Image processing device, image processing method, and recording medium
There is provided an image processing device including an input receiver of inputting an image obtained by photographing a specimen subjected to staining and a hardware processor. The hardware processor extracts a region subjected to the staining from the image as a cell region; extracts a region as a candidate region, the region being surrounded by the cell region and not being subjected to the staining. The hardware processor further extracts a feature amount of the candidate region; determines whether or not the candidate region is a cell region on a basis of the feature amount; and corrects the candidate region which is determined to be a cell region by the distinction means to be a cell region.
1. An image processing device comprising:
a hardware processor that:
receives an image obtained by photographing a specimen subjected to staining;
extracts a region subjected to the staining from the image as a cell region;
extracts a hole region as a candidate region, the hole region being surrounded by the cell region and not being subjected to the staining;
extracts a feature amount of the candidate region;
determines, based on the feature amount, the candidate region is an unstained inner portion of a cell nucleus; and
corrects the unstained inner portion of the cell nucleus.
2. The image processing device according to claim 1 , wherein the hardware processor performs binarization processing so that the candidate region is a cell region.
3. The image processing device according to claim 1 ,
wherein the hardware processor sets a threshold value of the feature amount on a basis of at least one of a cancer type, progression of cancer, a method of collecting tissue, and a staining method, and
determines, on a basis of the feature amount and the threshold value, whether or not the candidate region is a cell region.
4. The image processing device according to claim 1 , wherein the feature amount includes a distinction result by a machine learning method.
5. The image processing device according to claim 1 , wherein the feature amount includes at least one of an area, a circularity, a convex hull rate of an area, a number of concave/convex portions, and a curvature of the candidate region.
6. The image processing device according to claim 1 , wherein the feature amount includes at least one of an average, a variation, and a difference of pixel value inside of or around the candidate region.
7. The image processing device according to claim 1 , wherein the feature amount includes at least one of an edge intensity, an edge direction, and a variation in an edge direction of the candidate region.
8. An image processing method comprising:
inputting an image obtained by photographing a specimen subjected to staining;
extracting a region subjected to the staining from the image as a cell region;
extracting a hole region as a candidate region, the hole region being surrounded by the cell region and not being subjected to the staining;
extracting a feature amount of the candidate region;
determining, based on the feature amount, the candidate region is an unstained inner portion of the cell nucleus; and
correcting the unstained inner portion of the cell nucleus.
9. A non-transitory computer readable medium storing an image processing program to cause a computer to:
receive an image obtained by photographing a specimen subjected to staining;
extract a region subjected to the staining from the image as a cell region;
extract a hole region as a candidate region, the hole region being surrounded by the cell region and not being subjected to the staining;
extract a feature amount of the candidate region;
determine, based on the feature amount, the candidate region is an unstained inner portion of a cell nucleus; and
correct the unstained inner portion of the cell nucleus.