Analysing stem cells with artificial intelligence
Stem cells are particularly interesting for medical research as they can "morph into any other cell, be it skin or heart muscle cells", said Dr. Johannes Bauer, joint founder of the biotechnology start-up TRI Thinking Research Instruments in Hamburg. From the researchers' point of view, stem cells should multiply but remain in their original state, if possible. However, precise knowledge of the cells exact state is crucial to research projects and successful experiments. This is where "Visual Artificial Intelligence and Drugs Research (VAIDR) comes in - an all-in-one system consisting of microscope, computing hardware and software, developed by Dr Bruno Chilian, a co-founder of TRI, and Bauer.
Generating comparable data
AI analyses the state of cells, the degree of differentiation and their quality as cells may die depending on the experiment. "Our product handles image data acquisition, permanent and secure management of primary and metadata, and provides automatic, objective, consistent AI-assisted image analysis and evaluates the results," Bauer pointed out. Depending on the project, thousands, sometimes hundreds of thousands of images have to be analysed and theoretically viewed through a microscope which is ideal for AI. Bauer remarked: "AI has the huge advantage of not tiring after lunch, but continues to evaluate based on the same benchmarks thereby generating comparable data." The integrated solution can be trained to address any issue, such as the search for new active substances in new medicines. "And the user does not have to have any knowledge of AI for that task," he noted.
SAIBR - cell analysis in bioreactors
VAIDR is designed to relieve researchers of time-consuming, monotonous analytical tasks which can cause human beings to make mistakes. The system is currently being used by the Institute of Human Genetics at the University Hospital Hamburg-Eppendorf (UKE), the Max Delbrück Centre for Molecular Medicine in Berlin and the cell biology group at Public Health England. Bauer and Chilian are now gearing up to launch their "Seeing AI for Bio Reactors" (SAIBR) product. Bauer explains the approach using the example of stem cells and implant therapy. Tissue is grown from stem cells to to implant healthy heart tissue after a heart attack. "Large quantities of stem cells, which are grown in large containers called bioreactors, are needed for this research. However, we cannot look inside to see how the cells are developing. SAIBR allows us to monitor the cultures on a large scale."
Read the other parts of our series: