Diagnostic imaging support system and diagnostic imaging apparatus
A diagnostic imaging support system includes: an image input unit that receives an input of an image; a first processing unit that detects a region of interest including a subject of interest from the image, classifies the detected region of interest, and outputs a first classification result; and a second processing unit that subjects the region of interest to semantic segmentation, classifies the region of interest based on a result of the semantic segmentation, and outputs a second classification result. When a predetermined condition related to the region of interest is met, the first classification result is employed as indicating a class of the region of interest, and, when the condition is not met, the second classification result is employed as indicating the class of the region of interest.
1. A diagnostic imaging support system comprising:
a processor comprising hardware,
wherein the processor is configured to:
receive an input of an image;
detect a region of interest including a subject of interest from the image, classify the detected region of interest, and output a first classification result; and
subject the region of interest to semantic segmentation, classify the region of interest based on a result of the semantic segmentation, and output a second classification result,
wherein:
when a predetermined condition related to a shape of the region of interest identified based on pixels classified by the semantic segmentation as belonging to the region of interest is met, the first classification result is employed as indicating a class of the region of interest, and, when the condition is not met, the second classification result is employed as indicating the class of the region of interest, and
the condition requires that a ratio of a minor diameter to a major diameter of an ellipse obtained by subjecting a set of pixels classified by the semantic segmentation as belonging to the region of interest to elliptical approximation be equal to or greater than a predetermined threshold value and that an acute angle formed by a major axis of the ellipse and a longitudinal direction of the image be within a predetermined angular range.
2. The diagnostic imaging support system according to claim 1 , wherein each of an output of the first classification result and an output of the second classification result is implemented by performing a process using a convolutional neural network.
3. A non-transitory computer readable medium encoded with a program executable by a computer, the program being executable by the computer to cause the computer to perform functions comprising:
receiving an input of an image;
detecting a region of interest including a subject of interest from the image, classifying the detected region of interest, and outputting a first classification result; and
subjecting the region of interest to semantic segmentation, classifying the region of interest based on a result of the semantic segmentation, and outputting a second classification result,
wherein when a predetermined condition related to a shape of the region of interest identified based on pixels classified by the semantic segmentation as belonging to the region of interest is met, the first classification result is employed as indicating a class of the region of interest, and, when the condition is not met, the second classification result is employed as indicating the class of the region of interest;
the condition requires that a ratio of a minor diameter to a major diameter of an ellipse obtained by subjecting a set of pixels classified by the semantic segmentation as belonging to the region of interest to elliptical approximation be equal to or greater than a predetermined threshold value and that an acute angle formed by a major axis of the ellipse and a longitudinal direction of the image be within a predetermined angular range.
4. A diagnostic imaging method executed by a processor comprising hardware, the method comprising:
receiving an input of an image;
detecting a region of interest including a subject of interest from the image, classifying the detected region of interest, and outputting a first classification result; and
subjecting the region of interest to semantic segmentation, classifying the region of interest based on a result of the semantic segmentation, and outputting a second classification result,
wherein when a predetermined condition related to a shape of the region of interest identified based on pixels classified by the semantic segmentation as belonging to the region of interest is met, the first classification result is employed as indicating a class of the region of interest, and, when the condition is not met, the second classification result is employed as indicating the class of the region of interest,
the condition requires that a ratio of a minor diameter to a major diameter of an ellipse obtained by subjecting a set of pixels classified by the semantic segmentation as belonging to the region of interest to elliptical approximation be equal to or greater than a predetermined threshold value and that an acute angle formed by a major axis of the ellipse and a longitudinal direction of the image be within a predetermined angular range.