US 11,704,796 B2
Estimating bone mineral density from plain radiograph by assessing bone texture with deep learning
Kang Zheng, Bethesda, MD (US); Yirui Wang, Bethesda, MD (US); Shun Miao, Bethesda, MD (US); Changfu Kuo, Taiwan (CN); and Chen-I Hsieh, Taiwan (CN)
Assigned to PING AN TECHNOLOGY (SHENZHEN) CO., LTD., Shenzhen (CN)
Filed by Ping An Technology (Shenzhen) Co., Ltd., Shenzhen (CN)
Filed on Jan. 5, 2021, as Appl. No. 17/142,187.
Claims priority of provisional application 62/988,713, filed on Mar. 12, 2020.
Claims priority of provisional application 62/988,628, filed on Mar. 12, 2020.
Claims priority of provisional application 62/958,965, filed on Jan. 9, 2020.
Prior Publication US 2021/0212647 A1, Jul. 15, 2021
Int. Cl. G06T 7/00 (2017.01); G06T 7/73 (2017.01); G06N 3/08 (2023.01); A61B 6/00 (2006.01); G06T 7/11 (2017.01); G06N 3/04 (2023.01); G16H 30/40 (2018.01); G16H 50/30 (2018.01); G06F 18/213 (2023.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 10/44 (2022.01)
CPC G06T 7/0012 (2013.01) [A61B 6/469 (2013.01); A61B 6/505 (2013.01); A61B 6/5217 (2013.01); G06F 18/213 (2023.01); G06N 3/04 (2013.01); G06N 3/08 (2013.01); G06T 7/11 (2017.01); G06T 7/73 (2017.01); G06V 10/44 (2022.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G16H 30/40 (2018.01); G16H 50/30 (2018.01); G06T 2207/10116 (2013.01); G06T 2207/20072 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30012 (2013.01); G06T 2207/30052 (2013.01); G06T 2207/30168 (2013.01); G06V 2201/033 (2022.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method for radiographic bone mineral density (BMD) estimation, the method comprising:
receiving a plain radiograph;
detecting landmarks for a bone structure included in the plain radiograph;
extracting a region of interest (ROI) from the plain radiograph based on the detected landmarks, comprising:
detecting anatomical landmarks at a center of the femur head when detecting the landmarks included in the plain radiograph for the bone structure;
determining a size of the ROI relevant to a distance between two femur heads, wherein the ROI is a rectangular region containing a femur head, a greater trochanter, and a lesser trochanter, segmented from the plain radiograph; and
determining a center of the ROI based on locations of the femur heads; and
estimating the BMD for the ROI extracted from the plain radiograph by using a deep neural network.