Improving cancer diagnosis and treatment using data analytics

Project overview

Cancers develop from cells in our bodies that divide in an uncontrolled manner. While tumours develop over time, cancer cells progressively acquire more changes in their DNA, which could include small mutations or larger gains and losses of parts of the DNA or chromosomes.

Large datasets that document these genetic and genomic changes in human cancers are becoming increasingly more readily available.

This enables us to perform data analytics in order to better understand how cancers develop over time, in individual cancer patients, in groups of patients with the same type of cancer, as well as how this differs between patients affected by different types of cancer, such as breast, lung, brain cancer or melanoma.

We can also use cancer genetic data to predict patient prognosis and improve diagnosis.

Finally, the genetic abnormalities change how cancer cells respond to drugs. Therefore, we can use data analytical methods, such as machine learning, to identify new combinations between cancer genetic changes and drug sensitivity or drug resistance. Our goal is to improve personalized cancer treatment by predicting which drugs will work best for individual cancer patients.


  • Better understanding of how cancers develop over time.
  • Predict prognosis of cancer patients.
  • Improved diagnosis of cancer.
  • Improved treatment of cancer.

Project team

For more information about this project, contact Dr Pascal Duijf.


Melanoma cancer cells of human under microscope