CN-121985910-A - Diagnostic support program and diagnostic support system

Abstract

To analyze a dynamic image of a living body and display the analysis result. [ solution ] A diagnosis assistance program for analyzing an image of a living body and displaying the analysis result, the diagnosis assistance program causing a computer to execute a process of acquiring a plurality of frame images, a process of acquiring a periodic image signal that varies at a fixed period from the plurality of frame images, a process of extracting waveforms of all or a part of the periodic image signal, and a process of collating the extracted waveforms with model waveforms acquired in advance and detecting correlation therebetween. Also included is a process of supplementing the discrete periodic image signal to form a continuous waveform. Further, the correlation is detected based on at least one of a phase, an amplitude, or a frequency of a waveform of all or a part of the periodic image signal.

Inventors

• YOSHIDA NORIHITO

Assignees

• 帕拉梅维亚私人有限公司
• 美迪欧特股份有限公司
• 阿部武彦

Dates

Publication Date

20260505

Application Date

20240411

Priority Date

20230411

Claims (17)

1. 1. A diagnosis assisting program for analyzing an image of a living body and displaying an analysis result, the program being configured to cause a computer to execute: a process for acquiring a plurality of frame images; A process for acquiring a periodic image signal varying at a predetermined period from the plurality of frame images; A process for extracting all or a part of the waveform of the periodic image signal, and And a process for matching the extracted waveform with a pre-acquired model waveform and detecting correlation therebetween.
2. 2. The diagnosis assisting program according to claim 1, further comprising a process for supplementing the periodic image signal, which is discrete, and forming a continuous waveform.
3. 3. The diagnosis assisting program according to claim 1, wherein the correlation is detected based on at least one of a phase, an amplitude, or a frequency of all or a part of the waveform of the periodic image signal.
4. 4. The diagnosis assisting program according to claim 1, configured to cause the computer to execute a process for visualizing the extracted waveform based on the detected correlation.
5. 5. A diagnosis assisting program for analyzing an image of a living body and displaying an analysis result, the program being configured to cause a computer to execute: a process for acquiring a plurality of frame images; A process for acquiring a periodic image signal varying at a predetermined period from the plurality of frame images; A process for extracting all or a portion of the frequency of the periodic image signal; A process for matching the extracted frequencies with pre-acquired model frequencies and detecting correlations between them.
6. 6. The diagnostic aid of claim 5, further comprising a process for identifying a waveform based on phase, amplitude and frequency of all or a portion of the periodic image signal.
7. 7. The diagnostic aid of claim 6, configured to cause the computer to perform a process for visualizing the identified waveforms.
8. 8. A diagnosis assisting program for analyzing an image of a living body and displaying an analysis result, the program being configured to cause a computer to execute: a process for acquiring a plurality of frame images; a process for acquiring a periodic image signal varying at a predetermined period from the plurality of frame images, and A process for creating a model by training an artificial intelligence AI to learn all or a portion of the periodic image signal.
9. 9. The diagnosis assistance program according to claim 8, further comprising a process for detecting a correlation between the model and the periodic image signals acquired from the plurality of frame images.
10. 10. The diagnosis assisting program according to claim 1, further comprising a process for extracting a signal having a maximum amplitude among the extracted waveforms, Wherein the signal with the maximum amplitude is visualized based on the detected correlation.
11. 11. The diagnosis support program according to claim 1 or 5, further comprising a process for applying a logarithmic transformation to the periodic image signals acquired from the plurality of frame images, Wherein each of the processes is performed using the periodic image signal after the logarithmic transformation.
12. 12. The diagnosis support program according to claim 1 or 5, further comprising: A process for applying a fourier transform to the periodic image signal; a process for extracting a signal having said correlation from said signal after said Fourier transform, and A process for applying an inverse fourier transform to the extracted signal.
13. 13. The diagnostic aid of claim 1, further comprising a process for interpolating the waveform using a waveform representing motion of a standard model created by training an artificial intelligence AI to learn images of a standard organ.
14. 14. The diagnosis assisting program according to claim 5, further comprising a process of extracting pixels using a frequency of movement of a standard model created by training an artificial intelligence AI to learn an image of a standard organ, wherein a pixel value of the pixels varies with time according to the frequency of movement of the standard model.
15. 15. A diagnosis assisting program for analyzing an image of a living body and displaying an analysis result, the program being configured to cause a computer to execute: A process for acquiring a video composed of a plurality of frame images; A process for creating a model by training an artificial intelligence AI to learn the video, and A process for predicting a measured value of the living body using the model.
16. 16. A diagnosis assisting program for analyzing an image of a living body and displaying an analysis result, the program being configured to cause a computer to execute: A process for acquiring a video composed of a plurality of frame images; A process for acquiring a periodic image signal varying in a predetermined period from the video; a process for creating a model by training an artificial intelligence AI to learn the periodic image signal, and A process for predicting a phenomenon occurring in the living body using the model.
17. 17. A diagnosis assistance system configured to analyze an image of a living body and display an analysis result, comprising: an input interface configured to acquire a plurality of frame images, and A base module configured to: Acquiring a periodic signal image varying at a predetermined period from the plurality of frame images; Extracting all or a part of the waveform of the periodic image signal, and The extracted waveform is matched with a model waveform acquired in advance, and correlation between them is detected.

Description

Diagnostic support program and diagnostic support system Technical Field The present invention relates to a technique for analyzing an image of a human body and displaying the analysis result. Background When a doctor performs lung diagnosis based on dynamic images of the chest, it is very important to observe time-series dynamic chest images captured while the subject is in a natural respiratory state. Spirometers, RI (radio isotope) examinations, plain X-ray images from which morphological data can be obtained, and CT (computed tomography) scans, which are all well known methods for assessing lung function, are helpful for acquiring physiological data. However, it is not easy to efficiently acquire physiological data and morphological data. In recent years, a method for capturing a dynamic image of a human chest using a semiconductor image sensor such as an FPD (flat panel detector) has been tried, and used for diagnosis. For example, non-patent document 1 ("radiological investigation of basic imaging characteristics of a large-scale image intensifier TV digital chest radiography system: 4 months in 1987; 22:328-335) discloses a technique for generating a differential image representing a difference in signal value between a plurality of frame images constituting a dynamic image, determining a maximum value of each signal value from the differential image, and displaying the maximum value. Further, patent document 1 discloses a technique in which a lung field region is extracted from each of a plurality of frame images indicating a dynamic state of a human chest, the lung field region is divided into a plurality of small regions, and the divided small regions are associated with each other and analyzed among the plurality of frame images. According to this technique, a feature quantity indicating the movement of the divided small region is displayed. Prior art literature Patent literature Patent document 1 Japanese patent publication No. 5874636 Non-patent document Non-patent document 1, "basic imaging characteristics of large image intensifier TV digital chest radiography System", radiological investigation 4 months in 1987, 22:328-335. Disclosure of Invention Technical problem However, as in the technique described in non-patent document 1, displaying only the maximum inter-frame difference value for each pixel of a moving image does not allow a doctor to easily grasp a pathological condition. Further, as in the technique described in patent document 1, displaying only the feature quantity is still insufficient to grasp the pathological condition. In particular, conventionally, respiration and blood pressure are measured after a subject is placed in a resting state, and thus measurement can be performed only in this condition, however, it is desirable to acquire biological information such as respiration and blood pressure in a dynamic condition. That is, it is desirable to grasp the overall dynamic respiratory state and vascular dynamics of a human body as a subject, and display an image indicating actual movement based on the waveform, or frequency, or waveform and frequency of respiration, heart, or blood vessels, or blood flow in the portal area, or based on the trend of change of the image. The present invention has been made in view of such circumstances, and aims to provide a diagnosis assisting program capable of analyzing an image of a human body and displaying the analysis result. More specifically, the present invention aims at calculating values usable for diagnosis by specifying waveforms of new measured data of a dynamic target or by quantifying matching degrees with respect to waveforms and frequencies (Hz) that have been acquired and other mismatch degrees, and at generating images usable for diagnosis by visualizing these values. Technical proposal (1) In order to achieve the above object, the present application has adopted the following means. That is, according to one aspect of the present disclosure, a diagnosis assisting program for analyzing an image of a living body and displaying an analysis result is configured to cause a computer to execute a process for acquiring a plurality of frame images, a process for acquiring a periodic image signal that varies at a predetermined period from the plurality of frame images, a process for extracting all or a part of waveforms of the periodic image signal, and a process for matching the extracted waveforms with a model waveform acquired in advance and detecting correlation therebetween. (2) Optionally, the diagnostic aid according to the present disclosure may further comprise a process for supplementing the discrete periodic image signals and forming a continuous waveform. (3) Optionally, in the diagnosis assisting program according to the present disclosure, the correlation may be detected based on at least one of a phase, an amplitude, or a frequency of all or a part of the waveforms of the periodic image signal. (4) Optionally, the diag