The Diagnostic Eye: Żaneta Świderska-Chadaj and the Art of Teaching Machines to See

The Pathologist's Gaze, Augmented

Stand in a pathology lab and watch a specialist examine a tissue slide. The process looks almost meditative: the slow adjustment of the microscope, the careful scanning, the moment of recognition when abnormal cells reveal themselves. This is diagnostic work at its most intimate, a human eye trained over years to detect patterns invisible to the untrained observer.

Now consider what happens when that gaze is augmented, extended, made computational. Żaneta Świderska-Chadaj, Associate Professor at Warsaw University of Technology and ELLIS Member at the ELLIS Unit Warsaw, has spent her career at precisely this intersection. Her research focus, as described in her recent ELLIS spotlight, centers on applying AI to computational pathology to enhance medical diagnosis through advanced data analysis.

The phrase sounds technical. The implications are not.

A Career Built at Intersections

Świderska-Chadaj's trajectory reveals something about how serious AI research actually develops. According to her profile at the Academy of Young Scientists, she obtained her habilitated doctorate in Biomedical Engineering from Warsaw University of Technology in 2022, following a PhD in Computer Science from the same institution in 2017. Her engineering degree came from Silesian University of Technology in 2012.

Between these credentials lies the more interesting story: a 2.5-year postdoctoral fellowship at the Computational Pathology research group at Radboudumc in Nijmegen, Netherlands (2017-2020), a study visit to Cedars-Sinai in the USA (2019), and a three-month internship at the University of Castilla La Mancha in Spain (2017). This is not a career built in isolation. It is a career built through movement, through exposure to different research cultures and clinical contexts.

From 2021 to 2025, she worked at Bayer, bringing her academic research into contact with pharmaceutical industry applications. Since 2025, she has led the Medical Image Data Analysis team at IDEAS Research Institute, where the work focuses on applying artificial intelligence methods to develop tools that support medical diagnostics based on image data.

What the Publications Reveal

Her Google Scholar profile tells a story through titles. Convolutional neural networks can accurately distinguish four histologic growth patterns of lung adenocarcinoma in digital slides (Scientific Reports, 2019). Learning to detect lymphocytes in immunohistochemistry with deep learning (Medical Image Analysis, 2019). Artificial intelligence to detect MYC translocation in slides of diffuse large B-cell lymphoma (Virchows Archiv, 2021).

These are not abstract exercises. Each paper represents a specific diagnostic challenge: distinguishing cancer growth patterns, detecting immune cells, identifying genetic translocations that affect treatment decisions. The work sits at the boundary between what a pathologist can see and what a computational system can learn to recognize.

Her research on kidney transplant biopsies, conducted with collaborators at Radboudumc, addresses the quantitative assessment of inflammatory infiltrates, a task where consistency between human observers has historically been difficult to achieve. The work on prostate cancer classification examines how rescanning and normalization affect neural network performance across multiple centers, a practical concern for any system intended for real-world deployment.

The Warsaw Node

The ELLIS Unit Warsaw, where Świderska-Chadaj serves as a member, represents something specific in the European AI landscape. Established in November 2023, it was the first ELLIS Unit in Poland and one of the first in Central-Eastern Europe. The main hosting institution is IDEAS Research Institute, with co-hosting from the University of Warsaw and Warsaw University of Technology since December 2025.

The unit's research agenda spans six areas: intelligent algorithms and learned data structures, efficient and sustainable machine learning for computer vision, machine learning and sequential decision-making, algorithmic game theory in security, interpretable artificial intelligence, and autonomous agents and alignment of language models. Healthcare AI, including medical image analysis, sits within this broader ecosystem.

What makes this configuration interesting is its explicit mission to bridge academia and industry, to translate research into practical applications. The unit participates in European projects including ELSA (European Lighthouse on Secure and Safe AI) and ELIAS (European Lighthouse of AI for Sustainability), connecting Polish researchers to continent-wide initiatives.

Recognition and Its Meaning

The awards accumulated over Świderska-Chadaj's career offer a map of institutional recognition: the Outstanding Young Scientist Award from Start FNP (2020), the Minister's Scholarship (2021), the POLITYKA Scientific Award (2022), and inclusion among the TOP 100 Women in AI by the Perspectives Foundation. She is a member of IEEE and actively participates in the Women in Tech monitoring program.

Her stated goal in participating in the #WomenInELLIS campaign is direct: to show that science has immense, vital space for women. The statement is simple. The context is not. Computational pathology, like much of AI research, has historically been male-dominated. The visibility of researchers like Świderska-Chadaj matters not as decoration but as evidence of what is possible.

The Diagnostic Future

Her vision for her career, as stated in the ELLIS spotlight, is leading an expert AI/ML team dedicated to medical image analysis. The purpose: To unlock new insights, accelerate precision diagnosis, and collaboratively tackle healthcare's toughest challenges.

The phrase precision diagnosis deserves attention. In pathology, precision means the difference between catching a cancer early and missing it, between identifying the right treatment target and prescribing something ineffective. The stakes are not abstract.

The Medical Image Data Analysis team she leads at IDEAS Research Institute works on specific technical challenges: classification (cancer cell vs. healthy cell), segmentation (tumor area delineation), and modeling of specific processes (cell differentiation). One main research area is developing algorithms to support histopathological diagnosis of lymphoma, a cancer of the lymphatic system where accurate classification directly affects treatment decisions.

What This Means for Europe

The European AI ecosystem is often discussed in terms of regulation, investment, and competition with American and Chinese systems. These discussions matter. But they can obscure the more fundamental question: what kind of AI is Europe actually building, and who is building it?

Świderska-Chadaj's work offers one answer. It is AI developed in close collaboration with clinical practice, tested across multiple centers and countries, designed to augment rather than replace human expertise. It is research that moves between academic institutions and industry, between Poland and the Netherlands and Spain and the United States.

The ELLIS network, with its emphasis on connecting top researchers across European borders, provides infrastructure for this kind of work. The IDEAS Research Institute, operating under the supervision of Poland's Ministry of Digital Affairs and Ministry of National Defence, represents a different model: state investment in AI research with explicit attention to technological sovereignty.

Whether these models can scale, whether they can produce systems that actually improve patient outcomes at population level, remains to be seen. The technical challenges are substantial. The regulatory environment is complex. The integration of AI tools into clinical workflows requires more than algorithmic accuracy; it requires trust, training, and institutional change.

The Artifact and the Question

A tissue slide under a microscope. A digital image on a screen. A neural network's probability distribution over diagnostic categories. These are artifacts, each encoding assumptions about what matters and how to see it.

Świderska-Chadaj's work sits at the intersection of these artifacts, translating between the pathologist's trained eye and the computational system's learned representations. The question her research implicitly asks is not whether machines can see, but what it would mean for them to see well, to see in ways that serve patients and clinicians rather than replacing them.

Żaneta Świderska-Chadaj

I believe precise AI in diagnostics can redefine patient care.

The belief is not naive. It is grounded in years of research, in publications that address specific diagnostic challenges, in collaborations that span institutions and countries.

Whether that belief proves justified will depend on factors beyond any individual researcher's control: regulatory frameworks, clinical adoption, the broader politics of healthcare systems. But the work itself, the careful development of systems that can distinguish cancer growth patterns and detect immune cells and identify genetic translocations, represents something worth paying attention to.

The diagnostic eye, augmented. The question is what it will see.

The question of what it means to see, and to see accurately, is not merely technical. It is cultural. It is ethical. It is, increasingly, computational. For those building Europe's AI future, the conversation happening in Vienna on May 19 at Human x AI Europe offers a rare chance to engage with precisely these stakes.