Medical imaging data processing apparatus and method

• US 9,773,325 B2
• Filed: 04/02/2015
• Issued: 09/26/2017
• Est. Priority Date: 04/02/2015
• Status: Active Grant

• Current Assignee: Toshiba Medical Systems Corporation (Canon Inc.)
• Original Assignee: Toshiba Medical Systems Corporation (Canon Inc.)
• Inventors: Plakas, Costas, Razeto, Marco
• Primary Examiner(s): Yentrapati, Avinash
• Application Number:
  US14/677,007
• Publication Number:
  US 20160292848A1
• Time in Patent Office: 908 Days
• Field of Search: None
• CPC Class Codes: A61B 8/085 for locating body or organic structures, e.g. tumours, calculi, blood vessels, nodules
• CPC Class Codes: A61B 8/5207 involving processing of raw data to produce diagnostic data, e.g. for generating an image reconstruction G06T11/00
• CPC Class Codes: A61B 8/5223 for extracting a diagnostic or physiological parameter from medical diagnostic data for algorithms to analyse biomedical images G06T7/0012
• CPC Class Codes: G06T 11/003 Reconstruction from projections, e.g. tomography
• CPC Class Codes: G06T 2207/10132 Ultrasound image
• CPC Class Codes: G06T 2207/20156 Automatic seed setting
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 2207/30242 Counting objects in image
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment

Abstract

A medical imaging data processing apparatus comprises a setting circuitry configured to set a plurality of seeds at different locations in medical image data, a processing circuitry configured to select at least one seed from among the plurality of seeds and expand the at least one selected seed, and a region identifying circuitry configured to identify at least one target region based on a result of the expansion.

Claims

1. A medical imaging data processing apparatus, comprising:

setting circuitry configured to set a plurality of seeds at different locations in medical image data by distributing the plurality of seeds pseudorandomly or regularly;

processing circuitry configured to

estimate, from at least part of the medical image data, a first statistical distribution for pixels or voxels of a target class and at least one further statistical distribution for pixels or voxels of at least one further class;

select at least one seed from among the plurality of seeds by determining whether each seed belongs to the target class in dependence on the estimated first statistical distribution and the estimated at least one further distribution, and selecting seeds that are determined to belong to the target class, and

expand the at least one selected seed; and

region identifying circuitry configured to identify at least one target region based on a result of the expansion, wherein each identified target region comprises tissue belonging to the target class.

2. The apparatus according to claim 1, wherein

the processing circuitry is further configured to obtain at least one candidate region as a result of the expansion;

and

the region identifying circuitry is further configured to identify the at least one target region based on the at least one candidate region.

3. The apparatus according to claim 1, wherein each identified target region comprises a desired anatomical structure.

4. The apparatus according to claim 1, wherein each obtained candidate region and each identified target region comprises tissue belonging to the target class.

5. The apparatus according to claim 4, wherein each identified target region comprises an ovarian follicle, and the target class comprises follicle tissue.

6. The apparatus according to claim 3, wherein at least one of:

the region identifying circuitry is further configured to automatically measure each identified target region, thereby to obtain a measurement of the desired anatomical structure;

the region identifying circuitry is further configured to automatically count each identified target region to count the desired anatomical structure or structures; and

the region identifying circuitry is further configured to automatically determine a position of each identified target region.

7. The apparatus according to claim 4, wherein expanding the at least one selected seed comprises expanding each selected seed until a boundary between the target class and at least one further class is reached.

8. The apparatus according to claim 4, wherein expanding the at least one selected seed comprises expanding each seed using at least one of a level-set algorithm and a region growing algorithm.

9. The apparatus according to claim 8, wherein the level-set or region growing algorithm is configured to distinguish between the target class and at least one further class based on intensity.

10. The apparatus according to claim 2, wherein at least one of:

identifying the at least one target region comprises combining two or more adjacent or overlapping candidate regions to form a single candidate region;

identifying the at least one target region comprises splitting one candidate region into two or more candidate regions; and

identifying the at least one target region comprises selecting at least one candidate region in dependence on candidate region size and shape.

11. The apparatus according to claim 4, wherein the processing circuitry is further configured to distinguish between the target class and at least one further class.

12. The apparatus according to claim 11, wherein the first statistical distribution and the at least one further statistical distribution each comprise a distribution of at least one of intensity, gradient, shape of the region grown, local gradient, texture appearance indicators, and higher order derivatives.

13. The apparatus according to claim 4, wherein the target class comprises follicle tissue, and wherein the processing circuitry is further configured to distinguish between follicle tissue, tissue between follicles, and tissue outside follicles.

14. The apparatus according to claim 1, further comprising pre-processing circuitry configured to at least one of:

identify an area of interest in the medical image data; and

filter at least part of the medical image data using at least one of a noise-reducing filter, an edge-preserving filter, a median filter, and an anisotropic filter.

15. The apparatus according to claim 1, wherein each seed comprises at least one of a point, a pixel, a voxel, a two-dimensional seed region of the area of interest, and a three-dimensional seed region of the area of interest.

16. The apparatus according to claim 1, wherein the medical image data comprises at least one of ultrasound data, CT data, cone-beam CT data, MR data, X-ray data, PET data, and SPECT data.

17. A medical imaging data processing method, comprising:

setting a plurality of seeds at different locations in medical image data by distributing the plurality of seeds pseudorandomly or regularly;

estimating, from at least part of the medical image data, a first statistical distribution for pixels or voxels of a target class and at least one further statistical distribution for pixels or voxels of at least one further class;

selecting at least one seed from among the plurality of seeds by determining whether each seed belongs to the target class in dependence on the estimated first statistical distribution and the estimated at least one further distribution, and selecting seeds that are determined to belong to the target class;

expanding the at least one selected seed; and

identifying at least one target region based on the result of the expansion, wherein each identified target region comprises tissue belonging to the target class.

18. A non-transitory computer-readable storage medium storing a computer program comprising computer-readable instructions that are executable to perform a method comprising:

setting a plurality of seeds at different locations in medical image data by distributing the plurality of seeds pseudorandomly or regularly;

estimating, from at least part of the medical image data, a first statistical distribution for pixels or voxels of a target class and at least one further statistical distribution for pixels or voxels of at least one further class;

selecting at least one seed from among the plurality of seeds by determining whether each seed belongs to the target class in dependence on the estimated first statistical distribution and the estimated at least one further distribution, and selecting seeds that are determined to belong to the target class;

expanding the at least one selected seed; and

identifying at least one target region based on a result of the expansion, wherein each identified target region comprises tissue belonging to the target class.

19. A medical imaging data processing apparatus, comprising:

setting circuitry configured to set at least one seed in medical image data by distributing the at least one seed pseudorandomly or regularly;

processing circuitry configured to

estimate, from at least part of the medical image data, a first statistical distribution for pixels or voxels of a target class and at least one further statistical distribution for pixels or voxels of at least one further class, the target class being follicles;

select a seed from among the at least one seed by determining whether each seed belongs to the target class in dependence on the estimated first statistical distribution and the estimated at least one further distribution, and selecting a seed that is determined to belong to the target class, and

vary an extent of the selected seed to obtain a candidate follicle region; and

region identifying circuitry configured to process the candidate follicle region to identify at least one follicle.

20. The apparatus according to claim 19, wherein the setting of the at least one seed in the medical image data is in dependence on user input.

21. The apparatus according to claim 19, wherein the region identifying circuitry is further configured to automatically measure each follicle.

22. The apparatus according to claim 19, wherein the region identifying circuitry is further configured to automatically count each follicle.