JBT Journal
Advanced

Raqamli Transformatsiya va Sun’iy Intellekt

SUN’IY INTELLEKT ASOSIDA BARIATRIK JARROHLIKDAN KEYINGI KLINIK HOLATLARNI BASHORATLASH: XGBOOST, LIGHTGBM VA SVM MODELLARINING QIYOSIY TAHLILI

Yuldoshev Yusuf Sheraliyevich1, Otaxonova Bahrixon Ibragimovna2, Olimjonov Orifjon Olimjon o‘g‘li2, Mirolimov Asilbek Alijon o‘g‘li2

1 O‘zbekiston Respublikasi Oliy ta’limfan va innovatsiyalar vazirligi huzuridagi Oliy attestatsiya komissiyasi

2 Muhammad al-Xorazmiy nomidagi Toshkent axborot texnologiyalari universiteti

Zenodo DOI: 10.5281/zenodo.20181349

Keywords

klinik risklarni bashoratlash Support Vector Machine Bariatrik jarrohlik klinik qaror qabul qilish tibbiy ma’lumotlar tahlili XAI SHAP ROC–AUC sun’iy intellekt XGBoost LightGBM postoperativ monitoring gradient boosting mashinaviy o‘rganish

Abstract

Mazkur tadqiqotda bariatrik jarrohlikdan keyingi klinik holatlarni bashoratlashda zamonaviy sun’iy intellekt algoritmlarining samaradorligi qiyosiy tahlil qilindi. Tadqiqotda 10000 ta bemorning anonimlashtirilgan longitudinal klinik ma’lumotlari asosida XGBoost, LightGBM va Support Vector Machine (SVM) modellarining bashoratlash imkoniyatlari baholandi. Ma’lumotlar bazasi demografik, antropometrik, laborator, operatsion va postoperativ monitoring ko‘rsatkichlarini o‘z ichiga oldi. Modellar bariatrik jarrohlikdan keyingi qayta gospitalizatsiya xavfi (readmission_30d), operatsiyadan keyingi asoratlar (complication_90d), 12 oylik metabolik yetishmovchilik holati (any_def_12m) hamda umumiy vazn yo‘qotish ko‘rsatkichlarini prognozlash uchun qo‘llanildi. Eksperimental natijalar XGBoost va LightGBM modellarining yuqori diskriminatsiya qobiliyatiga ega ekanligini ko‘rsatdi. Xususan, any_def_12m vazifasi bo‘yicha XGBoost modeli ROC–AUC = 0.916 va LightGBM modeli ROC–AUC = 0.914 natijalarni qayd etdi. SVM modeli esa nisbatan pastroq natija ko‘rsatgan bo‘lsa-da, alternativ yondashuv sifatida klinik tahlil uchun foydali ekanligi aniqlandi. Tadqiqot davomida ehtimolliklarni kalibratsiya qilish, sinflar balanssizligi muammosini hisobga olish hamda SHAP va LIME asosidagi tushuntiriladigan sun’iy intellekt (XAI) yondashuvlari qo‘llanildi. Shuningdek, tadqiqot natijalari asosida Tkinter platformasida ishlovchi klinik qarorlarni qo‘llab-quvvatlovchi desktop ilova ishlab chiqildi. Mazkur tizim bemor ko‘rsatkichlarini real vaqt rejimida tahlil qilish, individual klinik risklarni aniqlash hamda avtomatik tavsiyalar shakllantirish imkonini beradi. Tadqiqot natijalari bariatrik jarrohlikdan keyingi monitoring jarayonlarini optimallashtirish va klinik qaror qabul qilish samaradorligini oshirishda sun’iy intellekt modellarining yuqori amaliy ahamiyatga ega ekanligini ko‘rsatdi.

Views: 122 Published: 2026-04-28 Pages: 239-244 Volume: 4 Issue: 2

References

  1. V. Vapnik, Statistical Learning Theory, Wiley, 1998.
  2. C. Cortes and V. Vapnik, “Support-Vector Networks,” Machine Learning, vol. 20, no. 3, pp. 273–297, 1995.
  3. C. Cortes and V. Vapnik, “Support-Vector Networks,” Machine Learning, vol. 20, no. 3, pp. 273–297, 1995.
  4. World Health Organization, “Obesity and overweight,” WHO, 2024.
  5. A. Géron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, O’Reilly Media, 2022.
  6. T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016.
  7. American Society for Metabolic and Bariatric Surgery, “Bariatric Surgery Procedures,” ASMBS, 2024.
  8. H. Buchwald and D. M. Oien, “Metabolic/Bariatric Surgery Worldwide 2011,” Obesity Surgery, vol. 23, no. 4, pp. 427–436, 2013.
  9. L. Breiman, “Random Forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001.
  10. World Health Organization, “Obesity and overweight,” WHO, 2024.
  11. T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, Springer, 2009.
  12. S. Lundberg and S.-I. Lee, “A Unified Approach to Interpreting Model Predictions,” NeurIPS, 2017.
  13. T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” in
  14. E. Topol, “High-performance medicine: the convergence of human and artificial intelligence,” Nature Medicine, vol. 25, pp. 44–56, 2019.
  15. A. Esteva et al., “A guide to deep learning in healthcare,” Nature Medicine, vol. 25, pp. 24–29, 2019.
  16. G. Ke et al., “LightGBM: A Highly Efficient Gradient Boosting Decision Tree,” NeurIPS, 2017.
  17. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785–794.