Labeling method and computing device by detecting lesion region in images using a learned network function
A labeling method and a computing device by detecting lesion region in images using a learned network function. The method includes obtaining a medical image, detecting and displaying a region of interest corresponding to at least one subject disease through a learned network function, and labeling a lesion region corresponding to the at least one subject disease on the medical image according to the region of interest and user inputs. The labeling includes receiving a first user input for choosing a selected region within the region of interest, setting a reference pixel value based on the selected region, detecting and displaying a lesion estimation region having pixel values within a predetermined range from the reference pixel value, and correcting the lesion estimation region according to a second user input to label the lesion region.
1 . A labeling method comprising:
obtaining a medical image;
receiving the medical image, detecting a region of interest corresponding to at least one subject disease using a learned network function, and displaying the region of interest on the medical image; and
labeling a lesion region on the medical image corresponding to the at least one subject disease, based on the region of interest and user inputs to the region of interest,
wherein the labeling the lesion region comprises:
receiving a first user input for choosing a selected region within the region of interest,
setting a reference pixel value based on the selected region,
detecting, as at least one lesion estimation region, at least one region having pixel values within a predetermined pixel value range from the reference pixel value and displaying the at least one lesion estimation region on the region of interest, and
correcting the at least one lesion estimation region according to a second user input to the at least one lesion estimation region and labeling the at least one corrected lesion estimation region as the lesion region on the region of interest,
wherein the reference pixel value represents a color value or a brightness value of the selected region, and
wherein the pixel values of the at least one region detected as the at least one lesion estimation region represent the color value or the brightness value within the predetermined pixel value range from the reference pixel value.
2 . The labeling method according to claim 1 , further comprising: setting the at least one subject disease according to the user inputs.
3 . The labeling method according to claim 1 , further comprising: correcting at least one of the regions of interest according to the user inputs and cropping and storing a corrected region of interest from the medical image.
4 . The labeling method according to claim 1 , wherein the labeling the lesion region further comprises: increasing a resolution of the region of interest, and the receiving the first user input is performed after the increasing the resolution.
5 . The labeling method according to claim 4 , wherein the labeling the lesion region further comprises: returning the resolution of the region of interest on which the lesion region is labeled to an initial resolution thereof.
6 . The labeling method according to claim 1 , wherein the displaying the at least one lesion estimation region is performed by emphasizing an outer line of the at least one lesion estimation region.
7 . The labeling method according to claim 1 , further comprising: storing the region of interest on which the lesion region is labeled.
8 . The labeling method according to claim 1 , wherein the at least one lesion estimation region includes a plurality of lesion estimation regions.
9 . The labeling method according to claim 8 ,
wherein the plurality of lesion estimation regions include a first lesion estimation region and a second lesion estimation region,
wherein the first lesion estimation region has the pixel values within a first predetermined pixel value range from the reference pixel value, and
wherein the second lesion estimation region has the pixel values within a second predetermined pixel value range from the reference pixel value.
10 . A computing device for supporting labeling, the computing device comprising:
at least one memory for storing a computer program for labeling;
a wired or wireless communication interface configured to obtain a medical image; and
at least one processor receiving the medical image, detecting a region of interest corresponding to at least one subject disease using a learned network function, and displaying the region of interest on the medical image, and labeling a lesion region on the medical image corresponding to the at least one subject disease, based on the region of interest and user inputs to the regions of interest,
wherein the at least one processor receives a first user input for choosing a selected region within the region of interest, sets a reference pixel value based on the selected region, detects, as at least one lesion estimation region, at least one region having pixel values within a predetermined pixel value range from the reference pixel value and displays the at least one lesion estimation region on the region of interest, corrects the at least one lesion estimation region according to a second user input to the at least one lesion estimation region, and labels the at least one corrected lesion estimation region as the lesion region on the region of interest,
wherein the reference pixel value represents a color value or a brightness value of the selected region, and
wherein the pixel values of the at least one region detected as the at least one lesion estimation region represent the color value or the brightness value within the predetermined pixel value range from the reference pixel value.
11 . The computing device according to claim 10 , wherein the at least one processor sets the at least one subject disease according to the user inputs.
12 . The computing device according to claim 10 , wherein the at least one processor corrects at least one of the regions of interest according to the user inputs and crops and stores a corrected region of interest from the medical image.
13 . The computing device according to claim 10 , wherein the at least one processor increases a resolution for the region of interest and receives the first user input for choosing the selected region in the region of interest whose resolution is increased.
14 . The computing device according to claim 13 , wherein the at least one processor returns the resolution of the region of interest on which the lesion region is labeled to an initial resolution thereof.
15 . The computing device according to claim 10 , wherein the at least one processor displays an outer line of the at least one lesion estimation region detected in the region of interest.
16 . The computing device according to claim 10 , wherein the at least one processor stores the region of interest on which the lesion region is labeled.
17 . The computing device according to claim 10 , wherein the at least one lesion estimation region includes a plurality of lesion estimation regions.
18 . The computing device according to claim 17 ,
wherein the plurality of lesion estimation regions include a first lesion estimation region and a second lesion estimation region,
wherein the first lesion estimation region has the pixel values within a first predetermined pixel value range from the reference pixel value, and
wherein the second lesion estimation region has the pixel values within a second predetermined pixel value range from the reference pixel value.