Method, software program and system for detecting image irregularities in video endoscopic instrument produced images
A method for detecting an image irregularity in one or more still images or video images produced by an endoscopic instrument. The method including: producing the one or more still images or video images; transmitting the one or more still images or video images to a processor comprising hardware; detecting, with the processor, a presence or an absence of the image irregularity in the one or more still images or video images; and where the presence of the image irregularity is detected, issuing a notification to a user about the presence of the image irregularity.
1 . A method for detecting an image irregularity in one or more images produced by an endoscopic instrument, the method comprising:
receiving, by a processor, the one or more images from the endoscopic instrument during an operation using the endoscope instrument, wherein the one or more images are produced by the endoscopic instrument during the operation using the endoscope instrument;
detecting, with the processor, a presence or an absence of the image irregularity in the one or more images; and
when the presence of the image irregularity is detected:
classifying, by the processor, the detected image irregularity into a type of defect of the endoscope instrument, wherein the type of defect includes one of mechanical, internal and electronic defects of the endoscope instrument; and
issuing, by the processor, a notification to a display, wherein the notification indicates the presence of the image irregularity in the one or more images and a prompt for a corrective action to address the classified type of defect of the endoscope instrument.
2 . The method according to claim 1 , wherein detecting the presence of the image irregularity comprises performing, by the processor, image processing on the one or more images.
3 . The method according to claim 2 , wherein the image processing includes at least one of:
a color tinge analysis of images with neutral color settings and proper white balance;
an analysis of irregular pixels, pixel clusters, rows or columns of pixels in the one or more images, and pixels that deviate from surrounding pixels; and
an analysis of smearing in the one or more images.
4 . The method according to claim 2 , wherein detecting the absence or the presence of the image irregularity and the classifying comprise inputting the one or more images to a convolutional neural network (CNN) to the processor, the CNN being trained for the detection of the image irregularity in endoscopic images, and the CNN performing a detection and the classification of the detected image irregularity in the one or more images.
5 . The method according to claim 4 , wherein the training of the CNN comprises a training image dataset of still training images or video training images featuring various image irregularities of known types and locations that are labelled with a classifier indicating one or more of the class of the respective image irregularity and an indication of the location of the image irregularity.
6 . The method according to claim 5 , wherein the indication of the location of the image irregularity comprises one or more of feature extraction and fine tuning.
7 . The method according to claim 1 , wherein the issuing of a notification about the presence of the image irregularity comprises optically or acoustically signalling to a user of the endoscope instrument.
8 . The method according to claim 1 , wherein issuing the notification comprises one or more of:
logging in to a clinical endoscopic reprocessing system that notifies a user charged with supervising one or more of the reprocessing of the endoscopic instrument and the reprocessing system itself; and
logging in to a locally installed or cloud-based supervisory system that notifies the user charged with supervising the reprocessing of the endoscopic instrument.
9 . The method according to claim 1 , further comprising:
subsequent to the issuing, determining that the determined image irregularity requires repair of the endoscope instrument; and
issuing the notification to include a recommendation in the prompt to remove the endoscopic instrument having the image irregularity from normal operation and sending the endoscopic instrument for the repair.
10 . The method according to claim 1 , wherein the notification further includes at least one of a location of the image irregularity, a classification of the image irregularity.
11 . The method according to claim 1 , further comprising subsequent to the issuing, automatically ordering a replacement endoscopic instrument upon detection of the image irregularity.
12 . A non-transitory computer-readable storage medium storing instructions that cause a computer to at least perform:
receiving one or more images from an endoscopic instrument during an operation using the endoscope instrument, wherein the one or more images are produced by the endoscopic instrument during the operation using the endoscope instrument;
detecting a presence or an absence of an image irregularity in the one or more images; and
when the presence of the image irregularity is detected:
classifying, by the processor, the detected image irregularity into a type of defect of the endoscope instrument, wherein the type of defect includes one of mechanical, internal and electronic defects of the endoscope instrument; and
issuing, by the processor, a notification to a display, wherein the notification indicates the presence of the image irregularity in the one or more images and a prompt for a corrective action to address the classified type of defect of the endoscope instrument.
13 . The computer-readable storage medium according to claim 12 , wherein the instructions cause the computer to detect and classify the presence of image irregularity by performing including instructions for at least one of image processing and inputting the one or more images to a convolutional neural network (CNN).
14 . A system for detecting an image irregularity in one or more images produced by an endoscopic instrument, the system comprising:
at least one endoscopic instrument configured to capture the one or more images;
a display having a screen, the display being configured to display the one or more images on the screen; and
one or more processors comprising hardware, the one or more processors being configured to:
receive image signals corresponding to the one or more images from the endoscopic instrument captured by the endoscopic instrument during an operation using the endoscope instrument;
provide the image signals corresponding to one or more images to the display to display the one or more images,
detect a presence or an absence of the image irregularity in the one or more images; and
when the presence of the image irregularity is detected:
classify the detected image irregularity into a type of defect of the endoscope instrument, wherein the type of defect includes one of mechanical, internal and electronic defects of the endoscope instrument; and
issue a notification to the display, wherein the notification indicates the presence of the image irregularity in the one or more images and a prompt for a corrective action to address the classified type of defect of the endoscope instrument.
15 . The system of claim 14 , wherein the one or more processors comprise:
an image processor configured to receive the image signals corresponding to the one or more images from the endoscopic instrument and provide the image signals to display; and
an evaluation processor configured to receive the image signals from the image processor, detect the presence or the absence of the image irregularity in the one or more images and, when the presence of the image irregularity is detected, issuing the notification to the display,
wherein the image processor is further configured to transmit the image signals to the evaluation processor.
16 . The system of claim 14 , further comprising at least one of a reprocessing device for the at least one endoscopic instrument and a connection to a supervisory system.
17 . The system of claim 16 , wherein the supervisory system is one of a local supervisory system or a cloud-based supervisory system.