Systems Medicine, Digital Twins and AI
Metabolic Inflammation and Carcinogenesis of the Liver
We build digital twins of human physiology — mechanistic, AI-powered models that predict how disease and therapy play out in every individual patient.
Data scientist, data analyst, computational modeler, bioinformatician, group leader. Application of computational modeling and machine learning to biological, medical and clinical questions and data...
Mariia works on ATLAS, a decision support tool for hepatocellular carcinoma (HCC). Based on AI methods, ATLAS processes all relevant patient data from databases,...
Shubhankar worked on enhancing our understanding of enalapril's pharmacokinetics and dynamics based on a physiologically based modeling approach. The work will build on top...
Michelle develops a physiologically based pharmacokinetic/pharmacodynamic (PBPK) model of the sulfonylurea glimepiride. The objective is to enhance our understanding of the underlying causes of...
Mariia workes on a physiologically based model of lisinopril. Key questions are the effect of hepatic and renal disease on lisinopril pharmacokinetics and pharmacodynamics....
Amanda develops a physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) model of the diuretic hydrochlorothiazide. The objective is to enhance our understanding of how hydrochlorothiazide changes urinary...
Kim is developing a physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) model of the GLP-1 receptor agonist dulaglutide to better understand its absorption, distribution, metabolism, and effects...
Jessica Cruz is developing a physiologically based pharmacokinetic (PBPK) model of the leukotriene receptor antagonist montelukast to better understand its absorption, distribution, metabolism, and...
Shelee is developing a physiologically based pharmacokinetic (PBPK) model of the multikinase inhibitor lenvatinib to better understand its absorption, distribution, metabolism, and excretion. The...
Chang Linh is developing a physiologically based pharmacokinetic (PBPK) model of the β2-adrenergic receptor agonist albuterol (salbutamol) to better understand its absorption, distribution, metabolism,...
Sandeep is developing a physiologically based pharmacokinetic (PBPK) model of the multi-kinase inhibitor regorafenib to better understand its absorption, distribution, metabolism, and excretion. The...
Antonio will delve into the influence of parameter variability in PBPK models, focusing on the consequential effects for dynamic liver function assessments. The overarching...
Latest News
2026-08-05
Publication: EFECT Enables Reliable and Reproducible Stochastic Simulations Across Disciplines
We introduce EFECT, the Empirical Characteristic Function Equality Convergence Test, a universal statistical framework for quantifying the reproducibility of stochastic simulations across systems biology, epidemiology, finance, and machine learning.
2026-07-07
Poster at HIP 2026 day: A PBPK Model of the Leukotriene Receptor Antagonist Montelukast
Jessica Cruz Pérez presents her internship project developing a physiologically based pharmacokinetic (PBPK) model of montelukast to investigate interindividual variability in pharmacokinetics, including effects of hepatic metabolism, CYP2C9/CYP3A4 variability, drug-drug interactions, food...
2026-07-06
Preprint: A Physiologically Based Pharmacokinetic/Pharmacodynamic Model of the Diuretic Hydrochlorothiazide
We present a whole-body PBPK/PD model of the diuretic hydrochlorothiazide to investigate pharmacokinetic and pharmacodynamic variability in hypertensive patients, with a focus on renal, hepatic, and cardiac impairment.
2026-06-04
Poster at PAGE 2026: PK-DB 2.0 for Pharmacometric Modeling, AI, and Digital Twins
We present PK-DB 2.0, an open, FAIR, and AI-ready pharmacokinetics and pharmacodynamics database providing curated PK/PD data from more than 700 clinical studies for pharmacometric modeling, machine learning, and digital twin development....
2026-05-06
Preprint: An Open and Reproducible Digital Twin for Personalised Anticoagulant Therapy
We present an open whole-body PBPK/PD digital twin of apixaban to investigate pharmacokinetic and pharmacodynamic variability in personalised anticoagulant therapy, with a focus on renal and hepatic impairment, food intake, body weight,...
2026-04-30
Automated Segmentation of Hepatic Vessels and Lobules in Whole-Slide Images Using U-Net Models
Our latest publication introduces a deep-learning pipeline for automated segmentation of hepatic vessels, bile ducts, and lobules in whole-slide liver histology images.
2026-04-23
Master Thesis Antonio Alvarez
Antonio Alvarez submitted his Master thesis on reproducible Bayesian optimization and uncertainty quantification in pharmacokinetic models
2026-03-20
Preprint: A Digital Twin of Empagliflozin Pharmacokinetics and Pharmacodynamics
We present a whole-body PBPK/PD digital twin of empagliflozin to investigate pharmacokinetic and pharmacodynamic variability in type 2 diabetes mellitus, with a focus on renal and hepatic impairment, metabolite exposure, and dose-dependent...
2026-03-12
Keynote: Digital Twins and Dashboards for Personalized Drug Treatment
At the Patient Monitoring Roundtable (PMRT) at Charité – Universitätsmedizin Berlin, I presented how digital twins and interactive dashboards can support personalized drug treatment and clinical decision-making.
2026-01-27
Preprint: A Digital Twin of Canagliflozin Pharmacokinetics and Pharmacodynamics in Type 2 Diabetes Mellitus
We present a whole-body PBPK/PD digital twin of canagliflozin to investigate pharmacokinetic and pharmacodynamic variability in type 2 diabetes mellitus, with a focus on renal and hepatic impairment, metabolite exposure, and dose-dependent...
2026-01-30
Interview: Modelle so komplex wie der Körper selbst – Digitale Zwillinge in der Medizin
In an interview with Matthias König, the Gesellschaft für Informatik explores how digital twins are becoming as complex as the human body itself and how they enable personalized prediction, simulation, and decision...
2025-11-28
Successful Open Science Workshop 2025
The Open Science Workshop 2025 brought together participants from across the Berlin University Alliance for a full-day, hands-on introduction to reproducible research, FAIR data, and open-source workflows.
2025-11-20
Completion of the AI Teaching Certificate (KI-Lehrzertifikat)
Successful completion of the AI Teaching Certificate (200 AE) at Humboldt-Universität zu Berlin, covering AI foundations, ethics, and higher education didactics within the KI-Campus 2.0 and AI Skills Initiative.
2025-11-13
Humboldt Internship Program 2026 - Computational Modeling of Drug Detoxification Using Digital Twins – A Systems Medicine Approach
We offer Internships in 2026, academic three-month research stays at Humboldt-Universität zu Berlin. The application for the Summer Term Internships 2026 is open from 05 November to 18 December 2025.
2025-11-13
A Digital Twin of Dapagliflozin Pharmacokinetics and Pharmacodynamics in Type 2 Diabetes Mellitus
We present a physiologically based PBPK/PD digital twin of dapagliflozin to analyze pharmacokinetic and pharmacodynamic variability in type 2 diabetes mellitus, with a focus on dosing, hepatorenal impairment, and food effects.
2025-11-01
Faces of the Berlin University Alliance
I am very honored to be featured in the new film series launched for the 6th anniversary of the Berlin University Alliance (BUA). Watch my short portrait: 🎥. Learn more about...
2025-10-14
Digital Twin of Losartan for Precision Blood Pressure Therapy
We introduce a physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) model of losartan and its active metabolite E3174 as a digital twin to study variability in blood pressure regulation and support individualized antihypertensive therapy.
2025-10-20
COMBINE & RDA: Reproducibility by Design: Data and Workflows for Pharmacokinetic Digital Twins
Presenting at COMBINE2025 and the RDA assembly our approach to reproducible digitial twins.
2025-10-08
Digital Twin of Glimepiride Enables Personalized Diabetes Therapy
We present a digital pharmacokinetic twin of glimepiride that simulates how patient-specific factors affect drug behavior, supporting individualized and stratified treatment of type 2 diabetes.
2025-09-15
Assessing the Impact of AI and Digital Twins on Clinical Decision-Making in Hepatology and Hepatobiliary Surgery
Our latest preprint presents results from a survey study exploring how clinicians in hepatology and hepatobiliary surgery perceive artificial intelligence (AI) and digital twins in clinical decision-making, highlighting both enthusiasm for innovation...