AI-driven digital pathology and advanced imaging for early, precise, and personalized skin cancer diagnostics

Our Research Areas

Computational Pathology and Artificial Intelligence

The digitalization of pathology creates new opportunities to use artificial intelligence (AI) to improve cancer diagnostics and to uncover information in routinely stained tissue sections that cannot be identified through conventional histopathological assessment.

We use large collections of digitized H&E-stained whole-slide images to develop AI-based methods for the diagnosis and classification of skin tumours, detection of metastases, and prediction of tumour progression and treatment response. A particular focus is melanoma, where we develop computational biomarkers to identify tumours with a high risk of metastasis and to predict clinically relevant molecular alterations and treatment response.

By combining histopathology with clinical, genomic and molecular data, our goal is to develop robust biomarkers that can enable earlier identification of aggressive tumours, improve prognostic assessment and support more personalized cancer treatment.

Ex Vivo Fluorescence Confocal Microscopy

Ex vivo fluorescence confocal microscopy (FCM) enables the examination of fresh, unfixed tissue and the generation of high-resolution digital images within minutes.

We develop and evaluate FCM for rapid pre- and intraoperative diagnosis of skin cancer, assessment of surgical margins and detection of melanoma metastases in lymph nodes.

Our goal is to develop faster diagnostic workflows that can provide clinically relevant information at the time of a procedure, thereby reducing waiting times and contributing to more efficient and personalized cancer care.

Molecular Imaging and Cancer Lipidomics

Changes in tumour-cell metabolism are an important part of cancer development and progression. We are particularly interested in how the composition and spatial distribution of lipids change as tumours develop and become more aggressive.

Using Time-of-Flight Secondary Ion Mass Spectrometry (ToF-SIMS), we can map lipids and other molecules directly in tissue at high spatial resolution. This molecular information can be compared with histopathological changes in corresponding tissue regions.

Our studies have identified changes in tumour-cell lipids during melanoma progression. By combining molecular imaging with histopathology and clinical data, we aim to identify novel biomarkers of tumour progression and aggressiveness and to improve our understanding of the molecular changes that occur during cancer development.

Current collaborators:

• Prof John Paoli, Department of Dermatology, Institute of Clinical sciences, Sahlgrenska Academy
• Prof Roger Olofsson Bagge, Department of Surgery, Institute of Clinical sciences, Sahlgrenska Academy
• Adjunct Professor Max levin, Department of Oncology, Institute of Clinical sciences, Sahlgrenska Academy
• Associate professor Kari Nielsen, Department of Clinical Sciences, Lund University
• Prof John Fletcher, Department of Chemistry and Molecular Biology, University of Gothenburg
• Associate professor Ida Häggström, Chalmers University of Technology and Gothenburg University
• Associate professor Leonardo Vinicius Monteiro de Assis, Department of Chemistry and Molecular Biology, University of Gothenburg
• Associate Professor Aino Rönkä, Department of Oncology, University of Eastern Finland and Kuopio University Hospital
• Associate Professor Jaana Hartikainen, Research Director; Institute of Clinical Medicine, School of Medicine, Faculty of Health Sciences, University of Eastern Finland
• Associate Professor Gabriele Campanella, Senior data scientist, Icahn School of Medicine at Mount Sinai, NYC
• Professor Klaus Busan, Department of Pathology, Memorial Sloan Kettering Cancer Center, NYC

News and Media

Faster treatment for lung cancer with AI

Faster and more accurate skin cancer diagnoses with AI and advanced imaging techniques