US 11,810,292 B2
Disease characterization and response estimation through spatially-invoked radiomics and deep learning fusion
Anant Madabhushi, Shaker Heights, OH (US); Nathaniel Braman, Bethel Park, PA (US); and Jeffrey Eben, Mayfield Village, OH (US)
Assigned to Case Western Reserve University, Cleveland, OH (US)
Filed by Case Western Reserve University, Cleveland, OH (US)
Filed on Sep. 30, 2020, as Appl. No. 17/038,934.
Claims priority of provisional application 62/908,072, filed on Sep. 30, 2019.
Prior Publication US 2021/0097682 A1, Apr. 1, 2021
Int. Cl. G06T 7/00 (2017.01); G06T 7/11 (2017.01); G06V 10/764 (2022.01); G06V 10/771 (2022.01); G06V 10/774 (2022.01); G06V 10/80 (2022.01); G06V 10/82 (2022.01); G06N 20/10 (2019.01); G06F 17/16 (2006.01); G06V 10/20 (2022.01); G06V 10/776 (2022.01)
CPC G06T 7/0012 (2013.01) [G06F 17/16 (2013.01); G06N 20/10 (2019.01); G06T 7/11 (2017.01); G06V 10/255 (2022.01); G06V 10/764 (2022.01); G06V 10/771 (2022.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 10/809 (2022.01); G06V 10/82 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30096 (2013.01); G06V 2201/03 (2022.01)] 32 Claims
OG exemplary drawing
 
32. A method, comprising:
extracting a first set of radiomic features from a tumor segmented in a medical imaging scan and a second set of radiomic features from a segmented peri-tumoral region around the segmented tumor;
providing the first set of radiomic features to a first machine learning model and the second set of radiomic features to a second machine learning model;
providing the segmented tumor to a first deep learning model and the segmented peri-tumoral region to a second deep learning model;
receiving a first predicted prognosis for the tumor from the first machine learning model, a second predicted prognosis for the tumor from the second machine learning model, a third predicted prognosis for the tumor from the first deep learning model, and a fourth predicted prognosis for the tumor from the second deep learning model;
providing the first predicted prognosis, the second predicted prognosis, the third predicted prognosis, and the fourth predicted prognosis to a combination learning model; and
receiving a combined predicted prognosis for the tumor from the combination learning model.