I am a postdoctoral researcher at the Biomedical Data Science Center of Lausanne University Hospital (CHUV), where I work in Marianna Rapsomaniki’s AI/ML for Biomedicine group. I am an anatomical pathologist by training who builds computational systems to analyse pathology images at scale. My work sits at the intersection of morphological expertise and machine learning, combining the diagnostic eye of a pathologist with the quantitative rigour of computational image analysis. This dual skillset allows me to extract biological insight from histopathological data that extends well beyond manual review.

Before joining CHUV, I held an SNSF Postdoc.Mobility fellowship at MD Anderson Cancer Center in the Yuan Laboratory. I completed my pathology residency in Turkey and earned my PhD in translational and computational pathology at the University of Basel under the supervision of Prof. Luigi Terracciano.

Research Focus

Machine Learning for Digital Pathology

I develop machine learning and computer vision models that translate morphological knowledge into quantitative, automated pathology tools. Rather than treating algorithms as black boxes, I integrate domain expertise directly into model design so that outputs are interpretable and clinically meaningful. These systems enable single-cell-resolution mapping of cell types within heterogeneous tumour sections and support more accurate prediction of cancer survival and treatment response. The goal is not tool development for its own sake, but the principled integration of pathology knowledge with computational methods to produce biomarkers that matter for patient care.

Tumours as Evolving Ecosystems

I study tumours as complex, evolving ecosystems in which dynamic crosstalk among multiple cell types governs the evolutionary trajectory of cancer. Just as spatial organisation of species, predators, and habitats is central to understanding natural ecosystems, a spatially explicit approach is essential for understanding cancer progression and therapeutic response. My work examines how the spatial immune microenvironment shapes tumour behaviour across cancer types. Precancer pathology is of particular interest, as it offers a unique window into the earliest stages of malignant evolution.

Biology-Aware Image-Omics Integration

I combine pathological image analysis with spatial and bulk genomic data to decipher how genetically diverse cancer cells adapt to their microenvironment. By integrating image-based cell spatial mapping with bioinformatics and ecological statistics, we can study how cancers evolve and spread at high spatial resolution. This biology-aware approach aims to explain why cancer is so difficult to treat and to use ecological principles to guide more effective therapeutic strategies. The broader objective is to deliver scientific advances and clinical innovations by dissecting the spatial heterogeneity of the tumour microenvironment.

For a detailed overview of my projects and publications, see the Research & Publications page.

Collaborations

Throughout my career I have enthusiastically engaged in interdisciplinary collaborations spanning pathology, computational biology, genomics, and clinical oncology. I am always open for new partnerships that bring these fields together. If you are working on spatial biology, computational pathology, cancer evolution, or related areas, I would welcome the opportunity to connect.