Image diagnosis support apparatus, medical image acquisition apparatus, and computer readable recording medium

Abstract

The invention provides an image diagnosis support apparatus, a medical image acquisition apparatus, and a computer readable recording medium. The most appropriate image for a diagnostic target among a plurality of images is selected and accurate diagnosis support information is presented regardless of the type of a selected image, a modality, or the like. The image diagnosis support apparatus includes: a diagnostic information generation unit that generates diagnostic information based on a plurality of medical images; a reliability calculation unit that evaluates an image quality and calculates an image reliability for each of the plurality of medical images; and a degree-of-contribution calculation unit that calculates a degree of contribution of each of the plurality of medical images to the diagnostic information using an internal parameter indicating a degree of appropriateness of each medical image for a diagnostic target and the reliability calculated by the reliability calculation unit. An image for detection used by the diagnostic information generation unit is generated based on the degree of contribution.

Description

technical field [0001] The present invention relates to a technique for assisting diagnosis using images acquired by medical image acquisition devices such as magnetic resonance imaging devices (hereinafter referred to as MRI devices), X-ray imaging devices, and CT devices. Background technique [0002] In the medical field, image diagnosis for diagnosing diseases and lesions based on images captured by MRI apparatuses and CT apparatuses is widely performed. In recent years, a diagnostic support technology has also been developed that uses AI obtained by machine learning to determine the presence or absence of lesions, the degree of progression, or the degree of malignancy in measured images (Non-Patent Document 1, etc.). [0003] In support of image diagnosis using machine learning, AI is obtained by learning the relationship between them and images for each type of disease and each part, and one or more measurement images are used as input to determine the presence or abse...

AI Technical Summary

Problems solved by technology

[0006] In image diagnosis, when measuring multiple images of the same subject and selecting and using one or more images that are most suitable for the object of diagnosis from these multiple images, due to the influence of body movement, etc. or according to shooting conditions etc., artifacts may occur in parts or areas that should be diagnosed in the image that should be selected, making it impossible to use for diagnosis

In this case, although other images are considered as alternatives, their selection is not easy

This problem also exists even for images obtained in different modalities of image types, such as MR images, CT images, X-ray images, etc., and the technology described in Patent Document 1 cannot solve it

[0007] In addition, even if a predetermined image is selected, the contrast of the image will vary depending on the manufacturer (supplier) of the medical image acquisition device and the imaging conditions. For example, in the case of an MRI apparatus, the contrast of the image will vary depending on the magnetic field strength and imaging parameters. , the selected image may not have the same contrast as the image used to generate the learned AI, so when using the existing AI, it is impossible to perform image diagnosis assistance with high accuracy

Images