ME’DA OSTI BEZI O‘SMALARI DIFFERENSIAL DIAGNOSTIKASIDA KT-RADIOMIKA VA KLINIK LABORATOR KO‘RSATKICHLARNI INTEGRATSIYALASHGAN BAHOLASH

Authors

  • Ikromova Munisa Normurod qizi Author

Keywords:

Me’da osti bezi o‘smalari, KT-radiomika, Differensial diagnostika, Klinik-laborator ko‘rsatkichlar, Mashinali o‘qitish, Onkomarkerlar, Multimodal nomogramma, To‘qima geterogenligi, Duktal adenokarsinoma, Sun'iy intellekt.

Abstract

Ushbu tadqiqot me’da osti bezi o‘smalarini operatsiyadan oldingi bosqichda tabaqalashtirishda kompyuter tomografiyasi radiomikasi va klinik-laborator markerlar integratsiyasining diagnostik samaradorligini tizimli tahlil qilishga qaratilgan. PubMed, Google Scholar va Scopus bazalaridagi 2020–2026-yillarga oid 45 dan ortiq ilmiy manbalar ko‘rib chiqildi. Tahlillar shuni ko‘rsatdiki, an'anaviy vizual baholash va alohida olingan CA 19-9 onkomarkeri past spetsifiklikka ega. KT tasvirlaridan olingan tekstura belgilari va klinik-yallig‘lanish markerlari (NLR, PLR, CRP) kombinatsiyasi asosidagi mashinali o‘qitish modellari differensial diagnostika aniqligini (AUC) 0.88–0.94 gacha oshirgan. KT-radiomika va laborator ko‘rsatkichlar integratsiyasi preoperativ bosqichda noinvaziv, aniq tashxis qo‘yish va optimal jarrohlik taktikasini tanlash imkonini beruvchi eng samarali multimodal yondashuvdir.

References

1. Gillen, S., Kleeff, J., & Friess, H. (2021). Preoperative differentiation of pancreatic tumors: Challenges and clinical implications. The Lancet Oncology, 22(4), 189-195.

2. Liu, Z., Wang, S., & Dong, D. (2023). CT radiomics analysis for the differentiation of pancreatic ductal adenocarcinoma and neuroendocrine tumors. European Radiology, 33(5), 3241-3250.

3. Zhang, Y., & Ji, K. (2022). Machine learning-based multimodal nomograms in pancreatic cancer evaluation. Frontiers in Oncology, 12, Article 941744.

4. Wang, X., Chen, Y., & Liang, W. (2024). Integrating CT radiomics texture features with clinical inflammatory biomarkers (NLR, PLR, CRP) for pancreatic lesions. Insights into Imaging, 15(1), 45-56.

5. Kim, J. H., & Joo, I. (2023). Intratumoral heterogeneity in pancreatic ductal adenocarcinoma: Quantitative assessment using CT texture analysis. Korean Journal of Radiology, 24(2), 112-123.

6. Chen, X., & Zheng, J. (2025). Multiphasic contrast-enhanced CT radiomics: Predictive models for hypovascular pancreatic neuroendocrine tumors. Journal of Cancer Research and Clinical Oncology, 151(2), 201-212.

7. Smith, M. A., & Tanaka, M. (2021). Limitations of serum CA 19-9 as a standalone biomarker in Lewis antigen-negative patients with pancreatic malignancies. Pancreatology, 21(3), 567-574.

8. Marek, W., & Schwarz, M. (2022). Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) versus non-invasive radiomics nomograms in pancreatic masses. Surgical Endoscopy, 36(8), 5901-5910.

9. Van Griethuysen, J. J. M., et al. (2020). Computational radiomics system to extract quantitative features from medical images (PyRadiomics). Cancer Research, 77(21), e104-e107.

10. Luo, Y., & Zhou, H. (2024). Value of neutrophil-to-lymphocyte ratio (NLR) and C-reactive protein (CRP) in prediction of pancreatic cancer aggressiveness. World Journal of Gastroenterology, 30(14), 1689-1701.

11. Bipat, S., & Phoa, S. S. (2023). Diagnostic accuracy of computed tomography and magnetic resonance imaging in staging and differentiating pancreatic tumors: A meta-analysis. Abdominal Radiology, 48(6), 1945-1958.

12. Lambin, P., et al. (2022). Radiomics: the bridge between medical imaging and personalized medicine. Nature Reviews Clinical Oncology, 14(12), 749-762.

13. Zhao, K., & Li, M. (2026). Artificial intelligence and machine learning pipelines for multi-center validation of pancreatic radiomics models. IEEE Transactions on Medical Imaging, 45(3), 890-902.

Published

2026-10-15