CN115359878A

CN115359878A - ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting system and method

Info

Publication number: CN115359878A
Application number: CN202211049097.6A
Authority: CN
Inventors: 王可欣; 王霄英; 岳新; 王祥鹏
Original assignee: Beijing Smarttree Medical Technology Co Ltd
Current assignee: Beijing Smarttree Medical Technology Co Ltd
Priority date: 2022-08-30
Filing date: 2022-08-30
Publication date: 2022-11-18

Abstract

The invention provides an ICU bedside chest X-ray film automatic contrast diagnosis intelligent report system, which comprises: the patient information acquisition module acquires a current ICU bedside chest X-ray image of a patient, patient medical history information, an existing image and existing report information; the segmentation data acquisition module inputs the current ICU bedside chest X-ray image and the past image into a plurality of deep learning models for image segmentation; the image registration module performs image registration on the current ICU bedside chest X-ray image and the previous image, and outputs all registered images and the registered segmentation data; the contrast analysis module analyzes the change condition of related diseases on the current ICU bedside chest X-ray image; the structured report module automatically outputs final diagnosis data based on the change condition of the related diseases. The invention also discloses an ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting method. The invention can find potential risks, preferentially and effectively cure the patients and improve the working efficiency.

Description

ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting system and method

Technical Field

The invention relates to the field of medical information, in particular to an ICU bedside chest X-ray film automatic comparison diagnosis intelligent reporting system and method.

Background

Chest radiography is almost the most abundant examination in the image diagnosis task of modern general hospitals. Among them, intensive Care Unit (ICU) accounts for a large proportion of critical patients' bedside chest radiographs. In most intensive care units, routine daily chest examination of patients is the standard procedure. When the patient's condition changes, or after certain medical procedures are performed, a bedside chest X-ray examination may be performed multiple times a day. Since the examination time interval of the consecutive bedside chest radiographs is short, the comparison is made with both the latest examination and the earlier consecutive examination in order to find the progressive change. The purpose of the image examination is to detect critical findings in real time, to detect signs that may alter the treatment regimen, and to track image changes. The huge number of examinations and the repeated comparative observation among a plurality of examinations become the heavy diagnosis task for the image diagnostician.

Meanwhile, diagnosis of the bedside chest radiograph is difficult. The bedside chest film is different from the conventional chest film, and has the characteristics of photographic equipment, projection positions and patient conditions. The photographic equipment of chest film beside the bed is a movable X-ray machine, the conventional chest film is a fixed X-ray machine, the exposure capability of the movable X-ray machine is slightly lower, and the penetration capability to soft tissues is slightly worse than that of the fixed X-ray machine, so the image quality of part of patients is poor. The projection position of the bedside chest radiography is front-back position, the conventional chest radiography is back-front position, the front-back position can cause the amplification of the heart image, and the abnormality of mediastinum and the shape of the heart is difficult to judge. The patient of bedside chest film has serious illness, a plurality of medical devices for supporting and monitoring physiological functions are arranged in the body and on the surface of the patient, the patient of the conventional chest film generally has light illness, the medical devices in the body and on the surface of the patient are few, and the detection of important symptoms can be influenced due to more interference of the medical devices on the bedside chest film. Therefore, diagnosis of the bedside chest radiograph is difficult. Because the bedside chest radiograph has great diagnostic significance for critically ill patients, large number of comparison radiographs and poor image quality, the diagnosis of the bedside chest radiograph by an image diagnostician needs higher experience, and the patients have fast change of illness state and require rapid diagnosis. In the prior art, image diagnosticians diagnose X-ray images beside a bed one by one, and due to the large workload and the large diagnosis difficulty, misdiagnosis or treatment delay is easily caused, and the safety of patients cannot be guaranteed.

Disclosure of Invention

In view of the above, the main objective of the present invention is to provide an ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting system and method, which can automatically contrast and diagnose the current ICU bedside chest X-ray film and the past chest X-ray film immediately after the ICU bedside chest X-ray film is photographed, and immediately send out a warning prompt to related medical staff after a potential risk is found, so as to preferentially and effectively treat the patients, thereby solving the problems of misdiagnosis or delayed treatment caused by large diagnosis difficulty and workload in the prior art.

In order to achieve the purpose, the technical scheme of the invention is realized as follows:

in one aspect, the invention provides an intelligent reporting system for ICU bedside chest X-ray film automatic contrast diagnosis, comprising: the system comprises a patient information acquisition module, a segmentation data acquisition module, an image registration module, a contrast analysis module and a structured report module, wherein the patient information acquisition module is respectively connected with the segmentation data acquisition module, the image registration module, the contrast analysis module and the structured report module and is used for acquiring a current ICU bedside chest X-ray image of a patient, patient medical history information, a previous image and previous report information; wherein the previous image is the last ICU bedside chest X-ray image of the patient and all ICU bedside chest X-ray images before the last ICU bedside X-ray image; the segmentation data acquisition module is respectively connected with the patient information acquisition module, the image registration module and the structured report module and is used for inputting the current ICU bedside chest X-ray image and the previous image into a plurality of deep learning models for image segmentation to respectively obtain corresponding segmentation data; the types of segmentation data comprise segmentation data of a breast structure and segmentation data of a breast lesion; the image registration module is respectively connected with the patient information acquisition module, the segmentation data acquisition module, the contrast analysis module and the structured report module and is used for carrying out image registration on the current ICU bedside chest X-ray image and the previous image based on the segmentation data and outputting all registered images and the registered segmentation data; the contrast analysis module is respectively connected with the patient information acquisition module, the image registration module and the structured report module and is used for analyzing the change condition of related diseases on the chest X-ray image beside the current ICU bed based on all registered images, the registered segmentation data and the past report information; and the structured report module is connected with the patient information acquisition module, the segmentation data acquisition module, the image registration module and the comparative analysis module and is used for automatically outputting final diagnosis data based on the change condition of the related diseases.

Preferably, the segmentation data for the thoracic structure is: all chest imaging areas, lung field areas and mediastinum areas of the current ICU bedside chest X-ray image and the past image; the segmentation data of the breast lesion are: pleural cavity lesion areas and abnormal high-density lesion areas of the lung of the current ICU bedside chest X-ray image and the latest ICU bedside chest X-ray image; wherein, the pleural cavity pathological change region includes pneumothorax region, pleural effusion region.

Preferably, all images after registration are: the registered current ICU bedside chest X-ray image, the registered latest ICU bedside chest X-ray image and the registered latest before ICU bedside chest X-ray image are obtained; the registered segmentation data is: the chest X-ray image comprises a mediastinum region in a current ICU bedside chest X-ray image after registration, a pleural cavity lesion region in the current ICU bedside chest X-ray image after registration, a lung abnormal high-density lesion region in the current ICU bedside chest X-ray image after registration, a mediastinum region in a latest ICU bedside chest X-ray image after registration, a pleural cavity lesion region in a latest ICU bedside chest X-ray image after registration, a lung abnormal high-density lesion region in the latest ICU bedside chest X-ray image after registration and a mediastinum region in all ICU bedside chest X-ray images before the latest ICU bedside after registration.

Preferably, the comparative analysis module further comprises: the mediastinum form analysis unit is used for inquiring the evaluation condition of the mediastinum in the last report in the previous report information, judging the form change of the mediastinum on the chest X-ray image beside the ICU bed according to the evaluation condition of the mediastinum, and outputting the form change condition of the mediastinum; wherein the evaluation condition of the mediastinum is normal or abnormal; the change of the form of the mediastinum comprises the following steps: the diaphragm shadow is obviously enlarged, the diaphragm shadow is reduced, and the diaphragm shadow is not changed.

Preferably, the comparative analysis module further comprises: the pleural cavity lesion analysis unit is used for inquiring the evaluation condition of the pleural cavity in the last report in the previous report information, judging the change of the pleural cavity lesion on the current ICU bedside chest X-ray image according to the pleural cavity evaluation condition and outputting the change condition of the pleural cavity lesion; wherein the pleural cavity is evaluated as normal or abnormal; changes in pleural cavity lesions include: newly increased pleural cavity pathological changes, normal pleural cavity, obviously increased pleural cavity pathological changes, reduced pleural cavity pathological changes and unchanged pleural cavity pathological changes.

Preferably, the comparative analysis module further comprises: the lung lesion analysis unit is used for inquiring the evaluation condition of the lung lesion in the last report in the previous report information, judging the change of the lung lesion on the chest X-ray image beside the ICU bed at present according to the evaluation condition of the lung lesion and outputting the change condition of the lung lesion; wherein the evaluation condition of the lung lesion is normal or abnormal; changes in lung lesions include: newly added lung lesion, normal lung, obvious increase of lung lesion, reduction of lung lesion and no change of lung lesion.

Preferably, the method for judging the change of the diaphragm morphology on the chest X-ray image beside the ICU bed comprises the following steps: when the evaluation condition of the mediastinum is normal, comparing a mediastinum area in the registered last ICU bedside chest X-ray image with mediastinum areas in all images of ICU bedside chest X-rays before the last ICU bedside after registration by a mediastinum form analysis unit, searching a previous minimum mediastinum area, comparing and analyzing the previous minimum mediastinum area with the mediastinum area in the registered current ICU bedside chest X-ray image, and judging that the mediastinum shadow is obviously increased when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than a first preset threshold value compared with the previous minimum mediastinum area; when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than a second preset threshold value in comparison with the existing minimum mediastinum area, judging that the mediastinum shadow is increased; when the mediastinum area in the registered current ICU bedside chest X-ray image is increased by less than or equal to a second preset threshold value compared with the existing minimum mediastinum area, judging that the mediastinum image has no change; when the evaluation condition of the mediastinum is abnormal, a mediastinum form analysis unit compares and analyzes a mediastinum area in the registered latest ICU bedside chest X-ray image and a mediastinum area in the registered current ICU bedside chest X-ray image, and when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than the mediastinum area in the registered latest ICU bedside chest X-ray image by more than a first preset threshold value, the mediastinum shadow is judged to be remarkably increased; when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than the mediastinum area in the registered latest ICU bedside chest X-ray image by more than a second preset threshold value, judging that the mediastinum shadow is increased; when the mediastinum area in the current ICU bedside chest X-ray image after registration is smaller than the mediastinum area in the latest ICU bedside chest X-ray image after registration by a second preset threshold value, judging that the mediastinum shadow is reduced; and when the change of the mediastinum area in the current ICU bedside chest X-ray image after the registration is less than or equal to a second preset threshold value than the change of the mediastinum area in the last ICU bedside chest X-ray image after the registration, judging that the mediastinum shadow has no change.

Preferably, the judging of the change of the pleural cavity lesion on the current ICU bedside chest X-ray image comprises the following steps: when the pleural cavity evaluation condition is normal, the pleural cavity lesion analysis unit detects whether pleural cavity lesions exist in the registered current ICU bedside chest X-ray image, and if yes, new pleural cavity lesions are judged; if not, judging that the pleural cavity is normal; when the evaluation condition of the pleural cavity is abnormal, a pleural cavity lesion analysis unit compares and analyzes a pleural cavity lesion area in the registered latest ICU bedside chest X-ray image with a pleural cavity lesion area in the registered current ICU bedside chest X-ray image, and when the pleural cavity lesion area in the registered current ICU bedside chest X-ray image is larger than a pleural cavity lesion area in the registered latest ICU bedside chest X-ray image by more than a third preset threshold value, the pleural cavity lesion is judged to be remarkably increased; when the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is larger than the pleural cavity lesion area in the last ICU bedside chest X-ray image after registration by a fourth preset threshold value, judging that the pleural cavity lesion area is increased; when the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is smaller than the pleural cavity lesion area in the last ICU bedside chest X-ray image after registration by a fourth preset threshold value, judging that the pleural cavity lesion area is reduced; and when the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is changed by less than or equal to a fourth preset threshold value compared with the pleural cavity lesion area in the last ICU bedside chest X-ray image after registration, judging that the pleural cavity lesion is not changed.

Preferably, the determining the change of the lung lesion on the current ICU bedside chest X-ray image comprises: when the evaluation condition of the lung lesion is normal, the lung lesion analysis unit detects whether the registered current chest X-ray image beside the ICU bed has the lung lesion, and if so, the newly added lung lesion is judged; if not, judging that the lung is normal; when the evaluation condition of the lung lesion is abnormal, the lung lesion analysis unit compares and analyzes the abnormal high-density lesion area of the lung in the registered latest ICU bedside chest X-ray image with the abnormal high-density lesion area of the lung in the registered current ICU bedside chest X-ray image, and when the abnormal high-density lesion area of the lung in the registered current ICU bedside chest X-ray image is increased by more than a fifth preset threshold value than the abnormal high-density lesion area of the lung in the registered latest ICU bedside chest X-ray image, the lung lesion is judged to be remarkably increased; when the lung abnormal high-density lesion area of the pleural cavity lesion area in the current ICU bedside chest X-ray image after the registration is larger than a sixth preset threshold value than the increase of the lung abnormal high-density lesion area in the last ICU bedside chest X-ray image after the registration, the lung lesion area is judged to be increased; when the lung abnormal high-density lesion area in the registered current ICU bedside chest X-ray image is smaller than a sixth preset threshold value compared with the lung abnormal high-density lesion area in the registered latest ICU bedside chest X-ray image, the lung abnormal high-density lesion area is judged to be reduced; and when the change of the abnormal high-density lesion area of the lung in the pleural cavity lesion area in the current ICU bedside chest X-ray image after the registration is less than or equal to a sixth preset threshold value than the change of the abnormal high-density lesion area of the lung in the last ICU bedside chest X-ray image after the registration, judging that the lung lesion is not changed.

On the other hand, the invention also provides an ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting method, which comprises the following steps: acquiring a chest X-ray image of a patient beside a current ICU (intensive care unit) bed, patient medical history information, an existing image and past report information; wherein the previous image is the last ICU bedside chest X-ray image of the patient and all ICU bedside chest X-ray images before the last ICU bedside X-ray image; inputting a chest X-ray image and a past image beside a current ICU (intensive care unit) bed into a plurality of deep learning models for image segmentation to respectively obtain corresponding segmentation data; the types of segmentation data comprise segmentation data of a breast structure and segmentation data of a breast lesion; based on the segmentation data, carrying out image registration on the current ICU bedside chest X-ray image and the previous image, and outputting all registered images and the registered segmentation data; analyzing the change condition of related diseases on the chest X-ray image beside the current ICU bed based on all the registered images, the registered segmentation data and the previous report information; and automatically outputting final diagnosis data based on the change condition of the related diseases.

Preferably, the segmentation data for the thoracic structure is: all chest imaging areas, lung field areas and mediastinum areas of the current ICU bedside chest X-ray image and the past image; the segmentation data of the breast lesion are: pleural cavity lesion areas and abnormal high-density lung lesion areas of the current ICU bedside chest X-ray image and the latest ICU bedside chest X-ray image; wherein, the pleural cavity pathological change region includes pneumothorax region, pleural effusion region.

The invention has the technical effects that:

the system of the invention applies an AI model and a program based on rules to the automatic comparative analysis of the current ICU bedside chest X-ray film and the existing bedside chest X-ray film to obtain an intelligent diagnosis report system for bedside chest X-ray examination, accesses the intelligent diagnosis report system into a PACS/RIS to automatically obtain patient medical history information, existing images and existing report information, can segment the images after the image acquisition is finished, can automatically output the change condition of the mediastinum form, the change condition of pleural cavity lesion and the change condition of lung lesion by using the segmented data and the registered images and the registered segmented data regions of the existing images and the existing report information, automatically finishes evaluation, and automatically transmits the result into a structured report to improve the working efficiency of a doctor. More importantly, when the intelligent system finds that a potential risk exists, the intelligent system immediately sends out a warning prompt to relevant medical personnel so as to help the medical personnel to immediately pay attention to the abnormal conditions of patients, and the patients are preferably treated effectively, so that the safety of the patients is guaranteed.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention and do not constitute a limitation of the invention. In the drawings:

FIG. 1 is a schematic structural diagram of an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to an embodiment of the invention;

FIG. 2 is a schematic diagram of an ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting system according to a second embodiment of the invention;

FIG. 3 is a schematic structural diagram of an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 4 is a schematic diagram of a current chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 5 is a schematic diagram of a chest imaging region of a current chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 6 is a schematic diagram of a registered current chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 7 is a chest imaging region schematic diagram of a registered current chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 8 is a schematic diagram of a pleural effusion region of a registered current chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 9 is a schematic diagram of a chest X-ray image before registration in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

fig. 10 shows a schematic diagram of a chest imaging region of a past chest X-ray image before registration in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the present invention;

FIG. 11 is a diagram showing a registered past chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 12 is a schematic diagram of a chest imaging region of a registered past chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 13 is a schematic diagram of a pleural effusion region of a registered past chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 14 is a diagram of an ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting system according to a fourth embodiment of the present invention;

FIG. 15 shows a flowchart of an ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting method according to a fifth embodiment of the invention.

Detailed Description

The present invention will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Example one

FIG. 1 is a schematic structural diagram of an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to an embodiment of the invention; as shown in fig. 1, the system includes: a patient information acquisition module 10, a segmentation data acquisition module 20, an image registration module 30, a contrast analysis module 40, and a structured reporting module 50, wherein,

before acquiring the patient information, firstly, the image property of the patient is identified, whether the patient is a bedside chest X-ray examination is judged, and a qualitative judgment result (whether the patient is the bedside chest X-ray examination or a non-bedside chest X-ray examination) is output to a corresponding control of structured report 'technical evaluation'. If the examination item is judged to be ICU bedside chest X-ray examination, outputting an X-ray image of the patient for a subsequent AI diagnosis process; if the examination item is judged to be not ICU bedside chest X-ray examination, an AI diagnosis process is stopped, prompt information is sent, and relevant personnel take charge of processing and record the prompt information in a database.

The patient information acquisition module 10 is respectively connected with the segmentation data acquisition module 20, the image registration module 30, the contrast analysis module 40 and the structured report module 50, and is used for acquiring a current ICU bedside chest X-ray image of the patient, patient medical history information, a past image and past report information; wherein the previous image is the last ICU bedside chest X-ray image of the patient and all ICU bedside chest X-ray images before the last ICU bedside X-ray image;

the current major disease diagnosis and clinical manifestations of patients are acquired in RIS and electronic cases. Qualitative judgments data are output-diagnosis of the current major disease from RIS and electronic case patients, qualitative judgments-clinical presentations obtained from RIS and electronic cases. The above is achieved by an AI model and program.

The patient's current primary disease diagnosis obtained from the RIS and electronic cases is returned to the corresponding controls of the structured report "clinical assessment" for reference by the diagnostician;

clinical presentations obtained from the RIS and electronic cases are returned to the structured report "clinical presentation" controls for reference by the diagnostician.

The method comprises the steps of obtaining previous images and previous report information, realizing by a program, obtaining images of the same type of examination after the same patient enters an ICU from a PACS/RIS database, obtaining a latest ICU bedside chest X-ray examination report from an RIS system, and extracting positive data in the reports, wherein the method comprises the following steps: mediastinal abnormalities, pleural cavity abnormalities, pulmonary abnormalities.

All images of bedside chest X-ray examination after the same patient enters ICU in the PACS and a latest ICU bedside chest X-ray examination report in a structured report system; the image of the last ICU bedside chest X-ray examination, all images of the ICU bedside chest X-ray examination before the last time, the mediastinum assessment in the last report, the pleural cavity assessment in the last report, and the lung assessment in the last report are output.

The image of the last ICU bedside chest X-ray examination and all the images of the ICU bedside chest X-ray examination before the last ICU bedside are used for image comparison;

mediastinum assessment in the last report, pleural cavity assessment in the last report, and lung assessment in the last report are used for subsequent image comparison and report generation.

The segmentation data acquisition module 20 is respectively connected with the patient information acquisition module 10, the image registration module 30 and the structured report module 50, and is used for inputting the current ICU bedside chest X-ray image and the previous image into a plurality of deep learning models for image segmentation to respectively obtain corresponding segmentation data; the types of segmentation data comprise segmentation data of a breast structure and segmentation data of a breast lesion;

the segmentation data for the thoracic structure are: all chest imaging areas, lung field areas and mediastinum areas of the current ICU bedside chest X-ray image and the past image; the segmentation data of the breast lesion are: pleural cavity lesion areas and abnormal high-density lesion areas of the lung of the current ICU bedside chest X-ray image and the latest ICU bedside chest X-ray image; wherein, the pleural cavity pathological change region includes pneumothorax region, pleural effusion region.

The segmentation data of the thoracic structure are a thoracic imaging region, a lung field and a mediastinum region;

the breast focus segmentation data comprises a pleural cavity lesion area and a lung abnormal high-density focus area.

The breast imaging region is segmented in all images of the same PACS patient taken into an ICU bedside chest X-ray exam, and the breast imaging region data is output for subsequent image registration and segmentation of the breast structure.

And (3) segmenting lung field and mediastinum regions in all images of the same PACS patient subjected to ICU bedside chest X-ray examination, and outputting lung field region data for subsequent intra-pulmonary lesion segmentation and comparison. The mediastinal region was used for subsequent mediastinal morphology comparison.

Segmentation of pleural cavity lesion area: inputting a current ICU bedside chest X-ray image, an image of a latest ICU bedside chest X-ray examination and a chest imaging area; outputting areas of pneumothorax and pleural effusion, and performing qualitative judgment on data-pneumothorax, qualitative judgment on data-pleural effusion and qualitative judgment on data-hydropneumothorax;

when pneumothorax exists, the results are returned to the control of the structured report "pneumothorax". The pneumothorax region is used for subsequent image contrast.

When pleural effusion is present, the results are returned to the control for the structured report "pleural effusion". The pleural effusion region is used for subsequent image contrast.

In the presence of pneumothorax, the results are returned to the control of the structured report "pneumothorax". Areas of pneumothorax and pleural effusion are used for subsequent image contrast.

Segmentation of high-density lesion areas of lung abnormalities: inputting a current ICU bedside chest X-ray image, an image of a latest ICU bedside chest X-ray examination and chest imaging, outputting a lung abnormal high-density lesion area, and qualitatively judging the lung abnormal high-density lesion; when there is an abnormally high density of lesions in the lung, the results are returned to the control for the structured report "lung lesions". The abnormally high density lesion area of the lung is used for subsequent image contrast.

The image registration module 30 is respectively connected with the patient information acquisition module 10, the segmentation data acquisition module 20, the comparative analysis module 40 and the structured report module 50, and is used for carrying out image registration on the current ICU bedside chest X-ray image and the previous image based on the segmentation data, and outputting all registered images and the registered segmentation data;

the total images after registration are: the registered current ICU bedside chest X-ray image, the registered latest ICU bedside chest X-ray image and the registered latest before ICU bedside chest X-ray image are obtained;

the registered segmentation data is: mediastinum area in current ICU bedside chest X-ray image after registration, pleural cavity lesion area in current ICU bedside chest X-ray image after registration, lung abnormal high-density focus area in current ICU bedside chest X-ray image after registration, mediastinum area in latest ICU bedside chest X-ray image after registration, pleural cavity lesion area in latest ICU bedside chest X-ray image after registration, lung abnormal high-density focus area in latest ICU bedside chest X-ray image after registration, and mediastinum area in all ICU bedside chest X-ray images before latest registration

Specifically, the method comprises the following steps: inputting a current ICU bedside chest X-ray image, an image of a latest ICU bedside chest X-ray examination, all images of a latest ICU bedside chest X-ray examination and a chest imaging area; outputting the registered current ICU bedside chest X-ray image, the registered image of the latest ICU bedside chest X-ray examination, the registered all images of the ICU bedside chest X-ray examination before the latest ICU bedside, the mediastinum in the registered current ICU bedside chest X-ray image, the pneumothorax in the registered current ICU bedside chest X-ray image, the pleural effusion in the registered current ICU bedside chest X-ray image, the abnormal high-density lesion in the registered current ICU bedside chest X-ray image, the mediastinum in the registered latest ICU bedside chest X-ray image, the pneumothorax in the registered latest ICU bedside chest X-ray image, the pleural effusion in the registered latest ICU bedside chest X-ray image, the abnormal high-density lesion in the registered latest ICU bedside chest X-ray image and the post-registration chest region in the registered all images of the latest ICU bedside X-ray examination, and the chest data after registration are used for breast segmentation.

The contrast analysis module 40 is respectively connected with the patient information acquisition module 10, the image registration module 30 and the structured report module 50, and is used for analyzing the change condition of related diseases on the chest X-ray image beside the current ICU bed based on all registered images, the registered segmentation data and the past report information;

and the structured reporting module 50 is connected with the patient information acquisition module 10, the segmentation data acquisition module 20, the image registration module 30 and the comparative analysis module 40 and is used for automatically outputting final diagnosis data based on the change condition of the related diseases.

The structured report module integrates the findings of all functional modules to derive an overall diagnostic impression. Based on rules built into the structured report, the final diagnostic data is automatically obtained and returned to the diagnostic impression of the structured report.

And if the critical value exists in the diagnosis impression, sending prompt information, processing by related personnel and recording in a database.

All data, all images are stored in a structured report database.

The embodiment of the invention applies an AI model and a program based on rules to the automatic comparative analysis of the current ICU bedside chest X-ray film and the existing bedside chest X-ray film to obtain an intelligent diagnosis report system for bedside chest X-ray examination, accesses the intelligent diagnosis report system into a PACS/RIS to automatically obtain patient medical history information, existing images and existing report information, can segment the images after the image acquisition is finished, can automatically output the change condition of the mediastinum form, the change condition of pleural cavity lesion and the change condition of lung lesion by using the segmented data and the existing images and the existing report information through the registered images and the registered segmented data area, automatically finishes evaluation, and automatically transmits the result into a structured report to improve the working efficiency of a doctor. More importantly, when the intelligent system finds that the potential risk exists, the intelligent system immediately sends out warning prompts to relevant medical care personnel so as to help the medical care personnel to pay attention to the abnormal conditions of the patients immediately, and the patients are treated and treated effectively in a preferential manner, so that the safety of the patients is guaranteed.

Example two

FIG. 2 is a schematic structural diagram of an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a second embodiment of the invention; as shown in fig. 2, the comparative analysis module 40 further includes: the mediastinum form analysis unit 402 is used for inquiring the evaluation condition of the mediastinum in the last report in the previous report information, judging the change of the mediastinum form on the chest X-ray image beside the current ICU bed according to the evaluation condition of the mediastinum, and outputting the change condition of the mediastinum form; wherein the evaluation condition of the mediastinum is normal or abnormal; the diaphragm morphological change condition comprises: the image of the mediastinum is obviously enlarged, the image of the mediastinum is reduced, and the image of the mediastinum is not changed.

Specifically, the method comprises the following steps: and comparing the mediastinum area in the current ICU bedside chest X-ray image after registration with the mediastinum area in the image of the latest ICU bedside chest X-ray examination after registration and the mediastinum areas in all images of the ICU bedside chest X-ray examination before the latest ICU bedside after registration, and judging whether obvious mediastinum form change exists or not.

Inputting a current ICU bedside chest X-ray image after registration, an image of a latest ICU bedside chest X-ray examination after registration, all images of an ICU bedside chest X-ray examination before the latest ICU bedside after registration, a mediastinum region in the current ICU bedside chest X-ray image after registration, mediastinum in the latest ICU bedside chest X-ray image after registration, mediastinum regions in all images of an ICU bedside chest X-ray examination before the latest ICU bedside after registration, and mediastinum evaluation in a latest report; outputting the prior minimum mediastinum area, qualitative judgment-mediastinum form change, qualitative judgment-mediastinum is obviously increased, and prompt information-mediastinum shadow is obviously increased.

And when the diaphragm form changes, returning the result to a control of a structured report of the diaphragm form change, sending prompt information when the diaphragm is remarkably increased, and recording the prompt information in a database by related personnel for processing.

Analysis of diaphragmatic morphological changes:

when the evaluation condition of mediastinum is normal, comparing mediastinum areas in the registered latest ICU bedside chest X-ray image with mediastinum areas in all images of the latest ICU bedside chest X-ray image before registration, searching a minimum previous mediastinum area, comparing and analyzing the minimum previous mediastinum area with the mediastinum areas in the registered current ICU bedside chest X-ray image, and judging that the mediastinum image is remarkably increased when the mediastinum area in the registered current ICU bedside chest X-ray image is increased by more than a first preset threshold (for example, 20%) compared with the minimum previous mediastinum area; when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than a second preset threshold (for example, 10%) than the existing minimum mediastinum area, judging that the mediastinum shadow is increased; when the mediastinum area in the registered current ICU bedside chest X-ray image is increased by less than or equal to a second preset threshold value compared with the existing minimum mediastinum area, judging that the mediastinum image has no change;

when the evaluation condition of the mediastinum is abnormal, the mediastinum form analysis unit compares and analyzes a mediastinum area in the registered latest ICU bedside chest X-ray image with a mediastinum area in a registered current ICU bedside chest X-ray image, and when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than the mediastinum area in the registered latest ICU bedside chest X-ray image by more than a first preset threshold value, the mediastinum shadow is judged to be obviously increased; when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than the mediastinum area in the registered latest ICU bedside chest X-ray image by more than a second preset threshold value, judging that the mediastinum shadow is increased; when the mediastinum area in the registered current ICU bedside chest X-ray image is smaller than the mediastinum area in the registered latest ICU bedside chest X-ray image by less than a second preset threshold, judging that the mediastinum shadow is reduced; and when the change of the mediastinum area in the current ICU bedside chest X-ray image after the registration is less than or equal to a second preset threshold value than the change of the mediastinum area in the last ICU bedside chest X-ray image after the registration, judging that the mediastinum shadow has no change.

EXAMPLE III

FIG. 3 is a schematic structural diagram of an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention; as shown in fig. 3, the comparative analysis module 40 further includes: the pleural cavity lesion analysis unit 404 is used for inquiring the evaluation condition of the pleural cavity in the latest report in the previous report information, judging the change of the pleural cavity lesion on the chest X-ray image beside the current ICU bed according to the pleural cavity evaluation condition, and outputting the change condition of the pleural cavity lesion; wherein the pleural cavity is evaluated as normal or abnormal; changes in pleural cavity lesions include: newly increased pleural cavity pathological changes, normal pleural cavity, obviously increased pleural cavity pathological changes, reduced pleural cavity pathological changes and unchanged pleural cavity pathological changes.

Specifically, the method comprises the following steps: and comparing the areas of the pleural cavity lesion in the current ICU bedside chest X-ray image after registration with the areas of the pleural cavity lesion in the latest ICU bedside chest X-ray checked image after registration, and judging whether the significant pleural cavity lesion change exists or not. Inputting a current ICU bedside chest X-ray image after registration, an image of a latest ICU bedside chest X-ray examination after registration, areas of pneumothorax and pleural effusion in the current ICU bedside chest X-ray image after registration, areas of pneumothorax and pleural effusion in the image of the latest ICU bedside chest X-ray examination after registration, and an evaluation of a pleural cavity area in a latest report; outputting qualitative judgment, namely changes of pleural cavity lesions, qualitative judgment, remarkable increase of pleural cavity lesions, and prompt information, namely remarkable increase of pleural cavity lesions.

When there is a pleural cavity morphology change, the result is returned to the control of the structured report "pleural cavity morphology change". When the pleural cavity lesion is remarkably increased, prompt information is sent, and the prompt information is processed by related personnel and recorded in a database.

Analysis of pleural cavity lesions:

judge the change of pleural cavity pathological change on current ICU bedside chest X-ray image, include: when the pleural cavity evaluation condition is normal, the pleural cavity lesion analysis unit detects whether pleural cavity lesion exists in the registered current ICU bedside chest X-ray image, and if yes, the newly added pleural cavity lesion is judged; if not, judging that the pleural cavity is normal;

when the evaluation condition of the pleural cavity is abnormal, the pleural cavity lesion analysis unit compares a pleural cavity lesion area in the registered latest ICU bedside chest X-ray image with a pleural cavity lesion area in the registered current ICU bedside chest X-ray image, and when the pleural cavity lesion area in the registered current ICU bedside chest X-ray image is larger than a pleural cavity lesion area in the registered latest ICU bedside chest X-ray image by more than a third preset threshold (for example, 20%), the pleural cavity lesion is judged to be remarkably increased; when the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is larger than the pleural cavity lesion area in the last ICU bedside chest X-ray image after registration by a fourth preset threshold (for example, 10%), judging that the pleural cavity lesion area is increased; when the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is smaller than a fourth preset threshold value compared with the pleural cavity lesion area in the latest ICU bedside chest X-ray image after registration, the pleural cavity lesion area is judged to be reduced; and when the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is changed by less than or equal to a fourth preset threshold value compared with the pleural cavity lesion area in the last ICU bedside chest X-ray image after registration, judging that the pleural cavity lesion is not changed.

The following is an example:

FIG. 5 is a schematic diagram of a chest imaging area of a current chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 9 is a diagram showing an image of a previous chest X-ray before registration in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 10 is a schematic diagram of a chest imaging region of a past chest X-ray image before registration in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

FIG. 11 shows a schematic diagram of registered past chest X-ray images in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention;

fig. 12 shows a schematic diagram of a chest imaging region of a registered past chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the present invention;

FIG. 13 is a schematic diagram of a pleural effusion region of a registered past chest X-ray image in an ICU bedside chest X-ray automatic contrast diagnosis intelligent reporting system according to a third embodiment of the invention; analysis of effusion lesions, as shown in FIGS. 4-13

Acquiring medical history information to obtain the current main disease diagnosis: pleural effusion

Acquiring previous report information, wherein the positive data is as follows: pleural effusion

Followed by image registration

And comparing the pleural effusion region in the registered previous chest X-ray image with the pleural effusion region in the registered current chest X-ray image, wherein the current pleural effusion region is increased by more than 20% compared with the previous pleural effusion region, and the pleural cavity lesion is considered to be remarkably increased.

Example four

FIG. 14 is a diagram of an ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting system according to a fourth embodiment of the present invention; as shown in fig. 14, the comparative analysis module 40 further includes: the lung lesion analysis unit 406 is configured to query an evaluation condition of a lung lesion in a latest report in the previous report information, judge a change of the lung lesion in a chest X-ray image of a bedside of the ICU at present according to the evaluation condition of the lung lesion, and output a change condition of the lung lesion; wherein the evaluation condition of the lung lesion is normal or abnormal; changes in lung lesions include: newly added lung lesion, normal lung, obvious increase of lung lesion, reduction of lung lesion and no change of lung lesion.

Specifically, the method comprises the following steps: and comparing the area of the lung lesion in the current ICU bedside chest X-ray image after registration with the area of the lung lesion in the image of the latest ICU bedside chest X-ray examination after registration, and judging whether obvious change of the lung lesion exists or not. Inputting a current ICU bedside chest X-ray image after registration, an image of a latest ICU bedside chest X-ray examination after registration, lung lesions in the current ICU bedside chest X-ray image after registration, lung lesion areas in the image of the latest ICU bedside chest X-ray examination after registration and lung lesion evaluation in a latest report; outputting qualitative judgment, namely change of lung lesion, qualitative judgment, significant increase of lung lesion and prompt information, namely significant increase of lung lesion;

when there is a lung lesion morphological change, the result is returned to the control of the structured report "lung lesion morphological change". When the lung pathological changes are remarkably increased, prompt messages are sent, and the prompt messages are processed by related personnel and recorded in a database.

Analysis of changes in lung lesions:

the method for judging the change of lung lesion on the current ICU bedside chest X-ray image comprises the following steps: when the evaluation condition of the lung lesion is normal, the lung lesion analysis unit detects whether the registered current ICU bedside chest X-ray image has the lung lesion, and if so, the registered current ICU bedside chest X-ray image is judged to have the newly added lung lesion; if not, judging that the lung is normal; when the evaluation condition of the lung lesion is abnormal, the lung lesion analysis unit compares and analyzes the abnormal high-density lesion region of the lung in the registered latest ICU bedside chest X-ray image with the abnormal high-density lesion region of the lung in the registered current ICU bedside chest X-ray image, and when the abnormal high-density lesion region of the lung in the registered current ICU bedside chest X-ray image is increased by more than a fifth preset threshold (for example, 20%) than the abnormal high-density lesion region of the lung in the registered latest ICU bedside chest X-ray image, the lung lesion is judged to be remarkably increased; when the lung abnormal high-density lesion area of the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is larger than the lung abnormal high-density lesion area in the last ICU bedside chest X-ray image after registration by more than a sixth preset threshold value (10%), judging that the lung lesion area is increased; when the lung abnormal high-density lesion area in the registered current ICU bedside chest X-ray image is smaller than a sixth preset threshold value compared with the lung abnormal high-density lesion area in the registered latest ICU bedside chest X-ray image, the lung abnormal high-density lesion area is judged to be reduced; and when the lung abnormal high-density lesion area in the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is less than or equal to a sixth preset threshold value compared with the change of the lung abnormal high-density lesion area in the last ICU bedside chest X-ray image after registration, judging that the lung lesion is not changed.

EXAMPLE five

FIG. 15 is a flow chart of an ICU bedside chest radiography automatic contrast diagnostic intelligence reporting method according to a fifth embodiment of the invention; as shown in fig. 15, the method includes the steps of:

before acquiring the patient information, firstly, the image property of the patient is identified, whether the patient is a bedside chest X-ray examination is judged, and a qualitative judgment result (whether the patient is the bedside chest X-ray examination or a non-bedside chest X-ray examination) is output to a corresponding control of a structured report 'technical evaluation'. If the examination item is judged to be ICU bedside chest X-ray examination, outputting an X-ray image of the patient for a subsequent AI diagnosis process; if the examination item is judged to be not ICU bedside chest X-ray examination, an AI diagnosis process is stopped, prompt information is sent, and relevant personnel take charge of processing and record the prompt information in a database.

Step S501, obtaining a chest X-ray image of a patient at the bedside of the ICU, patient medical history information, an existing image and existing report information; wherein the previous image is the last ICU bedside chest X-ray image of the patient and all ICU bedside chest X-ray images before the last ICU bedside X-ray image;

The patient's current primary disease diagnosis obtained from the RIS and electronic cases is returned to the corresponding control of the structured report "clinical assessment" for reference by the diagnostician;

Step S502, inputting a plurality of deep learning models into a chest X-ray image and a past image beside a current ICU bed for image segmentation to respectively obtain corresponding segmentation data; the types of segmentation data comprise segmentation data of a breast structure and segmentation data of a breast lesion;

the segmentation data for the thoracic structure are: all chest imaging areas, lung field areas and mediastinum areas of the current ICU bedside chest X-ray image and the previous image; the segmentation data of the breast lesion are: pleural cavity lesion areas and abnormal high-density lesion areas of the lung of the current ICU bedside chest X-ray image and the latest ICU bedside chest X-ray image; wherein, the pleural cavity pathological change region includes pneumothorax region, pleural effusion region.

The segmentation data of the thoracic structure are a thoracic imaging region, a lung field and a mediastinal region;

And (3) segmenting lung field and mediastinum regions in all images of the same PACS patient subjected to ICU bedside chest X-ray examination, and outputting lung field region data for subsequent intra-pulmonary lesion segmentation and comparison. The mediastinal region was used for subsequent mediastinal morphology comparisons.

Segmentation of abnormal high-density lesion areas of the lung: inputting a current ICU bedside chest X-ray image, an image of a latest ICU bedside chest X-ray examination and chest imaging, outputting a lung abnormal high-density lesion area, and performing qualitative judgment on the lung abnormal high-density lesion; when there is a lung aberrant high density lesion, the result is returned to the control of the structured report "lung lesion". The abnormally high density lesion areas of the lungs are used for subsequent image contrast.

Step S503, based on the segmentation data, carrying out image registration on the current ICU bedside chest X-ray image and the previous image, and outputting all registered images and the registered segmentation data;

all images after registration are: the registered current ICU bedside chest X-ray image, the registered latest ICU bedside chest X-ray image and the registered latest before ICU bedside chest X-ray image are obtained;

the registered segmentation data is: the chest X-ray image comprises a mediastinum region in a current ICU bedside chest X-ray image after registration, a pleural cavity lesion region in the current ICU bedside chest X-ray image after registration, a lung abnormal high-density lesion region in the current ICU bedside chest X-ray image after registration, a mediastinum region in a latest ICU bedside chest X-ray image after registration, a pleural cavity lesion region in a latest ICU bedside chest X-ray image after registration, a lung abnormal high-density lesion region in the latest ICU bedside chest X-ray image after registration and a mediastinum region in all ICU bedside chest X-ray images before the latest ICU bedside after registration.

Specifically, the method comprises the following steps: inputting a current ICU bedside chest X-ray image, an image of a latest ICU bedside chest X-ray examination, all images of a latest ICU bedside chest X-ray examination and a chest imaging area; outputting the registered current ICU bedside chest X-ray image, the registered image of the latest ICU bedside chest X-ray examination, the registered all images of the ICU bedside chest X-ray examination before the latest ICU bedside, mediastinum in the registered current ICU bedside chest X-ray image, pneumothorax in the registered current ICU bedside chest X-ray image, pleural effusion in the registered current ICU bedside chest X-ray image, pulmonary abnormal high-density lesion in the registered current ICU bedside chest X-ray image, mediastinum in the registered latest ICU bedside chest X-ray image, pneumothorax in the registered latest ICU bedside chest X-ray image, pleural effusion in the registered latest ICU bedside X-ray image, pulmonary abnormal high-density lesion in the registered latest ICU bedside X-ray image and mediastinum area in the registered all images of the latest ICU bedside X-ray examination for subsequent registration, and the registered subsequent segmentation data are used for comparison.

Step S504, based on all the registered images, the registered segmentation data and the past report information, analyzing the change condition of the related diseases on the current ICU bedside chest X-ray image;

the evaluation condition of the mediastinum in the latest report in the previous report information is inquired by a mediastinum form analysis list in the comparative analysis module, the change of the mediastinum form on the X-ray image of the chest beside the ICU bed is judged according to the evaluation condition of the mediastinum, and the change condition of the mediastinum form is output; wherein the evaluation condition of the mediastinum is normal or abnormal; the diaphragm morphological change condition comprises: the diaphragm shadow is obviously enlarged, the diaphragm shadow is reduced, and the diaphragm shadow is not changed.

Inputting a current ICU bedside chest X-ray image after registration, an image of a latest ICU bedside chest X-ray examination after registration, all images of an ICU bedside chest X-ray examination before the latest ICU bedside after registration, a mediastinum region in the current ICU bedside chest X-ray image after registration, a mediastinum in the latest ICU bedside chest X-ray image after registration, a mediastinum region in all images of an ICU bedside chest X-ray examination before the latest ICU bedside after registration, and mediastinum evaluation in a latest report; outputting the prior minimum mediastinum area, qualitative judgment-mediastinum form change, qualitative judgment-mediastinum is obviously increased, and prompt information-mediastinum shadow is obviously increased.

Analysis of diaphragmatic morphological changes:

when the evaluation condition of the mediastinum is normal, comparing a mediastinum area in the registered last ICU bedside chest X-ray image with mediastinum areas in all ICU bedside chest X-ray images before the last ICU bedside chest X-ray image after registration by a mediastinum form analysis unit 402, searching a previous minimum mediastinum area, comparing and analyzing the previous minimum mediastinum area with the mediastinum area in the registered current ICU bedside chest X-ray image, and judging that the mediastinum shadow is obviously increased when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than a first preset threshold (for example, 20%) larger than the previous minimum mediastinum area; when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than a second preset threshold (for example, 10%) than the existing minimum mediastinum area, judging that the mediastinum image is increased; when the mediastinum area in the registered current ICU bedside chest X-ray image is increased by less than or equal to a second preset threshold value compared with the existing minimum mediastinum area, judging that the mediastinum image has no change;

when the evaluation condition of the mediastinum is abnormal, the mediastinum form analysis unit compares and analyzes a mediastinum area in the registered latest ICU bedside chest X-ray image with a mediastinum area in a registered current ICU bedside chest X-ray image, and when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than the mediastinum area in the registered latest ICU bedside chest X-ray image by more than a first preset threshold value, the mediastinum shadow is judged to be obviously increased; when the mediastinum area in the current ICU bedside chest X-ray image after registration is larger than the mediastinum area in the last ICU bedside chest X-ray image after registration by a second preset threshold value, judging that the mediastinum shadow is increased; when the mediastinum area in the current ICU bedside chest X-ray image after registration is smaller than the mediastinum area in the latest ICU bedside chest X-ray image after registration by a second preset threshold value, judging that the mediastinum shadow is reduced; and when the change of the mediastinum area in the current ICU bedside chest X-ray image after the registration is less than or equal to a second preset threshold value than the change of the mediastinum area in the last ICU bedside chest X-ray image after the registration, judging that the mediastinum shadow has no change.

A pleural cavity lesion analysis unit in the comparative analysis module inquires the evaluation condition of the pleural cavity in the last report in the previous report information, judges the change of the pleural cavity lesion on the current ICU bedside chest X-ray image according to the pleural cavity evaluation condition, and outputs the change condition of the pleural cavity lesion; wherein the pleural cavity is evaluated as normal or abnormal; changes in pleural cavity lesions include: newly increased pleural cavity pathological changes, normal pleural cavity, obviously increased pleural cavity pathological changes, reduced pleural cavity pathological changes and unchanged pleural cavity pathological changes.

Specifically, the method comprises the following steps: and comparing the area of the pleural cavity lesion in the current ICU bedside chest X-ray image after registration with the area of the pleural cavity lesion in the image of the latest ICU bedside chest X-ray examination after registration, and judging whether obvious pleural cavity lesion change exists or not. Inputting a registered current ICU bedside chest X-ray image, a registered image of the latest ICU bedside chest X-ray examination, areas of pneumothorax and pleural effusion in the registered current ICU bedside chest X-ray image, areas of pneumothorax and pleural effusion in the registered image of the latest ICU bedside chest X-ray examination and evaluation of a pleural cavity area in a latest report; outputting qualitative judgment, namely changes of pleural cavity lesions, qualitative judgment, remarkable increase of pleural cavity lesions, and prompt information, namely remarkable increase of pleural cavity lesions.

Analysis of pleural cavity lesions:

judge the change of pleural cavity pathological change on current ICU bedside chest X line image, include: when the pleural cavity evaluation condition is normal, the pleural cavity lesion analysis unit detects whether pleural cavity lesions exist in the registered current ICU bedside chest X-ray image, and if yes, new pleural cavity lesions are judged; if not, judging that the pleural cavity is normal;

when the evaluation condition of the pleural cavity is abnormal, the pleural cavity lesion analysis unit compares a pleural cavity lesion area in the registered latest ICU bedside chest X-ray image with a pleural cavity lesion area in the registered current ICU bedside chest X-ray image, and when the pleural cavity lesion area in the registered current ICU bedside chest X-ray image is larger than a pleural cavity lesion area in the registered latest ICU bedside chest X-ray image by more than a third preset threshold (for example, 20%), the pleural cavity lesion is judged to be remarkably increased; when the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is larger than a fourth preset threshold (for example, 10%) than the pleural cavity lesion area in the latest ICU bedside chest X-ray image after registration, it is determined that the pleural cavity lesion area is increased; when the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is smaller than a fourth preset threshold value compared with the pleural cavity lesion area in the latest ICU bedside chest X-ray image after registration, the pleural cavity lesion area is judged to be reduced; and when the pleural cavity lesion area in the current ICU bedside chest X-ray image after registration is changed by less than or equal to a fourth preset threshold value compared with the pleural cavity lesion area in the last ICU bedside chest X-ray image after registration, judging that the pleural cavity lesion is not changed.

A lung lesion analysis unit in the comparative analysis module inquires the evaluation condition of the lung lesion in the latest report in the previous report information, judges the change of the lung lesion on the chest X-ray image beside the current ICU bed according to the evaluation condition of the lung lesion and outputs the change condition of the lung lesion; wherein the evaluation condition of the lung lesion is normal or abnormal; changes in lung lesions include: newly added lung lesion, normal lung, obvious increase of lung lesion, reduction of lung lesion and no change of lung lesion.

when there is a lung lesion morphological change, the result is returned to the control of the structured report "lung lesion morphological change". When the lung pathological changes are remarkably increased, prompt messages are sent and processed by related personnel and are recorded in a database.

Analysis of changes in lung lesions:

the method for judging the change of lung lesion on the current ICU bedside chest X-ray image comprises the following steps: when the evaluation condition of the lung lesion is normal, the lung lesion analysis unit detects whether the registered current ICU bedside chest X-ray image has the lung lesion, and if so, the registered current ICU bedside chest X-ray image is judged to have the newly added lung lesion; if not, judging that the lung is normal; when the evaluation condition of the lung lesion is abnormal, the lung lesion analysis unit compares and analyzes the abnormal high-density lesion region of the lung in the registered latest ICU bedside chest X-ray image with the abnormal high-density lesion region of the lung in the registered current ICU bedside chest X-ray image, and when the abnormal high-density lesion region of the lung in the registered current ICU bedside chest X-ray image is increased by more than a fifth preset threshold (for example, 20%) than the abnormal high-density lesion region of the lung in the registered latest ICU bedside chest X-ray image, the lung lesion is judged to be remarkably increased; when the lung abnormal high-density lesion area of the pleural cavity lesion area in the current ICU bedside chest X-ray image after the registration is larger than a sixth preset threshold (10%) than the lung abnormal high-density lesion area in the last ICU bedside chest X-ray image after the registration, the lung lesion area is judged to be increased; when the lung abnormal high-density lesion area in the current ICU bedside chest X-ray image after registration is smaller than a sixth preset threshold value compared with the lung abnormal high-density lesion area in the latest ICU bedside chest X-ray image after registration, the lung abnormal high-density lesion area is judged to be reduced; and when the change of the abnormal high-density lesion area of the lung in the pleural cavity lesion area in the current ICU bedside chest X-ray image after the registration is less than or equal to a sixth preset threshold value than the change of the abnormal high-density lesion area of the lung in the last ICU bedside chest X-ray image after the registration, judging that the lung lesion is not changed.

And step S505, automatically outputting final diagnosis data based on the change condition of the related diseases.

The structured reporting module integrates the findings of all functional modules to derive an overall diagnostic impression. Based on rules built into the structured report, the final diagnostic data is automatically obtained and returned to the diagnostic impression of the structured report.

All data, all images are stored in a structured report database.

From the above description, it can be seen that the above-described embodiments of the present invention achieve the following technical effects: the embodiment of the invention applies an AI model and a program based on rules to the automatic comparative analysis of the current ICU bedside chest X-ray film and the existing bedside chest X-ray film to obtain an intelligent diagnosis report system for bedside chest X-ray examination, accesses the intelligent diagnosis report system into a PACS/RIS to automatically obtain patient medical history information, existing images and existing report information, can segment the images after the image acquisition is finished, can automatically output the change condition of the mediastinum form, the change condition of pleural cavity lesion and the change condition of lung lesion by using the segmented data and the existing images and the existing report information through the registered images and the registered segmented data area, automatically finishes evaluation, and automatically transmits the result into a structured report to improve the working efficiency of a doctor. More importantly, when the intelligent system finds that the potential risk exists, the intelligent system immediately sends out warning prompts to relevant medical care personnel so as to help the medical care personnel to pay attention to the abnormal conditions of the patients immediately, and the patients are treated and treated effectively in a preferential manner, so that the safety of the patients is guaranteed.

It will be apparent to those skilled in the art that the modules or steps of the present invention described above can be implemented by a general purpose computing device, they can be centralized in a single computing device or distributed over a network of multiple computing devices, and they can alternatively be implemented by program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or fabricated separately as individual integrated circuit modules, or fabricated as a single integrated circuit module from multiple modules or steps. Thus, the present invention is not limited to any specific combination of hardware and software.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting system, which is characterized by comprising: a patient information acquisition module, a segmentation data acquisition module, an image registration module, a comparative analysis module, and a structured reporting module, wherein,

the patient information acquisition module is respectively connected with the segmentation data acquisition module, the image registration module, the comparison analysis module and the structured report module and is used for acquiring a current ICU bedside chest X-ray image of a patient, patient medical history information, a previous image and previous report information; the previous image is an ICU bedside chest X-ray image of the patient at the latest time and all ICU bedside X-ray images before the latest time;

the segmentation data acquisition module is respectively connected with the patient information acquisition module, the image registration module and the structured report module, and is used for inputting the current ICU bedside chest X-ray image and the previous image into a plurality of deep learning models for image segmentation to respectively obtain corresponding segmentation data; the types of segmentation data comprise segmentation data of a breast structure and segmentation data of a breast lesion;

the image registration module is respectively connected with the patient information acquisition module, the segmentation data acquisition module, the comparative analysis module and the structured report module, and is used for carrying out image registration on the current ICU bedside chest X-ray image and the previous image based on the segmentation data and outputting all registered images and the registered segmentation data;

the contrast analysis module is respectively connected with the patient information acquisition module, the image registration module and the structured report module, and is used for analyzing the change condition of related diseases on the chest X-ray image beside the current ICU bed based on all registered images, the registered segmentation data and the previous report information;

the structured report module is connected with the patient information acquisition module, the segmentation data acquisition module, the image registration module and the comparative analysis module, and is used for automatically outputting final diagnosis data based on the change condition of the related diseases.

2. An ICU bedside chest X-ray automatic contrast diagnostic intelligent reporting system as in claim 1 wherein said chest structure segmentation data is: all chest imaging areas, lung field areas and mediastinum areas of the current ICU bedside chest X-ray image and the existing image; the segmentation data of the breast lesion are as follows: pleural cavity lesion areas and abnormal high-density lesion areas of the current ICU bedside chest X-ray image and the latest ICU bedside chest X-ray image; wherein, pleural cavity pathological change region includes pneumothorax region, pleural effusion region.

3. The ICU bedside chest X-ray automatic contrast diagnostic intelligent reporting system of claim 2 wherein the full images after registration are: the registered current ICU bedside chest X-ray image, the registered latest ICU bedside chest X-ray image and the registered latest before ICU bedside chest X-ray image are obtained; the registered segmentation data is: the chest X-ray image comprises a mediastinum region in a current ICU bedside chest X-ray image after registration, a pleural cavity lesion region in the current ICU bedside chest X-ray image after registration, a lung abnormal high-density lesion region in the current ICU bedside chest X-ray image after registration, a mediastinum region in a latest ICU bedside chest X-ray image after registration, a pleural cavity lesion region in a latest ICU bedside chest X-ray image after registration, a lung abnormal high-density lesion region in the latest ICU bedside chest X-ray image after registration and a mediastinum region in all ICU bedside chest X-ray images before the latest ICU bedside after registration.

4. The ICU bedside chest X-ray automatic contrast diagnostic intelligent reporting system of claim 3, wherein said contrast analysis module further comprises: the mediastinum form analysis unit is used for inquiring the evaluation condition of mediastinum in the last report in the previous report information, judging the form change of the mediastinum on the X-ray image of the chest beside the current ICU bed according to the evaluation condition of the mediastinum, and outputting the form change condition of the mediastinum; wherein the evaluation condition of the mediastinum is normal or abnormal; the mediastinal morphological change condition comprises the following steps: the image of the mediastinum is obviously enlarged, the image of the mediastinum is reduced, and the image of the mediastinum is not changed.

5. The ICU bedside chest radiography (AFM) intelligent reporting system of claim 3, wherein said contrastive analysis module further comprises: the pleural cavity lesion analysis unit is used for inquiring the evaluation condition of the pleural cavity in the last report in the previous report information, judging the change of the pleural cavity lesion on the chest X-ray image beside the current ICU bed according to the evaluation condition of the pleural cavity, and outputting the change condition of the pleural cavity lesion; wherein the pleural cavity is assessed as normal or abnormal; the changes of the pleural cavity lesion comprise: newly increased pleural cavity pathological changes, normal pleural cavity, obviously increased pleural cavity pathological changes, reduced pleural cavity pathological changes and unchanged pleural cavity pathological changes.

6. The ICU bedside chest X-ray automatic contrast diagnostic intelligent reporting system of claim 3, wherein said contrast analysis module further comprises: the lung lesion analysis unit is used for inquiring the evaluation condition of the lung lesion in the latest report in the previous report information, judging the change of the lung lesion on the chest X-ray image beside the current ICU bed according to the evaluation condition of the lung lesion and outputting the change condition of the lung lesion; wherein the assessment of the lung lesion is normal or abnormal; the changes in the lung lesions include: newly added lung lesion, normal lung, obvious increase of lung lesion, reduction of lung lesion and no change of lung lesion.

7. The ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting system of claim 4, wherein the determining of mediastinal morphological changes on the current ICU bedside chest X-ray image comprises:

when the evaluation condition of the mediastinum is normal, the mediastinum form analysis unit compares a mediastinum area in the registered latest ICU bedside chest X-ray image with mediastinum areas in all images of ICU bedside chest X-ray before the latest ICU bedside after the registration, searches for a previous minimum mediastinum area, performs comparative analysis on the previous minimum mediastinum area and the mediastinum area in the registered current ICU bedside chest X-ray image, and judges that the mediastinum image is remarkably increased when the mediastinum area in the registered current ICU bedside chest X-ray image is increased by more than a first preset threshold value compared with the previous minimum mediastinum area; when the mediastinum area in the registered current ICU bedside chest X-ray image is larger than a second preset threshold value in comparison with the minimum mediastinum area in the past, judging that the mediastinum image is increased; when the mediastinum area in the registered current ICU bedside chest X-ray image is increased by less than or equal to a second preset threshold value compared with the minimum previous mediastinum area, judging that the mediastinum image has no change;

when the evaluation condition of the mediastinum is abnormal, the mediastinum form analysis unit compares and analyzes a mediastinum area in the registered latest ICU bedside chest X-ray image with a mediastinum area in the registered current ICU bedside chest X-ray image, and when the mediastinum area in the registered current ICU bedside chest X-ray image is increased by more than the first preset threshold value compared with the mediastinum area in the registered latest ICU bedside chest X-ray image, the mediastinum shadow is judged to be remarkably increased; when the mediastinum area in the current ICU bedside chest X-ray image after the registration is larger than the mediastinum area in the last ICU bedside chest X-ray image after the registration by more than a second preset threshold value, judging that the mediastinum shadow is increased; when the mediastinum area in the registered current ICU bedside chest X-ray image is smaller than the mediastinum area in the registered latest ICU bedside chest X-ray image by less than the second preset threshold, judging that the mediastinum shadow is reduced; and when the change of the mediastinum area in the current ICU bedside chest X-ray image after the registration is smaller than or equal to the second preset threshold value than the change of the mediastinum area in the last ICU bedside chest X-ray image after the registration, judging that the mediastinum shadow has no change.

8. The ICU bedside chest X-ray automatic comparative diagnosis intelligent reporting system of claim 5, wherein determining changes in pleural cavity lesions on the current ICU bedside chest X-ray image comprises:

when the pleural cavity evaluation condition is normal, the pleural cavity lesion analysis unit detects whether pleural cavity lesions exist in the registered current ICU bedside chest X-ray image, and if yes, the newly added pleural cavity lesions are judged; if not, judging that the pleural cavity is normal;

when the evaluation condition of the pleural cavity is abnormal, the pleural cavity lesion analysis unit compares and analyzes a pleural cavity lesion area in the registered latest ICU bedside chest X-ray image with a pleural cavity lesion area in the registered current ICU bedside chest X-ray image, and when the pleural cavity lesion area in the registered current ICU bedside chest X-ray image is increased by more than a third preset threshold value compared with the pleural cavity lesion area in the registered latest ICU bedside chest X-ray image, the pleural cavity lesion is judged to be remarkably increased; when the pleural cavity lesion area in the current ICU bedside chest X-ray image after the registration is larger than the pleural cavity lesion area in the last ICU bedside chest X-ray image after the registration by a fourth preset threshold value, judging that the pleural cavity lesion area is increased; when the pleural cavity lesion area in the current ICU bedside chest X-ray image after the registration is smaller than the pleural cavity lesion area in the latest ICU bedside chest X-ray image after the registration by a fourth preset threshold value, judging that the pleural cavity lesion area is reduced; and when the pleural cavity lesion area in the current ICU bedside chest X-ray image after the registration is changed by less than or equal to a fourth preset threshold value compared with the pleural cavity lesion area in the last ICU bedside chest X-ray image after the registration, judging that the pleural cavity lesion is not changed.

9. The ICU bedside chest X-ray automatic contrast diagnostic smart reporting system of claim 6 wherein said determining a change in a lung lesion on said current ICU bedside chest X-ray image comprises:

when the evaluation condition of the lung lesion is normal, the lung lesion analysis unit detects whether the registered current ICU bedside chest X-ray image has the lung lesion, and if so, the newly added lung lesion is judged; if not, judging that the lung is normal;

when the evaluation condition of the lung lesion is abnormal, the lung lesion analysis unit compares and analyzes the abnormal high-density lesion area of the lung in the registered latest ICU bedside chest X-ray image with the abnormal high-density lesion area of the lung in the registered current ICU bedside chest X-ray image, and when the abnormal high-density lesion area of the lung in the registered current ICU bedside chest X-ray image is increased by more than a fifth preset threshold value than the abnormal high-density lesion area of the lung in the registered latest ICU bedside chest X-ray image, the lung lesion is judged to be remarkably increased; when the lung abnormal high-density lesion area of the pleural cavity lesion area in the registered current ICU bedside chest X-ray image is larger than a sixth preset threshold value than the lung abnormal high-density lesion area in the registered latest ICU bedside chest X-ray image, judging that the lung lesion area is increased; when the lung abnormal high-density lesion area in the registered current ICU bedside chest X-ray image is smaller than the sixth preset threshold value compared with the lung abnormal high-density lesion area in the registered latest ICU bedside chest X-ray image, the lung abnormal high-density lesion area is judged to be reduced; and when the change of the lung abnormal high-density lesion area in the pleural cavity lesion area in the current ICU bedside chest X-ray image after the registration is smaller than or equal to a sixth preset threshold value than the change of the lung abnormal high-density lesion area in the last ICU bedside chest X-ray image after the registration, judging that the lung lesion is not changed.

10. An ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting method is characterized by comprising the following steps:

acquiring a chest X-ray image of a patient beside a current ICU (intensive care unit) bed, patient medical history information, an existing image and past report information; wherein the previous image is an ICU bedside chest X-ray image of the patient at the latest time and all ICU bedside X-ray images before the latest time;

inputting the current ICU bedside chest X-ray image and the previous image into a plurality of deep learning models for image segmentation to respectively obtain corresponding segmentation data; the types of segmentation data comprise segmentation data of a breast structure and segmentation data of a breast lesion;

based on the segmentation data, carrying out image registration on the current ICU bedside chest X-ray image and the previous image, and outputting all registered images and the registered segmentation data;

analyzing the change condition of related diseases on the current ICU bedside chest X-ray image based on the registered all images, the registered segmentation data and the previous report information;

and automatically outputting final diagnosis data based on the change condition of the related diseases.

11. The ICU bedside chest X-ray automatic contrast diagnostic intelligent reporting method of claim 10 wherein the chest structure segmentation data is: all chest imaging areas, lung field areas and mediastinum areas of the current ICU bedside chest X-ray image and the existing image; the segmentation data of the breast lesion are as follows: pleural cavity lesion areas and abnormal high-density lesion areas of the current ICU bedside chest X-ray image and the latest ICU bedside chest X-ray image; wherein, pleural cavity pathological change region includes pneumothorax region, pleural effusion region.

12. The ICU bedside chest X-ray automatic contrast diagnostic intelligent reporting method of claim 11, wherein said registered total images are: the registered current ICU bedside chest X-ray image, the registered latest ICU bedside chest X-ray image and the registered latest before ICU bedside chest X-ray image are obtained; the registered segmentation data is: the chest X-ray image comprises a mediastinum area in the current ICU bedside chest X-ray image after registration, a pleural cavity lesion area in the current ICU bedside chest X-ray image after registration, a lung abnormal high-density lesion area in the current ICU bedside chest X-ray image after registration, a mediastinum area in the last ICU bedside chest X-ray image after registration, a pleural cavity lesion area in the last ICU bedside chest X-ray image after registration, a lung abnormal high-density lesion area in the last ICU bedside chest X-ray image after registration and a mediastinum area in all the ICU bedside chest X-ray images before the last time after registration.