System and method for detecting central pulmonary embolism in CT pulmonary angiography images
A system and method for detecting central pulmonary embolisms in a subject's vasculature is provided. In some aspects, the method includes receiving, using the input, a set of images representing a vasculature of the subject's lungs, automatically analyzing the set of images to segment the main arteries associated with the subject's lungs and separate the main arteries from surrounding tissues. The method also includes automatically extracting central pulmonary embolism candidates from the set of images after segmenting and separating the main arteries, and automatically evaluating the central pulmonary embolism candidates in three-dimensional (3D) space by applying a series of rules. The method further includes automatically displaying a report indicating evaluated central pulmonary embolism candidates on a display.
1. A system for detecting pulmonary embolisms in a subject's vasculature, the system comprising:
an input configured to receive images acquired from a subject;
a hardware processor configured to process the images by at least:
i. receiving, using the input, a set of images representing a vasculature of the subject's lungs;
ii. analyzing the set of images to segment the main arteries associated with the subject's lungs and separate the main arteries from surrounding tissues;
iii. extracting pulmonary embolism candidates from the set of images after segmenting and separating the main; arteries by:
tracing a pulmonary trunk in the set of images in a cranial to caudal direction;
summing intensity values of binary image pixels based on the set of images to form a sum;
comparing the sum to a threshold; and
identifying a hypo-attenuating focus when the sum is greater than the threshold;
iv. evaluating the pulmonary embolism candidates in three-dimensional (3D) space; and
v. generating a report indicating evaluated candidate pulmonary embolisms; and
an output for displaying the report.
2. The system of claim 1 , wherein the set of images comprises computed tomography pulmonary angiography (CTPA) images.
3. The system of claim 1 , wherein the set of images comprises three-dimensional (3D) images.
4. The system of claim 1 , wherein processor is further caused to trace a right main artery from one side of the right main artery to another side of the right main artery and compare edges of the right main artery.
5. The system of claim 1 , wherein the processor is further configured to apply a logical operator to identify pieces that only appear in one of the set of images within the series of images and meet a threshold criteria and to identify these pieces meeting the threshold criteria as candidate pulmonary embolisms.
6. The system of claim 5 , wherein the threshold criteria include a size of more than 2 voxels in height and width or having an image intensity of between −100 HU and 200 HU.
7. A method for detecting central pulmonary embolisms in a subject's vasculature, the method comprising:
receiving, using the input, a set of images representing a vasculature of the subject's lungs;
automatically analyzing the set of images to segment the main arteries associated with the subject's lungs and separate the main arteries from surrounding tissues;
automatically extracting central pulmonary embolism candidates from the set of images after segmenting and separating the main arteries by:
automatically tracing a pulmonary trunk in the set of images in a cranial to caudal direction, summing intensity values of binary image pixels based on the set of images to form a sum, comparing the sum to a threshold;
identifying a hypo-attenuating focus when the sum is greater than the threshold;
automatically evaluating the central pulmonary embolism candidates in three-dimensional (3D) space by applying a series of rules; and
automatically displaying a report indicating evaluated central pulmonary embolism candidates on a display.
8. The method of claim 7 , wherein the set of images comprises computed tomography pulmonary angiography (CTPA) images.
9. The method of claim 7 , wherein the set of images comprises three-dimensional (3D) images.
10. The method of claim 7 , further comprising automatically tracing a right main artery from one side of the right main artery to another side of the right main artery and compare edges of the right main artery.
11. The method of claim 7 , further comprising automatically applying a logical operator to identify pieces that only appear in one of the set of images within the series of images and meet a threshold criteria and identifying these pieces meeting the threshold criteria as candidate pulmonary embolisms.
12. The method of claim 11 , wherein the threshold criteria include a size of more than 2 voxels in height and width or having an image intensity of between −100 HU and 200 HU.