Нові рекомендації щодо застосування байєсівського підходу за однією ознакою відносно досліджень показників онкохворих

• Дата: 2025
• Автори: Gupal, Anatolii, Tarasov , Andrii
• Формат: Стаття
• Мова: Ukrainian
• Опубліковано: V.M. Glushkov Institute of Cybernetics of NAS of Ukraine 2025
• Теми: рекомендації щодо вдосконалення розробки програмного забезпечення; байєсівська процедура розпізнавання; модифікована швидкість осідання еритроцитів; поверхневий плазмонний резонанс; білкові структури плазми крові; лазерний спектрограф; вторинний еритроцитоз; справжня поліцитемія
• Онлайн доступ: https://jais.net.ua/index.php/files/article/view/524
• Назва журналу: Problems of Control and Informatics

New approaches for software development using the Bayesian approach are described, which have not been used for one research indicator in the study of inflammatory processes in brain tumors using modified erythrocyte sedimentation rate, surface plasmon resonance in the study of blood plasma properties, and laser spectrograph indicators for indicators of plasma protein structures. New approaches are expedient to use for hematological diseases — secondary erythrocytosis and polycythemia vera. An increase in the speed of calculations is expected, which will make it possible to increase the total number of calculations performed. It will be necessary to evaluate what increases the speed of calculations, what optimizes the use of RAM. After receiving positive conclusions, it will become clear how best to use the Bayesian recognition approach based on several research indicators in recognizing diagnoses. The modified erythrocyte sedimentation rate method records the sedimentation of erythrocytes over time, the measurement of blood plasma protein structures using a laser spectrograph depends on the wave range of 0–1000 nm and has approximately 10 measurements per study. For both research methods, new approaches can be proposed to use the Bayesian approach when using Markov chains, Bayesian trees or other algorithms that consider the dynamics of measuring indicators of such studies. In the next work the relevant conclusions regarding the indicated new approaches for one research indicator and new advice on further improvements will be provided.