HomeResearchScientific and scientific-technical projects on grant and program-targeted financing ProjectsAP26194209 «Mathematical and algorithmic support for an intelligent clinical decision support system in haematology and diabetology based on blood gasometry technology»

AP26194209 «Mathematical and algorithmic support for an intelligent clinical decision support system in haematology and diabetology based on blood gasometry technology»

Name of the project: AP26194209 «Mathematical and algorithmic support for an intelligent clinical decision support system in haematology and diabetology based on blood gasometry technology»

Competition name: Competition for grant financing of scientific and (or) scientific and technical projects for 2025-2027.

Project Manager: Rakhmetullina Zhenisgul Tuleukhanovna, candidate of physical and Mathematical Sciences, Associate Professor

Identifiers:

ResearcherID Web of Science: GYG-2630-2022
ORCID: https://orcid.org/0000-0002-0554-7684
Scopus Author ID: 57220809777

Project Research Team

№	Full name	Project position	Identifiers (Scopus Author ID, Researcher ID, ORCID, if any) and links to related profiles
1.	Rakhmetullina Zhenisgul Tuleukhanovna	Project Manager	ResearcherID Web of Science: GYG-2630-2022 ORCID: https://orcid.org/0000-0002-0554-7684 Scopus Author ID: 57220809777
2.	Uvalieva Indira Makhmutovna Bukunova	leading researcher	ResearcherID Web of Science: N-9474-2017, ORCID: https://orcid.org/0000-0002-2117-5390 Scopus Author ID: 56512350900
3.	Amenova Farida Seitkumarovna	leading researcher	ORCID: https://orcid.org/ 0000-0001-6656-5037 Scopus Author ID: 57220237643
4.	MukashevaRozaUrumkanovna	senior researcher	ResearcherID Web of Science: GYG-2630-2022 ORCID: https://orcid.org/0000-0002-4722-4576 Scopus Author ID: 55367791800.
5.	Almira Shoshanovna,	senior researcher	Scopus Author ID: 57224522852
6.	TezekpayevaShynarTolegenovna	Research Associate	Author ID в Scopus: 000512347 ResearcherID Web of Science: GSE-2587-2022 ORCID: ID 0009-0006-9552-0288.
7.	NurumovZharasZholymkhanuly	junior researcher
8.	Shakhmetov Madiyar Nurushevich,	consultant-expert
9.	BorozenetsDavidRafaelevich	engineer programmer	ResearcherID Web of Science: KNS-2113-2024, ORCID: https://orcid.org/0009-0008-2017-5840 Scopus Author ID: 58753693300).
10.	AbishevaGulzatZhomartovna	project manager

Project abstract:

The development of intelligent clinical decision support systems based on artificial intelligence, machine learning, and mathematical modeling technologies can significantly improve the quality of medical care. Creation of algorithmic and mathematical software capable of: - integrating gasometry data with other laboratory and clinical indicators, to identify hidden patterns and relationships between parameters, predict the development of pathological conditions. Offering optimal clinical solutions is an urgent scientific task in the field of medical informatics and biomedical engineering.

The aim of the projectis to developscientific and methodologicalfoundations for mathematicalandalgorithmicsupport of an intelligentclinicaldecisionsupportsystembased on bloodgasometrytechnology, which willimprove the effectiveness of medicalrecommendationsinhematologyanddiabetology.

To achieve the goal, the following logically interrelated tasks are planned:

research and analysis of theoretical and practical aspects of clinical decision support processes based on blood gasometry technology;
designing a conceptual model of a clinical decision support system in hematology and diabetology based on blood gasometry technology;
development of mathematical models and algorithms to support clinical decisions in hematology and diabetology based on blood gasometry technology and machine learning methods;
development of a hybrid model for analyzing blood gasometry indicators based on the technology of ensemble machine learning methods to improve the accuracy of diagnosis and the effectiveness of medical solutions in hematology and diabetology;
software implementation of modules of the intelligent clinical decision support system for the tasks of hematology and diabetology;
experimental study of an intelligent clinical decision support system for hematology and diabetology and development of necessary technical documentation.

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.
2025 year	The results of research and analysis of theoretical and practical aspects of clinical decision support processes based on blood gasometry technology have been obtained. A conceptual model of a clinical decision support system in hematology and diabetology based on blood gasometry technology has been designed. 1) Rakhmetullina Zh.T., Uvalieva I.M., Tezekpaeva Sh.T. Theoretical and methodological foundations of building a clinical decision support system based on blood gasometry // Science, innovation, education: current issues of the XXI century: collection of articles of the XV International Scientific and Practical Conference. Penza: ICNS "Science and Education". – 2025. – pp. 31-35 (ISBN 978-5-00268-152-5) URL: https://naukaip.ru/wp-content/uploads/2025/11/MK-2491.pdf 2) Rakhmetullina Zh.T., Uvalieva I.M., Tezekpaeva Sh.T. Development of a conceptual model of a clinical decision support system in hematology and diabetology // Youth and science of the XXI century: current theoretical research: collection of articles of the IV International Scientific and Practical Conference. Penza: ICNS "Science and Education". – 2025. – pp. 17-21 (ISBN 978-5-00268-153-2) URL: https://naukaip.ru/wp-content/uploads/2025/11/MK-2492.pdf
2026 year	Mathematical models and algorithms will be developed to support clinical decisions in hematology and diabetology based on blood gasometry technology and machine learning methods. Mathematical models and algorithms have been developed to support clinical decisions in hematology and diabetology based on blood gasometry technology and machine learning methods. A computational model has been developed that integrates the equations of gas exchange and acid-base equilibrium, including the Henderson–Hasselbach ratio, an estimate of the anion difference and compensation mechanisms. The proposed algorithm makes it possible to formalize the interpretation of pH, PaCO2, HCO3, and degree of oxygenation, providing differentiation of metabolic and respiratory disorders. The results demonstrate an increase in the objectivity and reproducibility of diagnostics, as well as the possibility of integrating the model into clinical information systems. The developed approach helps to reduce diagnostic errors and accelerate decision-making in intensive care. The analysis of blood gasometry parameters based on ensemble machine learning methods for the diagnosis of diabetic ketoacidosis has also been implemented. The study used clinical data from the MIMIC-IV database, which included 86,362 observations of patients in intensive care units. Algorithms Random Forest, Gradient Boosting, AdaBoost, XGBoost and AutoML Ensemble were used to build intelligent models. Pre-processing of the data included removing identifying features, scaling parameters, analyzing class imbalances, and dividing the sample into training and test parts. The quality assessment of the models was performed using the metrics Accuracy, Precision, Recall, F1-score, ROC-AUC and PR-AUC. The results of the study showed the high effectiveness of ensemble methods in the diagnosis of diabetic ketoacidosis. The highest classification values were demonstrated by the AutoML Ensemble model, which provided the maximum values of ROC-AUC and PR-AUC. Based on the results of the work, 2 (two) articles have been published in publications recommended by the COKNVO and 2 (two) articles in the materials of international scientific conferences.: 1. Uvalieva I., Rakhmetullina Zh., Tezekpaeva Sh., Mukhamedova R. Mathematical models and algorithms for supporting clinical decisions in hematology based on blood gasometry technology // EKTU Journal of Information and Communication Sciences. – 2026. – Vol. 1, No. 1. – pp. 141-156. DOI: https://doi.org/10.51885/3134-8025_JICS_2026_1_11. 2. Uvalieva I.M., Rakhmetullina Zh.T., Tezekpaeva Sh.T., Borozenets D.R., Nurumov Zh.Zh. A clinical decision support model for the diagnosis of acid-base balance disorders based on blood gasometry based on the Henderson-Hasselbach equation // Bulletin of Shakarym, 2026, No. 2, pp. 573-582. DOI: https://doi.org/10.53360/2788-7995-2026-2(22)-61. 3. Borozenets D.R., Uvalieva I.M. Development and integration of a prototype continuous glucose monitoring system with mobile and cloud technologies //Artificial intelligence and innovations: new challenges and engineering solutions of the future: proceedings of the XII International Scientific and Technical Conference of Students, Undergraduates, Doctoral Students and Young Scientists dedicated to the Year of Digitalization and Artificial Intelligence (Ust-Kamenogorsk, April 9-10, 2026). - Ust–Kamenogorsk: D. Serikbaev Moscow Technical Technical University, 2026. – Ch. IV. – pp. 72-78. URL: /files/research/AllConfs/XII_MNTK/%D0%A1%D0%B1%D0%BE%D1%80%D0%BD%D0%B8%D0%BA/4_%D1%87%D0%B0%D1%81%D1%82%D1%8C_.pdf. 4. Tezekpaeva Sh.T., Rakhmetullina Zh.T., Nurumov Zh.Zh. Algorithms and models for the analysis of blood gasometry parameters for the assessment of acid-base state //Artificial intelligence and innovations: new challenges and engineering solutions of the future: proceedings of the XII International Scientific and Technical Conference of Students, Undergraduates, Doctoral Students and Young Scientists dedicated to the Year of Digitalization and Artificial Intelligence (Ust-Kamenogorsk, April 9-10, 2026). - Ust–Kamenogorsk: D. Serikbaev Moscow Technical Technical University, 2026. – Ch. IV. – C.197-203. URL: /files/research/AllConfs/XII_MNTK/%D0%A1%D0%B1%D0%BE%D1%80%D0%BD%D0%B8%D0%BA/4_%D1%87%D0%B0%D1%81%D1%82%D1%8C_.pdf.
2027 year

Year

The results obtained from the research.

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

2025 year

The results of research and analysis of theoretical and practical aspects of clinical decision support processes based on blood gasometry technology have been obtained. A conceptual model of a clinical decision support system in hematology and diabetology based on blood gasometry technology has been designed.

1) Rakhmetullina Zh.T., Uvalieva I.M., Tezekpaeva Sh.T. Theoretical and methodological foundations of building a clinical decision support system based on blood gasometry // Science, innovation, education: current issues of the XXI century: collection of articles of the XV International Scientific and Practical Conference. Penza: ICNS "Science and Education". – 2025. – pp. 31-35 (ISBN 978-5-00268-152-5) URL: https://naukaip.ru/wp-content/uploads/2025/11/MK-2491.pdf
2) Rakhmetullina Zh.T., Uvalieva I.M., Tezekpaeva Sh.T. Development of a conceptual model of a clinical decision support system in hematology and diabetology // Youth and science of the XXI century: current theoretical research: collection of articles of the IV International Scientific and Practical Conference. Penza: ICNS "Science and Education". – 2025. – pp. 17-21 (ISBN 978-5-00268-153-2) URL: https://naukaip.ru/wp-content/uploads/2025/11/MK-2492.pdf

2026 year

Mathematical models and algorithms will be developed to support clinical decisions in hematology and diabetology based on blood gasometry technology and machine learning methods.
Mathematical models and algorithms have been developed to support clinical decisions in hematology and diabetology based on blood gasometry technology and machine learning methods. A computational model has been developed that integrates the equations of gas exchange and acid-base equilibrium, including the Henderson–Hasselbach ratio, an estimate of the anion difference and compensation mechanisms. The proposed algorithm makes it possible to formalize the interpretation of pH, PaCO2, HCO3, and degree of oxygenation, providing differentiation of metabolic and respiratory disorders. The results demonstrate an increase in the objectivity and reproducibility of diagnostics, as well as the possibility of integrating the model into clinical information systems. The developed approach helps to reduce diagnostic errors and accelerate decision-making in intensive care. The analysis of blood gasometry parameters based on ensemble machine learning methods for the diagnosis of diabetic ketoacidosis has also been implemented. The study used clinical data from the MIMIC-IV database, which included 86,362 observations of patients in intensive care units. Algorithms Random Forest, Gradient Boosting, AdaBoost, XGBoost and AutoML Ensemble were used to build intelligent models. Pre-processing of the data included removing identifying features, scaling parameters, analyzing class imbalances, and dividing the sample into training and test parts. The quality assessment of the models was performed using the metrics Accuracy, Precision, Recall, F1-score, ROC-AUC and PR-AUC. The results of the study showed the high effectiveness of ensemble methods in the diagnosis of diabetic ketoacidosis. The highest classification values were demonstrated by the AutoML Ensemble model, which provided the maximum values of ROC-AUC and PR-AUC.

Based on the results of the work, 2 (two) articles have been published in publications recommended by the COKNVO and 2 (two) articles in the materials of international scientific conferences.:

1. Uvalieva I., Rakhmetullina Zh., Tezekpaeva Sh., Mukhamedova R. Mathematical models and algorithms for supporting clinical decisions in hematology based on blood gasometry technology // EKTU Journal of Information and Communication Sciences. – 2026. – Vol. 1, No. 1. – pp. 141-156. DOI: https://doi.org/10.51885/3134-8025_JICS_2026_1_11.

2. Uvalieva I.M., Rakhmetullina Zh.T., Tezekpaeva Sh.T., Borozenets D.R., Nurumov Zh.Zh. A clinical decision support model for the diagnosis of acid-base balance disorders based on blood gasometry based on the Henderson-Hasselbach equation // Bulletin of Shakarym, 2026, No. 2, pp. 573-582. DOI: https://doi.org/10.53360/2788-7995-2026-2(22)-61.

3. Borozenets D.R., Uvalieva I.M. Development and integration of a prototype continuous glucose monitoring system with mobile and cloud technologies //Artificial intelligence and innovations: new challenges and engineering solutions of the future: proceedings of the XII International Scientific and Technical Conference of Students, Undergraduates, Doctoral Students and Young Scientists dedicated to the Year of Digitalization and Artificial Intelligence (Ust-Kamenogorsk, April 9-10, 2026). - Ust–Kamenogorsk: D. Serikbaev Moscow Technical Technical University, 2026. – Ch. IV. – pp. 72-78. URL: /files/research/AllConfs/XII_MNTK/%D0%A1%D0%B1%D0%BE%D1%80%D0%BD%D0%B8%D0%BA/4_%D1%87%D0%B0%D1%81%D1%82%D1%8C_.pdf.

4. Tezekpaeva Sh.T., Rakhmetullina Zh.T., Nurumov Zh.Zh. Algorithms and models for the analysis of blood gasometry parameters for the assessment of acid-base state //Artificial intelligence and innovations: new challenges and engineering solutions of the future: proceedings of the XII International Scientific and Technical Conference of Students, Undergraduates, Doctoral Students and Young Scientists dedicated to the Year of Digitalization and Artificial Intelligence (Ust-Kamenogorsk, April 9-10, 2026). - Ust–Kamenogorsk: D. Serikbaev Moscow Technical Technical University, 2026. – Ch. IV. – C.197-203. URL: /files/research/AllConfs/XII_MNTK/%D0%A1%D0%B1%D0%BE%D1%80%D0%BD%D0%B8%D0%BA/4_%D1%87%D0%B0%D1%81%D1%82%D1%8C_.pdf.

2027 year

Date of last changes:
17.09.2026