Medical-image processing apparatus
View Patent ↗A medical-image processing apparatus according to an embodiment includes an extracting unit, a dividing unit, and an estimating unit. The extracting unit extracts a disease candidate region from a medical image. The dividing unit divides the disease candidate region into multiple partial regions. The estimating unit uses the feature value of each of the partial regions to estimate the disease state of the disease candidate region.
1. A medical-image processing apparatus comprising:
processing circuitry configured to:
extract a disease candidate region from a medical image;
divide the disease candidate region into an internal region and an external region, by performing erosion to the disease candidate region, the erosion being repeatedly performed so that a ratio between a size of the internal region and a size of the external region becomes a predetermined ratio;
extract a first feature vector of the internal region and a second feature vector of the external region;
generate a combining vector by combining the first feature vector and the second feature vector; and
estimate a disease state of the disease candidate region by using the combining vector,
wherein a number of elements of the combining vector is a total number of elements obtained by adding a number of elements of the first feature vector and a number of elements of the second feature vector wherein the processing circuitry is configured to: accept an estimation result in which a disease state of the disease candidate region is estimated; generate a classifier for a computer-aided diagnosis based on the combining vector and the estimation result; and estimate a disease state of a newly extracted disease candidate region by using the classifier.
2. The medical-image processing apparatus according to claim 1 , wherein the processing circuitry is configured to:
extract a third feature vector of disease candidate region before the dividing;
generate a combining vector by combining the first feature vector, the second feature vector and the third feature vector; and
estimate a disease state of the disease candidate region by using the combining vector,
wherein a number of elements of the combining vector is a total number of elements obtained by adding a number of elements of the first feature vector, a number of elements of the second feature vector, and a number of elements of the third feature vector.
3. The medical-image processing apparatus according to claim 1 , wherein the medical image is multiple medical image data sets.