Anouar Nouri
AI Researcher | Medical AI Specialist | Berlin
Hell-bent on developing transparent, trustworthy AI systems for healthcare. Specializing in causal inference, digital twin modeling, and medical imaging.
About Me
I'm Anouar Nouri, a researcher, builder, and runner living in Berlin. I am full of energy, full of ideas, and deeply curious about the world. I live my life to the fullest, constantly pushing my limits — intellectually, professionally, and personally.
My goal is simple: to learn as much as I can, build as much as I can, and help as much as I can, while caring for the people closest to my heart.
My work is centered on AI in medicine, with a focus on developing methods that make clinical decisions more precise, transparent, and trustworthy.
I believe technology should amplify the human mind, not replace it.
Research Focus
I aim to bridge the gap between cutting-edge AI methods and their safe, meaningful use in clinical practice.
Current Research Interests
Causal Inference in Healthcare
Individualized treatment effect estimation for stroke and oncology patients, including digital twin modeling.
Medical Imaging & Segmentation
Automated analysis of CT, MRI, and ultrasound, with Bayesian uncertainty estimation and marker-removal pipelines.
Multimodal Learning
Integrating imaging and clinical data for robust decision support.
Transparency & Responsibility in AI
Developing explainable models clinicians can trust.
Multi-Agent Pipelines
Designing modular orchestration systems for data retrieval, enrichment, and reflective processing to achieve complex tasks.
My long-term vision is to build clinically useful, interpretable, and robust AI systems that act as trusted allies to clinicians — especially in regions facing healthcare workforce shortages, such as my home country. By reducing waiting times, improving diagnostic accuracy, and supporting early interventions, I hope to contribute to higher survival rates and a better quality of life for patients who need it most.
Publications
Causal Inference for Ischemic Stroke Treatment: Individualized Effect Estimation
In preparation
Predicting Malignancy of Pediatric Head and Neck Lymphadenopathy by Applying Multi-modal Deep Learning Approaches
RSNA 2025
Deep Learning-based Multiclass Segmentation in Aneurysmal Subarachnoid Hemorrhage
PLOS ONE, June 24, 2024
A Comparative Study of Claim Extraction Techniques Leveraging Transformer-based Pre-trained Models
LTC'23, April 23, 2023
Professional Experience
Master's Thesis Researcher
Fraunhofer Institute & Charité AI Lab — 2024–Present
Project: "Causal Inference and Digital Twin Architectures for Individualized Treatment Effect Estimation in Ischemic Stroke"
- Developed a causal inference framework using multicenter stroke registries (German Stroke Registry, MR CLEAN)
- Built and evaluated DragonNet and TransTEE architectures with conformal calibration
- Explored digital twin modeling to simulate alternative treatment scenarios
- Built a novel evaluation framework for causal inference medical models
- Worked closely with clinicians and researchers to align modeling with real-world medical workflows
AI Researcher
Charité — Universitätsmedizin Berlin — May 2023–Present
- Collaborated with multidisciplinary teams of clinicians and AI researchers
- Developed brain CT segmentation pipelines and diffusion-based image generation methods
- Designed multimodal AI frameworks combining imaging and clinical data for cancer diagnosis
- Conducted explainable AI analyses for better interpretability in high-stakes settings
- Authored and co-authored scientific papers in medical AI
Software Developer
Complexium GmbH — September 2021–Present
- Developed internal retrieval-augmented generation pipelines for domain-specific applications
- Built NLP tools and visual analytics dashboards for structured and unstructured data
- Developed scalable web scraping and orchestration components for data collection
- Contributed to multi-agent system design for complex data processing workflows
Bachelor Thesis
Deloitte Deutschland — January 2022–June 2022
- Researched and fine-tuned transformer language models for claim extraction
- Developed novel hybrid ML techniques
- Conducted rigorous experimentation and evaluation
Education
M.Sc. Computer Science (Cognitive Systems)
Technische Universität Berlin — 2023–Present
Focus: Medical AI, Causal Inference, Digital Twins, Deep Learning
Exchange Semester
KU Leuven, Belgium — September 2024–February 2025
Neural Computing / Medical AI
B.Sc. Computational Engineering Science
Technische Universität Berlin — 2018–2023
Technical Skills
Machine Learning & AI
Medical Imaging, Segmentation, Causal Inference, Multimodal Learning, XAI, Responsible AI, Computer Vision, NLP, Data Engineering and Analysis, AI Ethics, Prompt Engineering
Programming & Tools
PyTorch, TensorFlow, FastAPI, Python, Docker, Nginx, Git, Linux
Engineering & DevOps
Web scraping, Data orchestration, Docker, Cloud Computing, DevOps
Languages
Native
Fluent
Fluent
Proficient
Beginner
Running & Personal Life
Running is my way of gaining as well as releasing my excess energy, dealing with stressful situations, and just pushing my limits and having fun.
Personal Bests
Berlin Half 2025
Berlin Marathon 2025
Marathon
Other Interests
Tennis, bouldering, VR gaming, sketching, and improvisational theater
Side Projects
FindMyRace.de
Berlin race results search engine for runners and endurance athletes.
RAG Pipeline Data Engineering UI DesignMedScan Filter Assistant
In progress — Deep learning system for marker removal in ultrasound scans.
Medical AI Vision EncoderThe Resonance
Creative sci-fi project — A narrative exploring morality, resurrection, and consciousness.
Creative WritingPhilosophy
"Just live."
Get In Touch
Let's connect and create something impactful together.