflammation of the lung parenchyma and breathing becomes unstable. Therefore, it is highly likely that the severity of pneumonia can be detected early by remotely monitoring RST in the evaluation system according to this embodiment. Also, the evaluation system according to this embodiment enables proactive medical treatment that emphasizes individuality, leading to an improvement in the QOL of patients such as avoiding early treatment and hospitalization and medical and economic effects.

[0034] (Supplementary Notes) As described above, the embodiments of the present invention have been explained, but the present invention is not limited to the above embodiments, and various changes are possible without departing from the spirit thereof.

[0035] In the above embodiment, the respiration waveform is obtained by removing waveforms due to body movements other than respiration and noise from the body pressure signal detected by the seat sensor. However, the present invention is not limited to this. For example, the respiration waveform may be obtained by a respiration airflow sensor attached to the skin surface near the nasal cavity of the target patient. Additionally, as a respiration signal detection device, a Depth sensor, a radar Doppler, an ultrasonic Doppler, etc. may be used. However, from the perspective of simplicity, the above embodiment is more preferable because it increases the burden on the patient himself / herself.

[0036] Also, in the above embodiment, the RST was used as an index for evaluating the severity of pneumonia. However, it is not particularly limited as long as it is an index indicating the instability of the respiration cycle or respiration frequency that is equivalent to the RST.

[0037] In addition to the RST, the respiration rate and heart rate may also be considered to evaluate the severity of pneumonia. In this case, the calculation unit 43 shown in FIG. 1 calculates the RST, respiration rate, and heart rate from the respiration waveform, and the evaluation unit 44 evaluates the severity of pneumonia based on the RST, respiration rate, and heart rate. Thereby, the accuracy of the evaluation can be further improved.

Example

[0038] Hereinafter, examples of the present invention will be described. Note that the present invention is not limited to the following examples.

[0039] The inventors of the present application remotely monitored the overnight RST from the body pressure signal of the seat sensor 21 for 40 patients nationwide in Japan from July 31 to November 19, 2020 (32 COVID-19 patients and 8 pneumonia patients). As the seat sensor 21, a body movement sensor manufactured by Sumitomo Riko Co., Ltd. was used.

[0040] As a result, the patients with "Moderate II·Severe" shown in the "Guidelines for COVID-19 Treatment" issued by the Ministry of Health, Labour and Welfare had significantly lower RST values compared to the patients with "Mild·Moderate I" (Figure 4). Similarly, the patients with "Medium or High" shown by the National Early Warning Score (NEWS) had significantly lower RST values compared to the patients with "Low or Low-medium" (Figure 5). In addition, the patients with pneumonia (with infiltration) shown by CT had significantly lower RST values compared to the patients without pneumonia (Figure 6).

[0041] When examining the RST at the time of hospitalization with the highest sensitivity and specificity for the presence or absence of future severe exacerbation, specifically for COVID-19 patients, the cut-off threshold was 26.3 seconds. The ROC curve at that optimal cut-off value is shown in Figure 7. The sensitivity was 0.727, the specificity was 0.700, and the area under the curve was 0.7227.

[0042] These results suggested that the RST threshold serving as the evaluation criterion for whether a patient falls under Moderate II or Severe is around 26 seconds. Therefore, when the RST is less than 20 to 30 seconds, it can be determined that the pneumonia has severely exacerbated to the extent that hospitalization is required.

Industrial Applicability

[0043] The present invention is particularly suitable for monitoring isolated COVID-19 positive individuals, but is not limited thereto. For example, by applying the present invention in nursing care facilities for the elderly and other healthcare facilities and monitoring the RST of the residents, early detection of aspiration pneumonia and the like becomes possible.

Explanation of Signs

[0044] 1 Evaluation system 2 Detection device 21 Sheet sensor 22 Measurement unit 3 Relay terminal 4 Evaluation device 41 Memory unit 42 Acquisition unit 43 Calculation unit 44 Evaluation Unit 5 Browsing Terminal D1 Evaluation Program

Claims

1. An evaluation device for evaluating the severity of pneumonia in a target patient, comprising: an acquisition unit that acquires the respiratory waveform of the target patient; a calculation unit that calculates the RST, which is the reciprocal of the standard deviation of the respiratory frequency, as an index indicating the instability of the respiratory cycle or respiratory frequency from the respiratory waveform; an evaluation unit that evaluates that the severity is high when the calculated RST is less than a threshold value; wherein the threshold value is 26 seconds. The evaluation device.

2. The evaluation device according to claim 1, wherein the pneumonia is pneumonia caused by COVID-19.

3. An evaluation system comprising a detection device that detects a signal including the respiratory waveform of the target patient, and the evaluation device according to claim 1 or 2.

4. The evaluation system according to claim 3, wherein the detection device includes a seat sensor provided under the target patient who is in bed, and the body pressure signal generated by the seat sensor is detected as the signal.

5. An evaluation method for evaluating the severity of pneumonia in a target patient by a computer, comprising: an acquisition step of acquiring the respiratory waveform of the target patient; a calculation step of calculating the RST, which is the reciprocal of the standard deviation of the respiratory frequency, as an index indicating the instability of the respiratory cycle or respiratory frequency from the respiratory waveform; an evaluation step of evaluating that the severity is high when the calculated RST is less than a threshold value; wherein the threshold value is 26 seconds. The evaluation method.

6. An evaluation program for operating a computer as each part of the evaluation device according to claim 1 or 2.

7. A computer-readable recording medium recording the evaluation program according to claim 6. ​

Images