Olga Teplytska

ORCID: 0009-0007-1849-0115
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Statistical Methods in Clinical Trials
  • Neural Networks and Applications
  • Machine Learning in Healthcare
  • Computational Drug Discovery Methods
  • Gene Regulatory Network Analysis
  • HER2/EGFR in Cancer Research
  • Diabetes Treatment and Management
  • Chronic Myeloid Leukemia Treatments
  • Cancer Treatment and Pharmacology
  • Renal Transplantation Outcomes and Treatments
  • PARP inhibition in cancer therapy
  • Control Systems and Identification
  • Artificial Intelligence in Healthcare
  • Lung Cancer Treatments and Mutations
  • COVID-19 diagnosis using AI
  • Bipolar Disorder and Treatment

University of Bonn
2021-2025

Exposure-efficacy and/or exposure-toxicity relationships have been identified for up to 80% of oral anticancer drugs (OADs). Usually, OADs are administered at fixed doses despite their high interindividual pharmacokinetic variability resulting in large differences drug exposure. Consequently, a substantial proportion patients receive suboptimal dose. Therapeutic Drug Monitoring (TDM), i.e., dosing based on measured concentrations, may be used improve treatment outcomes. The prospective,...

10.3390/cancers13246281 article EN Cancers 2021-12-14

Abstract Non-linear mixed-effects models are a powerful tool for studying heterogeneous populations in various fields, including biology, medicine, economics, and engineering. Here, the aim is to find distribution over parameters that describe whole population using model can generate simulations an individual of population. However, fitting these distributions data computationally challenging if description individuals complex large. To address this issue, we propose novel machine...

10.1101/2023.08.22.554273 preprint EN cc-by bioRxiv (Cold Spring Harbor Laboratory) 2023-08-23

Abstract A variety of classical machine learning approaches have been developed over the past ten years with aim to individualize drug dosages based on measured plasma concentrations. However, interpretability these models is challenging as they do not incorporate information pharmacokinetic (PK) disposition. In this work we compare well-known population PK modelling and a newly proposed scientific (SciML) framework, which combines knowledge disposition data-driven modelling. Our approach...

10.1101/2024.05.06.24306555 preprint EN cc-by medRxiv (Cold Spring Harbor Laboratory) 2024-05-08
Coming Soon ...