Image diagnosis assisting apparatus, image diagnosis assisting method and sample analyzing system
Reaction values indicating based on which areas in an input image a plurality of identification results representing the states of cells or tissues are judged are calculated, and differences between a plurality of reaction values are obtained to thereby calculate the degree to which cells or tissues in an area of interest likely are in each state. Thereby, each state of cells or tissues can be visualized at an accurate position in an area of interest; therefore, diagnostic assistance to pathologist becomes possible.
1. An image diagnosis assisting apparatus comprising:
a local area feature amount calculating unit that calculates a local area feature amount from an input image, wherein the local area feature amount is a three-dimensional array with dimensions including a feature amount number, a vertical position, and a horizontal position;
an identifying unit that performs a plurality of identification processes using the local area feature amount, including identifying a plurality of states of cells or tissues in the input image and a likelihood of each state;
a reaction value calculating unit that calculates a plurality of reaction values indicating a degree of contribution of each dimension of the local area feature amount to each identification result, based on which unit area in the input image each identification result is judged, using a calculation result of the local area feature amount and a result of the plurality of identification processes; and
a reaction value difference calculating unit that calculates differences between the plurality of reaction values.
2. The image diagnosis assisting apparatus according to claim 1 , including:
a staining shade classifying unit that classifies the input image according to staining shade before the local area feature amount calculating unit calculates the local area feature amount from the input image.
3. The image diagnosis assisting apparatus according to claim 1 wherein
the local area feature amount calculating unit updates a coefficient group to be used in calculation of a local area feature amount from the input image or a coefficient group to be used in a plurality of identification processes using the local area feature amount.
4. An image diagnosis assisting method comprising:
a local area feature amount calculation step of calculating, by an arithmetic operation unit, a local area feature amount from an input image, wherein the local area feature amount is a three-dimensional array with dimensions including a feature amount number, a vertical position, and a horizontal position;
an identification step of performing, by the arithmetic operation unit, a plurality of identification processes using the local area feature amount, including identifying a plurality of states of cells or tissues in the input image and a likelihood of each state;
a reaction value calculation step of calculating, by the arithmetic operation unit, a plurality of reaction values indicating a degree of contribution of each dimension of the local area feature amount to each identification result, based on which unit area in the input image each identification result is judged, using a calculation result of the local area feature amount and a result of a plurality of identification processes; and
a reaction value difference calculation step of calculating, by the arithmetic operation unit, differences between a plurality of reaction values,
wherein the local area feature amount is a three-dimensional array with dimensions including a feature amount number, a vertical position, and a horizontal position.
5. The image diagnosis assisting method according to claim 4 , including:
a staining shade classification step of classifying, by the arithmetic operation unit, the input image according to staining shade before the local area feature amount is calculated from the input image by the arithmetic operation unit at the local area feature amount calculation step.
6. The image diagnosis assisting method according to claim 4 , including:
an update step of updating, by the arithmetic operation unit, a coefficient group to be used in calculation of the local area feature amount from the input image or a coefficient group to be used in a plurality of identification processes using the local area feature amount.
7. A sample analyzing system comprising:
a local area feature amount calculating unit that calculates a local area feature amount from an input image, wherein the local area feature amount is a three-dimensional array with dimensions including a feature amount number, a vertical position, and a horizontal position;
an identifying unit that performs a plurality of identification processes using the local area feature amount, including identifying a plurality of states of cells or tissues in the input image and a likelihood of each state;
a reaction value calculating unit that calculates a plurality of reaction values indicating a degree of contribution of each dimension of the local area feature amount to each identification result, based on which unit area in the input image each identification result is judged, using a calculation result of the local area feature amount and a result of the plurality of identification processes;
a reaction value difference calculating unit that calculates differences between the plurality of reaction values;
a sample collecting unit that collects a sample based on the calculated information; and
an analyzing unit that performs analysis of the sample collected by the sample collecting unit,
wherein the local area feature amount is a three-dimensional array with dimensions including a feature amount number, a vertical position, and a horizontal position.
8. The sample analyzing system according to claim 7 , including:
a staining shade classifying unit that classifies the input image according to the staining shade before the local area feature amount calculating unit calculates the local area feature amount from the input image.
9. The sample analyzing system according to claim 7 , wherein
the local area feature amount calculating unit updates a coefficient group to be used in calculation of a local area feature amount from the input image or a coefficient group to be used in a plurality of identification processes using the local area feature amount.