By 'better', Prof. Schut refers to improving care for patients, but also to bringing improvements to the diagnostic and research laboratories. By 'bytes', he means smart technology, in particular AI - the ability of computer algorithms to draw conclusions without direct human interference. Prof. Schut is particularly focused on the question how scientific AI discoveries can be reliably and safely fast-tracked into the lab and in patient care, and how do you monitor, evaluate and evolve these applications?
Amsterdam UMC is already using AI in various ways, both in the research laboratory and in the clinical domains. One example is the treatment of patients with rheumatoid arthritis. These patients are first prescribed the effective, safe, and inexpensive drug methotrexate, but methotrexate does not work for everyone. An algorithm has been developed that can predict who will not respond to this drug, so other - more effective - medication can be prescribed immediately.
In the oncological care domain, Prof. Schut has created AI tools to analyze notes from general practitioners to develop clinical prediction models for the early detection of lung cancer. The clinical chemistry lab and the CCA have use AI-guided sequence analysis of mRNA in platelets for early detection of cancer. In current research of Schut, with the Liquid Biopsy Center AI is used for normalization of low-input sequencing data, which is of crucial importance for application of biomarker models in the clinic.
The more data an algorithm receives, the better it becomes at doing its job. But researchers and clinicians still are not sufficiently aware that AI models are extremely powerful tools in analyzing complex data. Therefore, Prof. Schut sees it as his mission to change this paradigm: "Unknown makes unloved,” he says. “That may be the reason why complex AI models are not applied more. We have a calling to spread the word."
For more information contact Prof. Martijn Schut.