IP Library Granted Patent US 12,737,409
Granted Patent B2
US 12,737,409 · App. 18/432,410 · Granted Sep 15, 2026

Method and apparatus for automated creation of search patterns

Inventor: Andre De Oliveira (Uttenreuth, DE)
Assignee: Siemens Healthineers AG
G06F16/383G06F11/0736G06F11/0793
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,737,409
App. No.
18/432,410
Granted
Sep 15, 2026
Kind
B2
Abstract

One or more example embodiments of the present invention describes a method for automated creation of search patterns on the basis of entries in a knowledge database. One or more example embodiments of the present invention further describes a search pattern dataset, a device for automated creation of search patterns, a problem solving method, a creation method, a training dataset and a device system.

Claims (31)

1 . A method for automated problem solving with a search pattern dataset based on entries in a knowledge database in which stored log passages of a protocol file of a device system are linked to at least one of problem analyses or problem solutions, the method comprising:

training a machine learning model using at least a training dataset including a search pattern dataset; and

creating the search pattern dataset, wherein the creating the search pattern dataset includes,

a) loading an entry from the knowledge database, the entry including at least one of explanatory text, images, and protocol files, and creating a search text from the entry, the entry corresponding to a problem in an imaging system;

b) identifying a log passage in the search text, the search text associated with text of the protocol file;

c) testing whether the identified log passage is relevant;

d) storing the log passage if the identified log passage is relevant or repeating steps b) and c) and identifying and checking a further log passage until no further log passage can be identified if the identified log passage is not relevant;

e) indicating a linkage of log passages in response to a text that adjoins the identified log passage in the search text;

f) storing the linkage together with a statement regarding to which log passages the linkage relates;

g) repeating the steps b) to f) with further log passages until a termination condition is reached;

h) generating a search pattern based on the stored log passages and stored linkages after reaching the termination condition wherein the search pattern corresponds to a ground truth based on a problem described in an entry, wherein the ground truth includes the problem in the imaging system and a solution to the problem in the imaging system; and

i) repeating the steps a) to h) for a plurality of entries in the knowledge database, wherein the search pattern dataset comprises the plurality of search patterns and the plurality of corresponding ground truths.

2 . The method of claim 1 , wherein the identifying the log passage includes searching the search text through for a number of known features of the log passages, the number of known features include one or more of the following elements:

a time stamp,

an identification,

a special character, or

a predetermined keyword.

3 . The method of claim 1 , wherein the testing includes:

investigating whether the identified log passage differs from normal text.

4 . The method of claim 1 , wherein the checking whether the linkage of the log passages is indicated includes,

searching a passage of the search text for predetermined key passages associated with corresponding linkages and, on recognition of at least one of the predetermined key passages, the corresponding linkage is associated with a preceding log passage.

5 . The method of claim 1 , wherein the search pattern is created based on text passages of the stored log passages.

6 . A computer program product comprising commands which, when executed by a processor, cause said processor to perform the method of claim 1 .

7 . A non-transitory computer-readable storage medium comprising commands which, when executed by a computer, cause said computer to perform the method of claim 1 .

8 . The method of claim 3 , wherein the investigating includes at least one of a colored emphasis of the log passage, an emphasis of a background of the log passage or a use of a textual emphasis.

9 . The method of claim 8 , further comprising:

adopting a particular emphasis in the testing in the entry in the knowledge database during a creation of the search text to check whether an identified log passage is relevant.

10 . The method of claim 5 , wherein two different log passages are linked to one another with a linkage based on a corresponding linkage between the two different log passages being indicated in the search text.

11 . The method of claim 10 , wherein the search pattern is additionally based upon time information items which are linked to the log passages.

12 . The method of claim 11 , wherein the search pattern is based upon a time difference between time stamps of two or more log passages.

13 . The method of claim 1 , wherein the imaging system is a medical imaging system.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 10, 2024
From: DE OLIVEIRA, ANDRE
To: SIEMENS HEALTHINEERS AG
Reel/Frame 067945/0439 →
Priority Claims (1)
DE 102023200969 · Feb 7, 2023 · national
Continuity (1)
Related Publication 20240265045A1 · Aug 8, 2024
References Cited (18)
US 20050193004A1 · Cafeo · 2005 [cited by examiner]
US 20150169288A1 · Yamamoto · 2015 [cited by examiner]
US 20150372855A1 · Kushmerick · 2015 [cited by examiner]
US 20160239487A1 · Potharaju · 2016 [cited by examiner]
US 20180333132A1 · Noda · 2018 [cited by examiner]
US 20190155709A1 · de Oliveira et al. · 2019 [cited by applicant]
US 20200005069A1 · Wang · 2020 [cited by examiner]
US 20200089983A1 · Manickam · 2020 [cited by examiner]
US 20220138777A1 · Ulla · 2022 [cited by examiner]
US 20240122555A1 · Pawlak · 2024 [cited by examiner]
US 20240126419A1 · Makhijani · 2024 [cited by examiner]
US 20240211330A1 · Branton · 2024 [cited by examiner]
US 20250013519A1 · Gao · 2025 [cited by examiner]
US 20250285749A1 · Naik · 2025 [cited by examiner]
EP 3486675A1 · 2019 [cited by applicant]
Zhukova, Nataly A. et al: “The Method for Searching Patterns In Log Files of Telecommunication Devices for Monitoring their State”; In: 2021 XXIV International Conference on Soft Computing and Measurements (SCM). IEEE, … [cited by applicant]
Lin Qingwei et al:“Log clustering based problem identification for online service systems.” Proceedings of the 38th International Conference on Software Engineering Companion. 2016. S.102-111. doi: 10.1145/2889160.28892… [cited by applicant]
“pytesseract 0.3.10”, in: https://pypi.org/project/pytesseract/#description [abgerufen am Jan. 24, 2024]. [cited by applicant]