US 12,118,719 B2
System and method for prediction of disease progression of pulmonary fibrosis using medical images
Hyun Kim, Los Angeles, CA (US); Yu Shi, Fremont, CA (US); Jonathan Gerald Goldin, Oakland, CA (US); and Weng Kee Wong, Oakland, CA (US)
Assigned to The Regents of the University of California, Oakland, CA (US)
Appl. No. 17/639,511
Filed by The Regents of the University of California, Oakland, CA (US)
PCT Filed Sep. 3, 2020, PCT No. PCT/US2020/049099
§ 371(c)(1), (2) Date Mar. 1, 2022,
PCT Pub. No. WO2021/046152, PCT Pub. Date Mar. 11, 2021.
Claims priority of provisional application 62/895,496, filed on Sep. 3, 2019.
Prior Publication US 2022/0327693 A1, Oct. 13, 2022
Int. Cl. G06T 7/00 (2017.01); A61B 5/00 (2006.01); A61B 6/00 (2006.01); A61B 6/03 (2006.01); G06V 10/25 (2022.01); G06V 10/77 (2022.01)
CPC G06T 7/0012 (2013.01) [A61B 5/0037 (2013.01); A61B 5/7267 (2013.01); A61B 5/7275 (2013.01); A61B 6/032 (2013.01); A61B 6/5217 (2013.01); G06V 10/25 (2022.01); G06V 10/7715 (2022.01); G06T 2207/10081 (2013.01); G06T 2207/30101 (2013.01); G06V 2201/03 (2022.01); G06V 2201/07 (2022.01)] 15 Claims
OG exemplary drawing
 
1. A system that classifies predictive region-of-interest (“ROIs”) of progression of idiopathic pulmonary fibrosis, the system comprising:
a processor configured to analyze images of a patient and indicate regions in the images expected to reflect progressive pulmonary fibrosis in the future using a machine learning algorithm;
a display to communicate the image of the patient indicating the regions in the images expected to reflect progressive pulmonary fibrosis in the future;
a memory accessible by the processor and having stored thereon the machine learning algorithm, wherein the machine learning algorithm was trained by:
(a) acquiring a set of computed tomography (CT) images of a plurality of patients;
(b) selecting a plurality of ROIs within the set of images, wherein each of the ROIs designates a label indicating progression of pulmonary fibrosis;
(c) training the machine learning algorithm by inputting the plurality of ROIs and the associated labels into the machine learning algorithm, wherein the machine learning algorithm identifies the ROIs in the set of images as indicating regions of pulmonary fibrosis within the set of images based on the features; and
(d) generating classification probabilities from the output of the machine learning algorithm, wherein the classification probabilities classifies regions in the set of images as indicating regions of expected to be progressive pulmonary fibrosis in the future within the set of images.