Medical diagnostic apparatus, method for operating medical diagnostic apparatus, and computer-readable recording medium

Patent metadata

Field	Value
Publication number	US-9655593-B2
Application number	US-201615180569-A
Country	US
Kind code	B2
Filing date	Jun 13, 2016
Priority date	Jun 11, 2014
Publication date	May 23, 2017
Grant date	May 23, 2017

Abstract

A medical diagnostic apparatus includes: a feature calculating unit configured to calculate a plurality of kinds of features that is based on luminance information or frequency spectrum information obtained from a received signal from a specimen, and that relates to attributes of a tissue of the specimen; a classification unit configured to: classify the attributes of the tissue of the specimen by using a feature determined among the plurality of kinds of features calculated by the feature calculating unit according to a classification item selected in advance from a plurality of classification items assigned for each classification purpose; and assign visual information according to a classification result to pixels of an image based on the received signal; and a feature image data generating unit configured to generate feature image data on which the visual information assigned to the pixels of the image based on the received signal is superimposed.

First claim

What is claimed is:

1. A medical diagnostic apparatus comprising:
a processor comprising hardware, wherein the processor is configured to:
perform a frequency analyzing process comprising analyzing a frequency of a received signal from a specimen to calculate a frequency spectrum;
perform a feature calculating process comprising calculating a plurality of kinds of features that is based on frequency spectrum information obtained from the received signal from the specimen, the plurality of kinds of features relating to attributes of a tissue of the specimen, the attributes of the tissue comprising at least two of a malignant tumor tissue, a benign tumor tissue, an endocrine tumor tissue, a mucinous tumor tissue, a normal tissue, and a vessel;
perform a classification process comprising:
classifying the attributes of the tissue of the specimen by using a feature determined among the plurality of kinds of features calculated in the feature calculating process according to a classification item selected in advance from a plurality of classification items assigned for each classification purpose, the plurality of classification items comprising at least two of tumor screening, distinction between malignancy and benignancy, and follow-up determination; and
assigning visual information according to a classification result to pixels of an image based on the received signal; and

perform a feature image data generating process comprising generating feature image data on which the visual information assigned to the pixels of the image based on the received signal is superimposed,
wherein the feature calculating process comprises an approximation process comprising:
approximating, by a linear expression, the frequency spectrum calculated by the frequency analyzing process; and
extracting, as the plurality of kinds of features, at least one of a slope of the linear expression an intercept of the linear expression, and an intensity defined by using the slope, the intercept, and a specific frequency in a frequency band of the frequency spectrum.

2. The medical diagnostic apparatus according to claim 1 , further comprising a classification information storage configured to store classification information including at least one of an attribute of a tissue to be classified corresponding to the classification item, a kind of a feature used for classification, and information associating visual information according to a value of the feature,
wherein the classification process comprises performing classification and assignment of the visual information by referring to the classification information storage.

3. The medical diagnostic apparatus according to claim 2 , further comprising:
a known specimen information storage configured to store known specimen information including at least one of a received signal from a specimen whose attribute is known and information calculated based on the received signal such that the known specimen information is associated with the attribute,
wherein the processor is configured to perform a classification information setting process comprising setting the classification information by using the known specimen information stored in the known specimen information storage.

4. The medical diagnostic apparatus according to claim 1 ,
wherein the feature calculating process comprises:
extracting a plurality of parameters based on the received signal from a predetermined area of the specimen; and
calculating the features by using the plurality of parameters.

5. The medical diagnostic apparatus according to claim 4 ,
wherein the feature calculating process comprises calculating, as the features, a statistic of parameters of same type among the plurality of parameters.

6. The medical diagnostic apparatus according to claim 1 ,
wherein the visual information is a variable constituting a color space.

7. The medical diagnostic apparatus according to claim 1 ,
wherein the visual information is luminance.

8. The medical diagnostic apparatus according to claim 1 , further comprising a display configured to display a feature image corresponding to the feature image data generated by the feature image data generating process.

9. The medical diagnostic apparatus according to claim 1 , further comprising an input device configured to receive an input for selecting the classification item.

10. The medical diagnostic apparatus according to claim 1 , further comprising an ultrasound probe configured to transmit an ultrasound wave to the specimen and to receive, as the received signal, an electric echo signal acquired by converting an ultrasound echo reflected from the specimen.

11. The medical diagnostic apparatus according to claim 10 ,
wherein the frequency analyzing process comprises analyzing the frequency of the echo signal to calculate the frequency spectrum, and
wherein the feature calculating process comprises calculating the plurality of kinds of features by using the frequency spectrum.

12. The medical diagnostic apparatus according to claim 1 ,
wherein the feature calculating process comprises calculating, as the plurality of kinds of features, texture features obtained by a texture analysis from luminance of each of the pixels of the image based on the received signal.

13. A method for operating a medical diagnostic apparatus that generates diagnostic image data based on a received signal from a specimen, the method comprising:
analyzing, by a frequency analyzing process, a frequency of the received signal from the specimen to calculate a frequency spectrum;
calculating, by a feature calculating process, a plurality of kinds of features in frequency spectrum information obtained from the received signal;
by a classification process, classifying attributes of a tissue of the specimen by using a feature determined among the plurality of kinds of features according to a classification item selected in advance from a plurality of classification items assigned for each classification purpose, and assigning visual information according to a classification result to pixels of an image based on the received signal, the attributes of the tissue comprising at least two of a malignant tumor tissue, a benign tumor tissue, an endocrine tumor tissue, a mucinous tumor tissue, a normal tissue, and a vessel, and the plurality of classification items comprising at least two of tumor screening, distinction between malignancy and benignancy, and follow-up determination; and
generating, by a feature image data generating process, feature image data on which the visual information assigned to the pixels of the image based on the received signal is superimposed,
wherein calculating the plurality of kinds of features comprises:
approximating, by a linear expression, the frequency spectrum calculated by the frequency analyzing process; and
extracting, as the plurality of kinds of features, at least one of a slope of the linear expression, an intercept of the linear expression, and an intensity defined by using the slope, the intercept, and a specific frequency in a frequency band of the frequency spectrum.

14. A non-transitory computer-readable recording medium with an executable program stored thereon, the program instructing a medical diagnostic apparatus that generates diagnostic image data based on a received signal from a specimen, to execute:
analyzing, by a frequency analyzing process, a frequency of the received signal from the specimen to calculate a frequency spectrum;
calculating, by a feature calculating process, a plurality of kinds of features in frequency spectrum information obtained from the rec

Assignees

• Olympus Corp

Inventors

• Kawashima Tomonao

Classifications

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06F18/24

  Classification techniques · CPC title

• A61B8/5223

  for extracting a diagnostic or physiological parameter from medical diagnostic data (for algorithms to analyse biomedical images G06T7/0012) · CPC title

• A61B8/14

  Echo-tomography · CPC title

• A61B8/467

  characterised by special input means · CPC title