АР23486396 «Models and methods for recognizing anatomical structures in MRI images in computer diagnostics tasks»

HomeResearchScientific and scientific-technical projects on grant and program-targeted financing ProjectsАР23486396 «Models and methods for recognizing anatomical structures in MRI images in computer diagnostics tasks»

АР23486396 «Models and methods for recognizing anatomical structures in MRI images in computer diagnostics tasks»

Competition name: Competition for program-targeted financing of scientific and scientific-technical programs for 2024-2026.

Project Manager: Tlebaldinova Aizhan Soltangalievna, PhD, associated professor.

Identifiers:

Scopus Author ID 56825933500 (https://www.scopus.com/authid/detail.uri?authorId=56825933500),
ORCID: 0000-0003-1271-0352 (https://orcid.org/0000-0003-1271-0352),
Mendeley.

Project Research Team

№	Full name	Project position	Identifiers (Scopus Author ID, Researcher ID, ORCID, if any) and links to related profiles
1.	Aizhan S. Tlebaldinova	Scientific director of the project (Chief Researcher)	Scopus Author ID: 56825933500 (https://www.scopus.com/authid/detail.uri?authorId=56825933500), ORCID: 0000-0003-1271-0352 (https://orcid.org/0000-0003-1271-0352), ResearcherID Web of Science: AAX-2461-2020 (https://www.webofscience.com/wos/author/record/2065080)
2.	Saule S. Smailova	Leading Researcher	Scopus Author ID: 55645191800 (https://www.scopus.com/authid/detail.uri?authorId=55645191800), ResearcherID Web of Science AAO-3775-2020 (https://www.webofscience.com/wos/author/record/2005603), ORCID: 0000-0002-8411-3584 (https://orcid.org/0000-0002-8411-3584).
3.	Saule K. Kumargazhanova	Leading Researcher	Scopus Author ID: 568262045000 (https://www.scopus.com/authid/detail.uri?authorId=56826204500), ResearcherID Web of Science: P-1875-2017 (https://www.webofscience.com/wos/author/record/1250411), ORCID: 0000-0002-6744-4023 (https://orcid.org/0000-0002-6744-4023).
4.	Ivan G. Kazancev	Leading Researcher	Scopus Author ID: 70043946600 (https://www.scopus.com/authid/detail.uri?authorId=7004394660), ResearcherID Web of Science: S-1049-2017 (https://www.webofscience.com/wos/author/record/1547717), ORCID: 0000-0002-8479-7349 (https://orcid.org/0000-0002-8479-7349).
5.	Zbigniew Omiotek	Leading Researcher	Scopus Author ID 55793789400 (https://www.scopus.com/authid/detail.uri?authorId=55793789400), ORCID 0000-0002-6614-7799 (https://orcid.org/0000-0002-6614-7799), ResearcherID Web of Science B-7028-2017 (https://www.webofscience.com/wos/author/record/1065785).
6.	Bagdat E. Balbosynov	Senior Researcher	Scopus Author ID 57995664100 (https://www.scopus.com/authid/detail.uri?authorId=57995664100), ORCID 0000-0003-0538-9564 (https://orcid.org/0000-0003-0538-9564).
7.	Markhaba A. Karmenova	Researcher	Scopus Author ID 57217096947 https://www.scopus.com/authid/detail.uri?authorId=57217096947), ORCID 0000-0002-3028-9461 (https://orcid.org/0000-0002-3028-9461), ResearcherID Web of Science ACV-9911-2022 (https://www.webofscience.com/wos/author/record/ACV-9911-2022).
8.	Aizhan M. Sydykova	Junior Researcher	-
Молодые ученые (до 40 лет + обучающие (Студенты, магистранты)) на анг
	Full name	Project position	Identifiers (Scopus Author ID, Researcher ID, ORCID, if any) and links to related profiles	Note (Teaching staff, student, Master's student, doctoral student)
1	Akerke K. Tankibayeva	Junior Researcher	ResearcherID Web of Science: JPX-1262-2023 (https://www.webofscience.com/wos/author/record/52025690), ORCID: 0009-0007-6053-4775 (https://orcid.org/0009-0007-6053-4775).	Doctoral student
2	Akbota S. Kumarkanova	Programmer	ORCID: 0000-0003-0517-2445 (https://orcid.org/0000-0003-0517-2445).	Teaching staff
3	Ivan N. Glinsky	Programmer	-	Master's student

Project abstract (relevance, purpose (no more than 250 words))….

The topic of the research is relevant because artificial intelligence technologies can assist orthopedic surgeons in making informed decisions about surgical intervention in MRI images by automatically highlighting anatomical structures. In addition, this reduces the likelihood of medical errors. Meniscal injuries can thus be diagnosed and assessed more accurately, resulting in more successful and high-quality treatment.

The purpose of this research is to develop improved models and methods for analyzing and recognizing anatomical structures in MRI images in order to enhance computer-aided diagnosis.

Expected results:

A set of annotated MRI images of the knee joint;
Preprocessing and improving the quality of MRI images using mathematical models;
Identifying damage to the meniscus of the knee joint using algorithms and models;
Software prototype for MRI diagnosis of knee meniscus injuries.

The end result of the project is a prototype software program assisting less experienced specialists in making medical decisions about meniscus injuries of the knee joint based on MRI images, which provided fast and accurate analysis and processing.

The practical significance of the research results lies in improving diagnostic accuracy, reducing errors and false-positive results, early detection of diseases, optimizing the examination process and increasing the efficiency of medical institutions.

The main expected socio-economic effect. This study will collect MRI and arthroscopy data of patients with knee meniscal injuries over the last 10 years in Kazakhstan. The developed models and methods will make it possible to recognize various anatomical structures, which will ensure their wide applicability in diagnostic tasks.

Expected and achieved results of the project:

Year	The results obtained from the research. Publications (with links to them) and patents; information for potential users.
2024 year	1) MRI and arthroscopy data on patients with meniscus damage of the knee joint for the last 10 years were collected in the regions of Kazakhstan (RGP on PCV ‘National Scientific Centre of Traumatology and Orthopaedics named after Academician N.D. Batpenov’). 2) An algorithm was developed that automates the process of determining the number of images, modes of their visualisation, depth of slices and other parameters, as well as exporting data in 2D-format and sorting them into folders according to the modes of visualisation. As a result, all these data are automatically entered into an Excel file 3) The technological chain of MRI images processing was developed, which includes the following stages: study and analysis, preprocessing, registration, framing of the area of interest, analysis and extraction of quantitative characteristics of images. 4) Mathematical models of image processing are developed. 5) Published article in the journal recommended by the SHEQAC RK: Tankibayeva A.K., Kumargazhanova S.K., Tlebaldinova A.S. ‘Review of methods of correction of intensity inhomogeneity of magnetic resonance images’ / Bulletin of Toraigyrov University. № 4(2024), 2024. С.332-344. 6) The article is prepared and submitted to the journal recommended by SHEQAC RK: Tlebaldinova A.S., Karmenova M.A., Kumargazhanova S.K., Smailova S.S. ‘Exploratory analysis of MRI images to improve the quality of medical imaging and diagnostics’/Proceedings of Kartu University. The article has been accepted for publication (expected date of publication 1 issue, March, 2025)
2025 year	An annotated database of images with markings of damaged menisci of the knee joint will be prepared. The results of the conducted experimental studies will be reported at the international conference, on the materials of which 1 article will be submitted. Improved models and methods for recognising meniscus injuries of the knee joint will be developed. One article will be submitted to a peer-reviewed scientific publication indexed and included in the 1st (first) and (or) 2nd (second) quartile of the impact factor in the Web of Science database and (or) having a CiteScore percentile in the Scopus database of at least 65 (sixty-five).
2026 year	A prototype of MRI software for the diagnosis of meniscus injuries of the knee joint will be developed. The results of the conducted experimental studies will be reported in the international conference, on the materials of which 1 article will be submitted. All stages of work on testing the software prototype will be carried out. One article will be submitted to a peer-reviewed scientific publication indexed and included in the 1st (first) and (or) 2nd (second) quartile of the impact factor in the Web of Science database and (or) having a CiteScore percentile of at least 65 (sixty-five) in the Scopus database.

Year

The results obtained from the research.

Publications (with links to them) and patents; information for potential users.

2024 year

1) MRI and arthroscopy data on patients with meniscus damage of the knee joint for the last 10 years were collected in the regions of Kazakhstan (RGP on PCV ‘National Scientific Centre of Traumatology and Orthopaedics named after Academician N.D. Batpenov’).

2) An algorithm was developed that automates the process of determining the number of images, modes of their visualisation, depth of slices and other parameters, as well as exporting data in 2D-format and sorting them into folders according to the modes of visualisation. As a result, all these data are automatically entered into an Excel file

3) The technological chain of MRI images processing was developed, which includes the following stages: study and analysis, preprocessing, registration, framing of the area of interest, analysis and extraction of quantitative characteristics of images.

4) Mathematical models of image processing are developed.

5) Published article in the journal recommended by the SHEQAC RK: Tankibayeva A.K., Kumargazhanova S.K., Tlebaldinova A.S. ‘Review of methods of correction of intensity inhomogeneity of magnetic resonance images’ / Bulletin of Toraigyrov University. № 4(2024), 2024. С.332-344.

6) The article is prepared and submitted to the journal recommended by SHEQAC RK: Tlebaldinova A.S., Karmenova M.A., Kumargazhanova S.K., Smailova S.S. ‘Exploratory analysis of MRI images to improve the quality of medical imaging and diagnostics’/Proceedings of Kartu University. The article has been accepted for publication (expected date of publication 1 issue, March, 2025)

2025 year

An annotated database of images with markings of damaged menisci of the knee joint will be prepared. The results of the conducted experimental studies will be reported at the international conference, on the materials of which 1 article will be submitted.

Improved models and methods for recognising meniscus injuries of the knee joint will be developed.

One article will be submitted to a peer-reviewed scientific publication indexed and included in the 1st (first) and (or) 2nd (second) quartile of the impact factor in the Web of Science database and (or) having a CiteScore percentile in the Scopus database of at least 65 (sixty-five).

2026 year

A prototype of MRI software for the diagnosis of meniscus injuries of the knee joint will be developed. The results of the conducted experimental studies will be reported in the international conference, on the materials of which 1 article will be submitted.

All stages of work on testing the software prototype will be carried out.

One article will be submitted to a peer-reviewed scientific publication indexed and included in the 1st (first) and (or) 2nd (second) quartile of the impact factor in the Web of Science database and (or) having a CiteScore percentile of at least 65 (sixty-five) in the Scopus database.

Infographics (photos, drawings, etc.)