Automated image system for scoring changes in quantitative interstitial lung disease
Automated image analysis systems and methods are disclosed to quantify change in fibrosis and interstitial lung disease. The system generates scoring changes in Quantitative Interstitial Lung Disease (QILD) by filtering the uploaded images to minimize cross-site variability within images, sampling from a grid of pixels or voxels within the CT images, classifying individual pixels or voxels within downloaded images based on one or more selected texture features, generating a QILD score for each image based on selected features within the image, and calculating a transition between QILD scores within the plurality of CT images.
1. A method for scoring changes in Quantitative Interstitial Lung Disease (QILD), comprising:
uploading a plurality of CT images of a patient's lung;
filtering the uploaded images to minimize cross-site variability within images;
generating a QILD score for each image based on selected features within the image; and
calculating a transition between QILD scores within the plurality of CT images.
2. A method as recited in claim 1 , further comprising:
sampling from a grid of pixels or voxels within the CT images.
3. A method as recited in claim 2 , further comprising:
classifying individual pixels or voxels within said downloaded images based on one or more selected texture features; and
wherein the QILD score is calculated as the quotient of a total number of pixels or voxels within a CT image that are classified as a disease type by a total number (counts) of grid samples within the CT image.
4. A method as recited in claim 1 , wherein calculating a transition between QILD scores comprises calculating transitional changes of hierarchical severity within the image.
5. A method as recited in claim 4 , wherein the transitional changes of hierarchical severity are calculated using a Markov Chain.
6. A method as recited in claim 2 , wherein classifying individual pixels or voxels is a function of a classifier model built using Support Vector Machine (SVM). programming.
7. A method as recited in claim 1 , wherein the transition between QILD scores is used for estimating a transitional change in one or more of: fibrotic reticulation, ground glass and normal patterns associated with the patient's lung.
8. A method as recited in claim 1 , wherein calculating a transition between QILD scores comprises generating a Markov Chain Transition Matrix (MCTM) of the QILD scores.
9. A method as recited in claim 8 , wherein the MCTM is a function of the fibrotic reticulation, ground glass and normal patterns associated with the patient's lung.
10. A system for scoring changes in Quantitative Interstitial Lung Disease (QILD), comprising:
a processor; and
programming executable on said processor for:
uploading a plurality of CT images of a patient's lung;
filtering the uploaded images to minimize cross-site variability within images;
generating a QILD score for each image based on selected features within the image; and
calculating a transition between QILD scores within the plurality of CT images.
11. A system as recited in claim 10 , further comprising:
sampling from a grid of pixels or voxels within the CT images.
12. A system as recited in claim 11 , said programming further configured for:
classifying individual pixels or voxels within said downloaded images based on one or more selected texture features; and
wherein the QILD score is calculated as the quotient of a total number of pixels or voxels within a CT image that are classified as a disease type by a total number (counts) of grid samples within the CT image.
13. A system as recited in claim 10 , wherein calculating a transition between QILD scores comprises calculating transitional changes of hierarchical severity within the image.
14. A system as recited in claim 13 , wherein the transitional changes of hierarchical severity are calculated using a Markov Chain.
15. A system as recited in claim 11 , wherein classifying individual pixels or voxels is a function of a classifier model built using Support Vector Machine (SVM) programming.
16. A system as recited in claim 10 , wherein the transition between QILD scores is used for estimating a transitional change in one or more of: fibrotic reticulation, ground glass and normal patterns associated with the patient's lung.
17. A system as recited in claim 10 , wherein calculating a transition between QILD scores comprises generating a Markov Chain Transition Matrix (MCTM) of the QILD scores.
18. A system as recited in claim 17 , wherein the MCTM is a function of the fibrotic reticulation, ground glass and normal patterns associated with the patient's lung.
19. A system for scoring changes in Quantitative Interstitial Lung Disease (QILD), comprising:
a processor; and
programming executable on said processor for:
uploading a plurality of CT images of a patient's lung;
filtering the uploaded images to minimize cross-site variability within images;
sampling from a grid of pixels or voxels within the CT images;
classifying individual pixels or voxels within said downloaded images based on one or more selected texture features;
generating a QILD score for each image based on selected features within the image;
wherein the QILD score is calculated as the quotient of a total number of pixels or voxels within a CT image that are classified as a disease type by a total number (counts) of grid samples within the CT image; and
calculating a transition between QILD scores within the plurality of CT images.
20. A system as recited in claim 19 :
wherein calculating a transition between QILD scores comprises calculating transitional changes of hierarchical severity within the image; and
wherein the transitional changes of hierarchical severity are calculated by generating a Markov Chain Transition Matrix (MCTM) of the QILD scores.
21. A system as recited in claim 20 , wherein the MCTM is a function of the fibrotic reticulation, ground glass and normal patterns associated with the patient's lung.
22. A system as recited in claim 19 , wherein filtering comprises filtering noise as a function of geometric, texture, and noised images associated with the CT images.
23. A system as recited in claim 19 , wherein the classified disease types comprise one or more of: fibrotic reticulation, ground glass and normal patterns.
24. A system as recited in claim 23 , wherein the fibrotic reticulation disease type comprises is combined patterns of lung fibrosis and honeycomb.
25. A system as recited in claim 19 , wherein the QILD score is the sum of Quantitative Fibrotic Reticulation (QFR), Quantitative Ground Glass (QGG), and Quantitative Normal Lung (QNL) scores.
26. A system as recited in claim 25 , where the QFR score is the sum of quantitative lung fibrosis and honeycomb.