Method and system for the detection, comparison and volumetric quantification of pulmonary nodules on medical computed tomography scans

• US 7,206,462 B1
• Filed: 07/27/2004
• Issued: 04/17/2007
• Est. Priority Date: 03/17/2000
• Status: Expired due to Fees

• Current Assignee: The General Hospital Corporation (Mass General Brigham, Inc.)
• Original Assignee: The General Hospital Corporation (Mass General Brigham, Inc.)
• Inventors: Ko, Jane P., Betke, Margrit
• Primary Examiner(s): Wu; Jingge
• Assistant Examiner(s): AZARIAN, SEYED H
• Application Number:
  US10/900,031
• Time in Patent Office: 994 Days
• Field of Search: 382/100, 382/128, 382129-134, 382/154, 382/168, 382/170, 382/172, 382/194, 382/203, 382/209, 382/237, 382/256, 382/266, 382/274, 382/305, 382/318, 382/280, 700/182, 600/424, 454/52, 710/120, 378/8, 378/21
• US Class Current: 382/280
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

A method and system for the automated detection and analysis of pulmonary nodules in computed tomographic scans which uses data from computed tomography (CT) scans taken at different times is described. The method and system includes processing CT lung images to identify structures in the images for one patient in each of a plurality of different studies and then relating the plurality of CT lung images taken from the patient in one CT study with a plurality of CT lung images taken from the same patient in at least one other different study to detect changes in structures within the CT images.

Claims

1. A method of processing computed tomographic (CT) images of a lung, the method comprising:

generating a first plurality of CT images of the lung at a first time;

generating a second plurality of CT images of the lung at another time;

selecting a first CT image from among the first plurality of CT images; and

relating the selected first CT image with the second plurality of images of the lung, wherein the relating comprises;

selecting a plurality of predetermined vertical regions of the lung;

identifying a first one of the predetermined vertical regions associated with the selected first CT image; and

selecting a second CT image from among the second plurality of CT images associated with the identified first one of the predetermined vertical regions, and

identifying a first one or more centroids in the selected first CT image and a second one or more centroids in the selected second CT image, wherein the first one or more centroids and the second one or more centroids are associated with at least one of a right lung or a left lung.

2. The method of claim 1, wherein the identifying a first one of the predetermined vertical regions comprises:

identifying, in the selected first CT image, at least one of a first CT image spine region, a first CT image sternum region, or a first CT image trachea region; and

comparing at least one of the first CT image spine region, the first CT image sternum region, or the first CT image trachea region with a respective at least one of a template of a spine region, a template of a sternum region, or a template of a trachea region, wherein the template of the spine region, the template of the sternum region, and the template of the trachea region are associated with the first one of the predetermined vertical regions of the lung.

3. The method of claim 1, wherein the relating the selected first CT image with the second plurality of images her comprises:

aligning the first one or more centroids with the second one or more centrioids.

19. The method of claim 2, wherein the identifying, in the first CT image, the at least one of the first CT image spine region, the first CT image sternum region, or the first CT image trachea region, comprises:

generating at least one histogram of pixel intensities associated with the first CT image projected along a respective at least one axis on the first CT image.

20. The method of claim 2, wherein the selecting the second CT image from among the second plurality of CT images comprises:

identifying, in the selected second CT image, at least one of a second CT image spine region, a second CT image sternum region, or a second CT image trachea region; and

comparing at least one of the second CT image spine region, the second CT image sternum region, or the second CT image trachea region with a respective at least one of the template of the spine region, the template of the sternum region, or the template of the trachea region, wherein the template of the spine region, the template of the sternum region, and the template of the trachea region are associated with the first one of the predetermined vertical regions of the lung.

21. The method of claim 20, wherein the identifying, in the second CT image, the at least one of the second CT image spine region, the second CT image sternum region, or the second CT image trachea region, comprises:

generating at least one histogram of pixel intensities associated with the second CT image projected along a respective at least one axis on the second CT image.

4. A method of processing computed tomographic (CT) images of a lung, the method comprising:

identifying a structure in a first CT image of the lung;

computing a first shape associated with the structure in the first CT image, wherein the first shape is computed as at least one of a first occupancy factor and a first ratio between a major axis of the structure and a minor axis of the structure, wherein the first occupancy factor is indicative of a portion of a predetermined shape occupied by the structure; and

classifying the structure.

5. The method of claim 4, wherein the classifying the structure includes classifying the structure as a selected one of a nodule, a potential nodule, or a vessel in accordance with the first shape.

6. The method of claim 4, wherein the first occupancy factor is a circular occupancy factor indicative of a portion of a first circle occupied by the structure.

7. The method of claim 4, further including:

computing a first volume of the structure in the first CT image; and

classifying the structure in accordance with at least one of the first shape or the first volume of the structure.

8. The method of claim 7, further including:

segmenting the first CT image of the lung into two or more predetermined horizontal regions; and

identifying a first horizontal region in which the structure resides from among the two or more predetermined horizontal regions in the first CT image, wherein the classifying the structure is made in accordance with at least one of the first shape, the first volume of the structure, or the identified first horizontal region.

9. The method of claim 4, further including:

segmenting the first CT image of the lung into two or more predetermined horizontal regions; and

identifying a first horizontal region in which the structure resides from among the two or more predetermined horizontal regions, wherein the classifying the structure is made in accordance with at least one of the first shape or the identified first horizontal region.

10. The method of claim 4, further including:

identifying the structure in a second CT image of the lung; and

computing a second shape associated with the structure in the second CT image, wherein the second shape is computed as at least one of a second occupancy factor and a second ratio between a major axis of the structure and a minor axis of the structure, wherein the second occupancy factor is indicative of a portion of a predetermined shape occupied by the structure.

11. The method of claim 10, wherein the classifying the structure includes classifying the structure as a selected one of a nodule, a potential nodule, or a vessel in accordance with at least one of the first shape or the second shape.

12. The method of claim 10, wherein the second occupancy factor is a circular occupancy factor indicative of a portion of a second circle occupied by the structure.

13. The method of claim 10, further including:

computing a second volume of the structure in the second CT image; and

classifying the structure in accordance with at least one of the first shape, the second shape, the first volume, or the second volume.

14. The method of claim 13, further including:

segmenting the second CT image of the lung into two or more predetermined horizontal regions; and

identifying a second horizontal region in which the structure resides from among the two or more predetermined horizontal regions in the second CT image, wherein the classifying the structure is made in accordance with at least one of the first shape, the second shape, the first volume, the second volume, the first horizontal region, or the second horizontal region.

15. The method of claim 10, further including:

segmenting the second CT image of the lung into two or more predetermined horizontal regions; and

identifying a second horizontal region in which the structure resides from among the two or more predetermined horizontal region;

wherein the classifying the structure is made in accordance with at least one of the first shape, the second shape, the first horizontal region, or the second horizontal region.

16. A method of processing computed tomographic (CT) images of a lung, the method comprising:

identifying a lung boundary in the CT image, wherein the identifying comprises;

applying pixel rules associated with the CT image to provide a preliminary lung boundary, wherein the pixel rules comprise;

identifying a starting pixel in the CT image associated with the lung boundary; and

identifying a next pixel associated with the preliminary lung boundary in accordance with pixel densities of pixels surrounding the starting pixel; and

applying boundary rules to the preliminary lung boundary to provide a final lung boundary.

17. The method of claim 16, wherein the boundary rules comprise:

computing a volume of a lung associated with the preliminary lung boundary; and

splitting the preliminary lung boundary if the volume of the lung is greater than a predetermined lung volume value.

18. The method of claim 17, wherein the boundary rules further comprise:

identifying a curvature associated with the preliminary lung boundary; and

altering the preliminary lung boundary if the curvature of the preliminary lung boundary is greater than a predetermined curvature value.