AI speeds up identification brain tumor type
AI speeds up identification brain tumor type
What type of brain tumor does this patient have? AI technology helps to determine this as early as during surgery, within 1.5 hours. This process normally takes a week. The new technology allows neurosurgeons to adjust their surgical strategies on the spot. Today, researchers from UMC Utrecht and researchers, pathologists and neurosurgeons from the Princess Máxima Center for pediatric oncology and Amsterdam UMC have published about this study in Nature.
Every year, 1,400 adults and 150 children are diagnosed with a tumor in the brain or spinal cord in the Netherlands. Surgery is often the first step taken in treatment. Currently, during surgery, neurosurgeons do not precisely know what type of brain tumor and what degree of aggressiveness they are dealing with. The exact diagnosis will usually only be available one week after surgery, after the tumor tissue has been visually and molecularly analyzed by the pathologist.
Deep-learning algorithm
Researchers from UMC Utrecht have developed a new ‘deep-learning algorithm’, a form of artificial intelligence, which significantly speeds up diagnosis.
Jeroen de Ridder, research group leader within UMC Utrecht and Oncode Institute: “Recently, Nanopore sequencing became available: a technology that helps to read DNA in real time. For this, we developed an algorithm that is equipped to learn from millions of simulated realistic ‘DNA snapshots’. With this algorithm, which we named ‘Sturgeon’, we can identify the tumor type within 20 to 40 minutes. And that is fast enough to directly adjust the surgical strategy, if necessary.”
Jeroen is in charge of a ‘bioinformatics lab’, consisting of 15 computational scientists. His team uses the latest developments in computer science, such as machine learning and artificial intelligence (AI), to analyze complex molecular datasets. That data is retrieved, for example, from tumor tissue taken from patients and is collected in biobanks.
"Modern technologies allow us to make enormously complex and rich measurements of, for instance, tumor biopsies," Jeroen says. "How do we ensure that that highly complex collection of measurement data leads to new fundamental insights about cancer? And how can we use that collected data to better diagnose and treat cancer?"
"To answer these questions, it is essential to design algorithms that can analyze large collections of molecular data, and that is exactly what bioinformatics focuses on. Although our research is fundamental in nature, we are driven to make sure that our findings will have a positive effect on patients' lives."
In deze folder bevindt zich extra informatie door middel van een video. Scan de bovenstaande QR-code met uw telefoon om een animatie over dit onderwerp bekijken. Of bekijk de video via:
Tested and trained with biobank
Analyzing large amounts of data faster and more precisely is what Jeroen and his team do. For this specific new algorithm, they cooperated closely with Bastiaan Tops, head of the Laboratory for Pediatric Oncology at the Princess Máxima Center.
Bastiaan brought together the new technology and the needs from the operating room, which was partly made possible by funding from the KiKa foundation. He provided the large amount of data that was needed to develop the algorithm: the Princess Máxima Center has maintained an extensive biobank for years. Among other things, this biobank stores tissue from children with brain tumors. The algorithm was trained and tested using this biobank.
“That we can now determine the type of brain tumor as soon as during surgery shows how technology can speed up diagnostics. And how we can use an existing biobank to develop a new technology, with great impact for future patients,” says Bastiaan.
Used during surgery
Consequently, the whole procedure was also carried out several times during actual brain surgeries, from taking the tissue in the operating room to determining the tumor type. In Utrecht, this happened with children, and in Amsterdam with adults. Total length of the procedure: 60 to 90 minutes.
The Princess Máxima Center has decided that the results of the technique are sufficiently reliable and is already using it with children for whom the outcome may determine the surgical strategy. Amsterdam UMC will also use the technique in daily practice, to help them speed up diagnosis.
Eelco Hoving, pediatric neurosurgeon and clinical director of neuro-oncology at the Máxima Center, is excited about the possibilities of DNA analysis during surgery: “During surgery, a small remnant of tumor tissue is sometimes deliberately left behind to prevent neurological damage. But if it will later turn out, for example, that the tumor is very aggressive, a second surgery may still be necessary to remove that last remnant. This will again create risks and stress for patients and their families. This can be avoided now because we will already know during the first surgery what type of tumor we are dealing with."
Comparative research
To use the new technique even more widely and structurally, more research is needed. For instance, more tumor types should be added to the algorithm. In this way, international standards will be met, allowing comparison of data. Also, the outcomes of the new and current (lengthier) method will be further compared, in collaboration with other (inter)national centers. This should make clear whether the new method will also contribute to patients' quality of life in the long term.
Jeroen de Ridder: “It is truly exciting that we have been able to actually make the step into clinical practice by combining all areas of expertise, from basic researchers to pathologists and surgeons. By doing so, we can help surgeons to optimize the outcome of brain tumor surgery.”
More information
'Ultra-fast deep-learned CNS tumour classification during surgery’: the Nature publication.
Oncode Investigator Jeroen de Ridder discusses new research on IO sequencing: video by Oncode Institute.
Sturgeon live: video showing the entire process, from operating room to real-time DNA analysis and back.
Health AI Labs
Artificial intelligence (AI) is part of UMC Utrecht's strategy. The innovative technology helps to work towards the healthcare of tomorrow (‘de zorg van morgen’), which should be innovative, sustainable, affordable and health professional friendly. All Health AI initiatives of UMC Utrecht and Utrecht University have been assembled, in order to promote wider and faster AI implementation in healthcare, education and research.
UMC Utrecht and Utrecht University house five Health AI Labs. Jeroen de Ridder is one of the directors of the AI Lab for Molecular Medicine. Here, we investigate how AI can improve the analysis of molecular data. This involves the use of large and unique biobanks and patient cohorts (studies in which large groups of patients are followed for long periods of time).