Processor for endoscope, program, information processing method, and information processing device
A processor for an endoscope includes: an endoscopic image acquisition unit that acquires an endoscopic image of a patient from the endoscope; a virtual endoscopic image acquisition unit that acquires a virtual endoscopic image reconstructed on the basis of a three-dimensional medical image obtained by capturing an image of the patient in advance; a virtual endoscopic image reconstruction unit that reconstructs a corrected virtual endoscopic image that matches most with the endoscopic image on the basis of the degree of matching between the virtual endoscopic image and the endoscopic image; and a diagnosis support information output unit that outputs diagnosis support information based on a feature parameter corrected according to a correspondence between each pixel of the endoscopic image and a distance image obtained from the corrected virtual endoscopic image.
1. A device for converting a signal from an endoscope into an image comprising:
a processor configured to
acquire an endoscopic image of a patient from the endoscope via a signal transmitted from the endoscope to the processor;
acquire a virtual endoscopic image reconstructed on the basis of a three-dimensional medical image obtained by capturing an image of the patient in advance;
reconstruct a corrected virtual endoscopic image that matches most with the endoscopic image on the basis of a degree of matching between the acquired virtual endoscopic image and the acquired endoscopic image; and
output to a display diagnosis support information based on a feature parameter corrected according to a correspondence between each pixel of the acquired endoscopic image and a distance image obtained from the reconstructed virtual endoscopic image, wherein
the display displays in words whether the endoscopic image shows the presence or absence of a lesion on a first screen in response to receiving the diagnosis support information output from the processor, and
the display displays in words the effective mass number of a selected pixel of the virtual endoscopic image on the first screen that also displays in words whether the endoscopic image shows the presence or absence of a lesion in response to receiving the diagnosis support information output from the processor.
2. The device according to claim 1 , wherein the processor is further configured to:
measure an insertion distance of the endoscope inserted into a body of the patient;
acquire bending history information of the endoscope; and
correct a Z coordinate, which is the insertion distance of the endoscope, a viewpoint direction, and a viewpoint position of a virtual endoscope according to the acquired bending history information.
3. The device according to claim 1 , wherein the processor is further configured to:
reconstruct the distance image on the basis of the virtual endoscopic image; and
correct a pixel value of each pixel of the corresponding endoscopic image according to a distance obtained from the distance image on the basis of the reconstructed distance image.
4. The device according to claim 3 , wherein the processor is further configured to correct the feature parameter of each image region of the corresponding endoscopic image on the basis of the reconstructed distance image.
5. The device according to claim 4 , wherein the processor is further configured to output the diagnosis support information including a tumor candidate on the basis of the corrected pixel value of each pixel or the corrected feature parameter.
6. The device according to claim 4 , wherein the processor is further configured to output a recognition result by using a trained image recognition model that outputs the recognition result in a case where the corrected endoscopic image corrected on the basis of the corrected pixel value of each pixel or the corrected feature parameter is input.
7. The device according to claim 1 , wherein the display displays in words whether the endoscopic image shows the presence or absence of a lesion, in association with the display of the endoscopic image in response to receiving the diagnosis support information output from the processor.
8. The device according to claim 1 , wherein the display displays, in addition to the words indicating the presence or absence of a lesion, the stage of a symptom of the patient in response to receiving the diagnosis support information output from the processor.
9. The device according to claim 1 , wherein the display displays information in
a management ID column,
a patient ID column,
a diagnosis content column, and
a diagnosis date and time column, in response to receiving the diagnosis support information output from the processor.
10. A device for converting a signal from an endoscope into an image comprising:
a processor configured to
acquire an endoscopic image of a patient from the endoscope via a signal transmitted from the endoscope to the processor;
acquire a virtual endoscopic image reconstructed on the basis of a three-dimensional medical image obtained by capturing an image of the patient in advance;
reconstruct a corrected virtual endoscopic image that matches most with the endoscopic image on the basis of a degree of matching between the acquired virtual endoscopic image and the acquired endoscopic image;
output to a display diagnosis support information based on a feature parameter corrected according to a correspondence between each pixel of the acquired endoscopic image and a distance image obtained from the reconstructed virtual endoscopic, wherein the display displays in words whether the endoscopic image shows the presence or absence of a lesion in response to receiving the diagnosis support information output from the processor;
reconstruct the distance image on the basis of the virtual endoscopic image;
correct a pixel value of each pixel of the corresponding endoscopic image according to a distance obtained from the distance image on the basis of the reconstructed distance image; and
correct the feature parameter of each image region of the corresponding endoscopic image on the basis of the reconstructed distance image, wherein
the display outputs, in response to receiving the diagnosis support information output from the processor, a recognition result by using a trained image recognition model that outputs the recognition result in a case where the corrected endoscopic image corrected on the basis of the corrected pixel value of each pixel of the corresponding endoscopic image or on the basis of the corrected feature parameter is input, and
the processor is further configured to output AI diagnosis support information using AI to output the recognition result.