US 12,339,932 B2
Classification system
Shinkichi Himeno, Yame-gun (JP)
Assigned to IRYOU JYOUHOU GIJYUTU KENKYUSHO CORPORATION, Yame-gun (JP)
Filed by IRYOU JYOUHOU GIJYUTU KENKYUSHO CORPORATION, Fukuoka (JP)
Filed on Aug. 31, 2021, as Appl. No. 17/462,703.
Application 17/462,703 is a continuation in part of application No. PCT/JP2020/008822, filed on Mar. 3, 2020.
Claims priority of application No. 2019-039332 (JP), filed on Mar. 5, 2019.
Prior Publication US 2021/0397905 A1, Dec. 23, 2021
Int. Cl. G06F 18/243 (2023.01); G06F 16/55 (2019.01); G06F 18/21 (2023.01); G06F 18/2413 (2023.01)
CPC G06F 18/24323 (2023.01) [G06F 16/55 (2019.01); G06F 18/217 (2023.01); G06F 18/2413 (2023.01)] 7 Claims
OG exemplary drawing
 
1. A classification system for estimating a diagnosis name from at least one symptom and/or finding comprising:
at least one memory storing instructions to perform operations; and
at least one processor configured to execute the instructions, the operations comprising:
(i) recording a plurality of candidate classes and class attributes for each of the plurality of candidate classes;
(ii) in a learning phase of the classification system, learning pairs of
(a) symptoms/findings and
(b) their correct candidate classes, which are diagnosis names;
(iii) in an operation phase of the classification system, acquiring symptoms/findings as the class attributes comprising attributes of an object to be classified;
(iv) in the operation phase of the classification system, through a classification using the class attributes acquired in (iii) as an input, obtaining a diagnosis name list, consisting of a list of a portion of the candidate classes selected from the plurality of candidate classes that is less than all of the plurality of candidate classes in (i); and
(v) checking the appropriateness of the classification made in (iv) by comparing the class attributes of an individual candidate class in the diagnosis name list with the class attributes in (i) corresponding to the individual candidate class.