When it comes to tumor removal, the main goal is to remove all cancerous cells from the body, but the only way to know if this was successful is to look at the tumor under a microscope after surgery. Fortunately, there may soon be a more effective technique to ensure patients are cancer cell-free with an AI-assisted microscope.
Developed by researchers at Rice University, the microscope can rapidly image large sections of tissue to allow doctors to examine tumor margins in real-time. The deep learning extended depth-of-field microscope, also known as DeepDOF, uses deep learning AI technology to speed up image collection and processing. Unlike traditional microscopes, which can only focus on objects on a level plane, there is no required trade-off between spatial resolution and depth-of-field, so the images are incredibly precise. DeepDOF is the first microscope designed with coexisting imaging and image analysis.
Traditional tumor analysis requires that the sample be prepared and sliced into extremely thin layers before being assessed under a microscope. Although accurate, this process is costly and time-consuming, meaning patients must wait for lab work to come back to determine the success of their surgery. DeepDOF uses a standard optical microscope in combination with an inexpensive optical phase mask and offers high-level imaging in as little as two minutes.
Clinical research is necessary to determine the technology’s success rate with actual surgical cancer patients, but if put into practice, DeepDOF would allow doctors to see tumor removal details nearly instantly and increase instances of complete cancer cell eradication.
