Endoscope image processing device
An image processing device including a housing and a processing circuit in the housing, the processing circuit including a processor and memory, the memory including a neural network model and proximity suppression logic, the neural network model comprising a single-pass neural network model trained with training images corresponding to an endoscopic procedure and defining anatomic references observable in the training images, wherein the neural network model is configured to process images, to detect the anatomic references in the images, and to output a set of anatomic references including identifiers and confidence values representing the likelihoods that the anatomic reference identifiers are correct, and wherein the proximity suppression logic is configured to change the confidence values of the anatomic references in the set of anatomic references based on a prior position of the endoscope to identify an anatomic reference indicative of the current position of the endoscope.
1 . An image processing device to facilitate an endoscopic procedure using an endoscope having an image sensor configured to generate images, the image processing device comprising:
a housing; and
a processing circuit in the housing, the processing circuit including a processor and memory, the memory comprising a neural network model and proximity suppression logic, the neural network model comprising a single-pass neural network model trained with training images corresponding to the endoscopic procedure and defining anatomic references observable in the training images, the processor being configured to process the neural network model and the proximity suppression logic,
wherein the neural network model is configured to process the images, to detect the anatomic references in the images, and to output a set of anatomic references including identifiers and confidence values representing the likelihoods that the anatomic reference identifiers are correct, and
wherein the proximity suppression logic is configured to change the confidence values of the anatomic references in the set of anatomic references based on a prior position of the endoscope to identify an anatomic reference indicative of the current position of the endoscope.
2 . The image processing device of claim 1 , wherein the proximity suppression logic comprises a proximity suppression map including anatomical references and weights corresponding to the anatomical references, the weights predetermined based on a proximity of an anatomical reference to the other anatomical references, wherein the proximity suppression logic matches the prior position of the endoscope with one of the anatomical references in the map and changes the confidence values by multiplying the confidence values and the respective weights.
3 . The image processing device of claim 1 , further comprising an organ model including representations of lumens of organs of a patient to be examined.
4 . The image processing device of claim 3 , further comprising graphical user interface (GUI) logic configured to present the organ model with a display screen communicatively connected to the image processing device, wherein the GUI logic is additionally configured to present with the display screen an indication of the current position of the endoscope.
5 . The image processing device of claim 4 , further comprising a medical device interface configured to receive the images from the endoscope, the image processing device configured to determine, based on the medical device interface, a type of endoscopic procedure and to select the organ model, from two or more organ models stored in the memory, based on the type of the endoscopic procedure.
6 . The image processing device of claim 4 , wherein the image processing device is further configured to determine a route comprising a plurality of anatomic references.
7 . The image processing device of claim 6 , wherein the image processing device is further configured to determine whether the current position of the endoscope is on the route and output a warning indication if the current position is not on the route.
8 . The image processing device of claim 1 , wherein the image processing device is configured to repeatedly store in a prior position database a position indication of the current position of the endoscope as the endoscope is moved during the endoscopic procedure and the current position changes, the position indication comprising at least one of the current position of the endoscope and/or the anatomic reference indicative of the current position of the endoscope, and the prior position database containing the position indications generated during the endoscopic procedure.
9 . The image processing device of claim 8 , wherein the image processing device is configured to determine, by comparing the position indications in the prior position database with the position indication of the current position, if the current position of the endoscope matches a prior position of the endoscope.
10 . The image processing device of claim 8 , wherein the image processing device is configured to generate a report documenting each of the position indications in the database.
11 . The image processing device of claim 10 , wherein the report depicts an organ model with markings reflecting prior endoscope positions stored in the prior position database.
12 . The image processing device of claim 10 , wherein the processor is configured to store, in response to a user input, an image together with first data indicative of a position of the endoscope in an organ model when the image was recorded, and wherein the report comprises the image and the first data.
13 . An endoscope system comprising:
the endoscope; and
an image processing device according to claim 1 .
14 . The endoscope system of claim 13 , further comprising a display screen.
15 . The endoscope system of claim 13 , further comprising a physical model of an organ of the human body, the organ comprising lumens and the physical model comprising model lumens, the physical model being a three-dimensional model, wherein the neural network model is trained with training images corresponding to the endoscopic procedure performed in the model lumens of the physical model.
16 . The endoscope system of claim 15 , wherein the physical model comprises visual markers detectable with the endoscope during the endoscopic procedure.