Junior Research Group Digital Biomarkers for Oncology

Dr. Titus Brinker

Artificial intelligence methods are applied to identify digital biomarkers - as shown here in a tissue section of the skin - in order to enable more precise diagnostics and improved therapy selection.
© dkfz.de

The development of digital biomarkers for non-invasive early detection, individualized diagnostic and therapeutic approaches as well as the prediction of tumor behavior is the research focus of our interdisciplinary research group. Scientists from the fields of medicine, molecular biology and data science focus on identifying patterns in images and genetic data with state-of-the-art techniques of artificial intelligence.

Two major projects coin our current activity: The Skin Classification Project (SCP) focuses on improving the accuracy of skin cancer screenings which currently affect ten million citizens in Germany alone. Via a broad network of 12 university hospitals throughout Germany, we develop and optimize AI-classifiers for clinical routine. Our long-term goal is to demonstrate successful on-site implementation in prospective clinical trials, making AI-assisted skin cancer diagnostics the new standard of care.

In a second project, the Tumor Behavior Prediction Initiative (TPI), tumor behavior is predicted on the basis of histologic slides and clinical follow-up data. A crucial advantage of this approach is that tissue samples are taken anyway in clinical cancer diagnostics. With the help of artificial intelligence, biomarkers are determined in the digitized slides, which in future could thus be used as a decision-making aid by the treating physician with minimal additional time and financial effort and represent a further building block for personalized medicine.

A special research focus of our research group is always to directly serve our oncologic patients at hand. Therefore, methods are developed to improve the explainability of the models and their robustness in everyday use. Our broad network of clinical cooperation partners as well as patient representatives is closely involved in the development of research questions and the subsequent data acquisition.


Dr. Titus Brinker
Digital Biomarkers for Oncology (C140)
Deutsches Krebsforschungszentrum
Im Neuenheimer Feld 223
69120 Heidelberg
Tel: +49 6221 425301

Selected Publications

  • Hekler A, Utikal JS, Enk AH, Hauschild A, Weichenthal M, Maron RC, Berking C, Haferkamp S, Klode J, Schadendorf D, Schilling B, Holland-Letz T, Izar B, von Kalle C, Fröhling S, Brinker TJ; Collaborators (2019). Superior skin cancer classification by the combination of human and artificial intelligence. Eur J Cancer, 120, 114-121
  • Brinker TJ, Hekler A, Enk AH, et al (2019). Deep learning outperformed 136 of 157 dermatologists in a head-to-head dermoscopic melanoma image classification task. Eur J Cancer, 113, 47-54.
  • Brinker TJ, Hekler A, Enk AH, Berking C, Haferkamp S, Hauschild A, Weichenthal M, Klode J, Schadendorf D, Holland-Letz T, von Kalle C, Fröhling S, Schilling B, Utikal JS (2019). Deep neural networks are superior to dermatologists in melanoma image classification. Eur J Cancer, 119, 11-17
  • Hekler A, Utikal JS, Enk AH, Solass W, Schmitt M, Klode J, Schadendorf D, Sondermann W, Franklin C, Bestvater F, Flaig MJ, Krahl D, von Kalle C, Fröhling S, Brinker TJ (2019). Deep learning outperformed 11 pathologists in the classification of histopathological melanoma images. Eur J Cancer, 119, 91-96
to top
powered by webEdition CMS