Department: Clinical Research Centre
Public PhD Defence date: 5 September 2022
Doctoral Thesis title: Predicting Type 2 Diabetes Remission After Sleeve Gastrectomy Using Clinical Data, Circulating microRNA, and Machine Learning Approach
Supervisor: Prof. dr hab. n. med. Adam Jacek Krętowski (Head of the Department of Endocrinology, Diabetology and Internal Medicine, MUB, Head of the Center for Clinical Research, MUB)
Reviewers:
Prof. dr hab. Dorota Zozulińska-Ziółkiewicz (Clinic of Internal Medicine and DiabetologyUniversity of Medical Science, Head of the Department)
Prof. dr hab. Tomasz Klupa Dr hab. Tomasz Klupa (Advanced Diabetological Technologies Laboratory at the Jagiellonian University)
Prof. dr hab. Elżbieta Bandurska-Stankiewicz (Head of the Clinical Department of Endocrinology, Diabetes and Internal Medicine at the Voivodeship Specialist Hospital in Olsztyn)
Current affiliation: assistant in Clninical Research centre Medical University of Białystok
Plans for the Future: Pursuing a research career in academia, starting with a Postdoctoral position at the Medical University of Białystok
Abstract of the Doctoral Thesis:
The aim of this doctoral dissertation is to profile pre-surgery serum miRNA from sleeve gastrectomy (SG) patients with T2D and develop prediction models using baseline clinical and miRNA data to predict T2D remission 12 months after surgery. Before SG, data and fasting serums were collected from 46 T2D patients. Two patients were excluded due to sample haemolysis. Six patients with unclear remission status were set aside for model evaluation. Model building was done with the remaining 38 patients. Variable selection was done using different approaches, including LASSO. LASSO was also used for model building. 68% of the patients achieved T2D remission. The best model (accuracy: 0.974) had four miRNAs (hsa-miR-32-5p, hsa-miR-382-5p, hsa-miR-1-3p, and hsa-miR-21-5p) and four clinical variables (T2D medication, sex, age, and fasting glucose). The results suggest that four serum miRNAs might be predictive biomarkers for T2D remission after SG, but validation studies are needed.