In the ever-evolving landscape of medical technology, a new study is published about ultra-fast deep-learned tumor classification during surgery (Vermeulen et al., 2023). While it might sound complex, in simple terms, it’s an AI technology that can determine the specific type of brain tumor a patient has (UMC Utrecht, 2023). This process normally takes a week, but with this new technology called ‘Sturgeon’, it can be done within 1.5 hours.
Currently, during surgery, neurosurgeons lack precise information regarding the specific type and level of aggressiveness of the brain tumor they are dealing with. They have to decide whether to remove some healthy brain tissue surrounding the tumor to ensure the entire tumor is removed, as failing to do so may leave malignant cells behind, potentially leading to the formation of a new tumor. The exact diagnosis typically becomes available one week after surgery, following the visual and molecular analysis of the tumor tissue by the pathologist. However, if it than turned out, for instance, that the tumor is highly aggressive, a second surgery may still be required to eliminate the remaining malignant cells. This will again create risks and anxiety for patients and their families (UMC Utrecht, 2023).
Now researchers from UMC Utrecht have developed a new method that uses Artificial Intelligence (AI) to provide real-time information about the status of the cell during surgery. This assists surgeon in making decisions regarding which tissue to remove and what to leave untouched (Leitch, 2023). The underlying technology behind this method is Nanopore sequencing. Jeroen de Ridder, research group leader within UMC Utrecht explained this as follows: ‘Nanopore sequencing is a technology that help 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’ (UMC Utrecht, 2023).
Sturgeon, with its ability to identify brain tumor types within just 1.5 hours during surgery, ‘represents how technology can speed up diagnostics’ (UMC Utrecht, 2023). This AI-driven innovation is not only reducing the time patients spend in uncertainty, but also empowering neurosurgeons to make more informed decisions on the operating table. As Sturgeoncontinues to advance, the possibilities for improving brain tumor diagnosis and treatment are boundless. With this cutting-edge technology, we are witnessing the dawn of a new era in the fight against brain tumors.
References
AI speeds up identification brain tumor type. (2023). UMC Utrecht. https://www.umcutrecht.nl/en/over-ons/nieuws/strategic-program-cancer/ai-speeds-up-identification-brain-tumor-type
Leitch, C. (2023, October 23). Sturgeon – This AI Diagnoses Brain Tumors During an Operation. Labroots. https://www.labroots.com/trending/clinical-and-molecular-dx/26033/sturgeon-ai-diagnoses-brain-tumors-operation
Vermeulen, C., Pagès-Gallego, M., Kester, L., Kranendonk, M. E. G., Wesseling, P., Verburg, N., De Witt Hamer, P. C., Kooi, E. J., Dankmeijer, L., van der Lugt, J., van Baarsen, K., Hoving, E. W., Tops, B. J., & de Ridder, J. (2023). Ultra-fast deep-learned CNS tumour classification during surgery. Nature. https://doi.org/10.1038/s41586-023-06615-2
Fascinating post!
You wrote the post in an easygoing manner, which made it enjoyable to read. Besides that, it is interesting to learn more about how AI can contribute to surgical efficiency. I can see this technology being used in the future in more ways than fighting tumors! However, I do wonder what the limitations and challenges of applying AI in this process are, for example, I can imagine the quality of data input might be critical. Let’s hope the innovation of AI in healthcare keeps expanding!