Method and apparatus for automated detection of target structures from medical images using a 3D morphological matching algorithm
First Claim
1. A device for detecting whether pulmonary nodules are present in a patient'"'"'s lung region from a three-dimensional (3D) data set representative of a volumetric image of the patient'"'"'s lung region, the device comprising:
- a processor configured to (1) identify contiguous structures in the 3D data set, (2) classify the identified contiguous structures according to a plurality of classifications, the classifications comprising a vessel contiguous structure classification and a non-vessel contiguous structure classification, wherein each non-vessel contiguous structure comprises a nodule candidate, (3) apply a first nodule detection operation to each vessel contiguous structure to determine a nodule status therefor, and (4) apply a second nodule detection operation to each non-vessel contiguous structure to determine a nodule status therefor, wherein the first nodule detection operation is different than the second nodule detection algorithm;
wherein the processor is further configured to apply the first nodule detection operation by (1) segmenting nodule candidate structures from surrounding vessel structures through a correlation of each vessel contiguous structure with a 3D morphological filter, and (2) determining a nodule status for each segmented nodule candidate;
wherein the processor is further configured to apply the second nodule detection operation by determining a nodule status for each non-vessel nodule candidate at least partially on the basis of geometric criteria;
wherein the processor is further configured to determine the nodule status for each non-vessel nodule candidate by comparing each nodule candidate with a compactness criteria; and
wherein the processor is further configured to compare each non-vessel nodule candidate with the compactness criteria by;
for each non-vessel nodule candidate, (1) determining a volume of that non-vessel nodule candidate, (2) determining a volume of the smallest 3D box that encloses that non-vessel nodule candidate, and (3) computing a ratio of the determined non-vessel nodule candidate volume to the determined box volume; and
for each non-vessel nodule candidate having a determined volume ratio less than approximately 0.5 or greater than approximately 1.5, determining that that non-vessel nodule candidate is not a pulmonary nodule.
0 Assignments
0 Petitions
Accused Products
Abstract
A method for the automated detection of target structures shown in digital medical images, the method of comprising: (1) generating a three dimensional (3D) volumetric data set of a patient region within which the target structure resides from a plurality of segmented medical image slices; (2) grouping contiguous structures that are depicted in the 3D volumetric data set to create corresponding grouped structure data sets; (3) assigning each grouped structure data set to one of a plurality of detection algorithms, each detection algorithm being configured to detect a different type of target structure; and (4) processing each grouped structure data set according to its assigned detection algorithm to thereby detect whether any target structures are present in the medical images. Preferably, the target structures are pulmonary nodules, and a specialized detection algorithm is applied to image data classified as a candidate for depicting perivascular nodules. To segment perivascular nodule candidates from surrounding vessels, the image data is preferably correlated with a plurality of 3D morphological filters.
-
Citations
68 Claims
-
1. A device for detecting whether pulmonary nodules are present in a patient'"'"'s lung region from a three-dimensional (3D) data set representative of a volumetric image of the patient'"'"'s lung region, the device comprising:
-
a processor configured to (1) identify contiguous structures in the 3D data set, (2) classify the identified contiguous structures according to a plurality of classifications, the classifications comprising a vessel contiguous structure classification and a non-vessel contiguous structure classification, wherein each non-vessel contiguous structure comprises a nodule candidate, (3) apply a first nodule detection operation to each vessel contiguous structure to determine a nodule status therefor, and (4) apply a second nodule detection operation to each non-vessel contiguous structure to determine a nodule status therefor, wherein the first nodule detection operation is different than the second nodule detection algorithm; wherein the processor is further configured to apply the first nodule detection operation by (1) segmenting nodule candidate structures from surrounding vessel structures through a correlation of each vessel contiguous structure with a 3D morphological filter, and (2) determining a nodule status for each segmented nodule candidate; wherein the processor is further configured to apply the second nodule detection operation by determining a nodule status for each non-vessel nodule candidate at least partially on the basis of geometric criteria; wherein the processor is further configured to determine the nodule status for each non-vessel nodule candidate by comparing each nodule candidate with a compactness criteria; and wherein the processor is further configured to compare each non-vessel nodule candidate with the compactness criteria by; for each non-vessel nodule candidate, (1) determining a volume of that non-vessel nodule candidate, (2) determining a volume of the smallest 3D box that encloses that non-vessel nodule candidate, and (3) computing a ratio of the determined non-vessel nodule candidate volume to the determined box volume; and for each non-vessel nodule candidate having a determined volume ratio less than approximately 0.5 or greater than approximately 1.5, determining that that non-vessel nodule candidate is not a pulmonary nodule. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20)
-
-
21. A device for detecting whether pulmonary nodules are present in a patient'"'"'s lung region from a three-dimensional (3D) data set representative of a volumetric image of the patient'"'"'s lung region, the device comprising:
-
a processor configured to (1) identify contiguous structures in the 3D data set, (2) classify the identified contiguous structures according to a plurality of classifications, the classifications comprising a vessel contiguous structure classification and a non-vessel contiguous structure classification, wherein each non-vessel contiguous structure comprises a nodule candidate, (3) apply a first nodule detection operation to each vessel contiguous structure to determine a nodule status therefor, and (4) apply a second nodule detection operation to each non-vessel contiguous structure to determine a nodule status therefor, wherein the first nodule detection operation is different than the second nodule detection algorithm; wherein the processor is further configured to apply the first nodule detection operation by (1) segmenting nodule candidate structures from surrounding vessel structures through a correlation of each vessel contiguous structure with a 3D morphological filter, and (2) determining a nodule status for each segmented nodule candidate; wherein the processor is further configured to apply the second nodule detection operation by determining a nodule status for each non-vessel nodule candidate at least partially on the basis of geometric criteria; wherein the processor is further configured to determine the nodule status for each non-vessel nodule candidate by comparing each nodule candidate with an elongation criteria; and wherein the processor is further configured to compare each non-vessel nodule candidate with the elongation criteria by; for each non-vessel nodule candidate, (1) determining a length of a major axis of the smallest rectangle or ellipse that encloses that non-vessel nodule candidate, (2) determining a length of a minor axis of the smallest rectangle or ellipse that encloses that non-vessel nodule candidate, (3) computing a ratio of the major axis length to the minor axis length; and for each non-vessel nodule candidate having a determined elongation axis ratio greater than approximately 3.0, determining that that non-vessel nodule candidate is not a pulmonary nodule. - View Dependent Claims (22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38)
-
-
39. A device for detecting whether pulmonary nodules are present in a patient'"'"'s lung region from a three-dimensional (3D) data set representative of a volumetric image of the patient'"'"'s lung region, the device comprising:
-
a processor configured to (1) identify contiguous structures in the 3D data set, (2) classify the identified contiguous structures according to a plurality of classifications, the classifications comprising a vessel contiguous structure classification and a non-vessel contiguous structure classification, wherein each non-vessel contiguous structure comprises a nodule candidate, (3) apply a first nodule detection operation to each vessel contiguous structure to determine a nodule status therefor, and (4) apply a second nodule detection operation to each non-vessel contiguous structure to determine a nodule status therefor, wherein the first nodule detection operation is different than the second nodule detection algorithm; wherein the processor is further configured to apply the first nodule detection operation by (1) segmenting nodule candidate structures from surrounding vessel structures through a correlation of each vessel contiguous structure with a 3D morphological filter, and (2) determining a nodule status for each segmented nodule candidate wherein the processor is further configured to apply the second nodule detection operation by determining a nodule status for each non-vessel nodule candidate at least partially on the basis of geometric criteria; wherein the processor is further configured to determine the nodule status for each non-vessel nodule candidate by comparing each nodule candidate with an elongation criteria; and wherein the processor is further configured to compare each non-vessel nodule candidate with the elongation criteria by; for each non-vessel nodule candidate, (1) determining a maximum eigenvalue from coordinates of the voxels of that non-vessel nodule candidate, (2) determining a minimum eigenvalue from coordinates of the voxels of that non-vessel nodule candidate, (3) computing a ratio of the maximum eigenvalue to the minimum eigenvalue; and for each non-vessel nodule candidate having a determined elongation eigenvalue ratio greater than approximately 3.0, determining that that non-vessel nodule candidate is not a pulmonary nodule. - View Dependent Claims (40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56)
-
-
57. A device for analyzing a 3D data set representative of a patient'"'"'s lung region, the device comprising:
-
a processor configured to (1) group the data set into data subsets, each subset being representative of a contiguous structure, (2) identify each data subset that corresponds to a vessel, and (3) segment any perivascular nodule candidates from each identified subset by correlating that identified subset with at least one 3D morphological filter that is tuned to an expected shape of a perivascular nodule; wherein the processor is further configured to perform the correlation by convolving each identified subset with the at least one filter to thereby compute a correlation value; and wherein the processor is further configured to perform the convolution by computing the correlation value between the identified subset (I) and the filter (F) by a Fast Fourier Transform (FFT) according to the formula; - View Dependent Claims (58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68)
-
Specification