US 12,469,129 B2
Prediction device, prediction system, control method, and control program
Kenichi Watanabe, Tokyo (JP); Masayuki Kyomoto, Ritto (JP); Masahiko Hashida, Toyonaka (JP); Shintaro Honda, Tokyo (JP); Takashi Sasaki, Chigasaki (JP); Toru Moro, Tokyo (JP); and Sakae Tanaka, Tokyo (JP)
Assigned to KYOCERA Corporation, Kyoto (JP); and The University of Tokyo, Tokyo (JP)
Appl. No. 17/911,066
Filed by KYOCERA Corporation, Kyoto (JP); and THE UNIVERSITY OF TOKYO, Tokyo (JP)
PCT Filed Mar. 12, 2021, PCT No. PCT/JP2021/009963
§ 371(c)(1), (2) Date Sep. 12, 2022,
PCT Pub. No. WO2021/182595, PCT Pub. Date Sep. 16, 2021.
Claims priority of application No. 2020-044699 (JP), filed on Mar. 13, 2020.
Prior Publication US 2023/0111852 A1, Apr. 13, 2023
Int. Cl. G06T 7/00 (2017.01); G06V 10/82 (2022.01)
CPC G06T 7/0012 (2013.01) [G06V 10/82 (2022.01); G06T 2207/10081 (2013.01); G06T 2207/10088 (2013.01); G06T 2207/10132 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A prediction system for predicting a disease of a joint of a subject comprising:
an image acquisition device configured to acquire a medical image showing a joint of a subject at a first point in time;
a first prediction unit configured to output first prediction information regarding the joint at a second point in time after a predetermined period of time from the first point in time, wherein the first prediction unit comprises a first prediction model configured to estimate the first prediction information from the medical image, the first prediction unit further configured to output the first prediction information to a display of a terminal device; and
a second prediction unit configured to output third prediction information indicating a method of intervention in the subject and an effect of the intervention, wherein the second prediction unit comprises a second prediction model configured to estimate the third prediction information from the first prediction information, the second prediction unit further configured to output the third prediction information to the display of the terminal device;
wherein the second prediction unit comprises a neural network trained using effect information as the second prediction model, the effect information being time series data regarding a joint-related symptom acquired from a plurality of individual patients to whom the interventions were applied at a plurality of past points in time.