how can we help you
For a typical microscope, there is a trade-off between spatial resolution and depth of field, which means that only objects at the same distance from the lens can focus clearly. Even a few millionths of a meter near or far from the objective of a microscope, the features will be blurred. As a result, lab microscope samples are usually thin and mounted between slides.
At present, the section is used to examine the margins of the tumor, and it is not easy to prepare. The removed tissue is usually sent to a hospital laboratory, where experts either freeze it or prepare it with chemicals, then make extremely thin sections and attach them to slides. This process is time-consuming and requires specialized equipment and trained workers. Hospitals rarely can examine slides at the edge of tumors during surgery, and relevant health institutions in many parts of the world lack the necessary equipment and expertise.
DeepDOF uses a deep learning neural network and an expert system, which can study a large number of data to make decisions similar to human. The researchers showed it 1200 images from a database of tissue slices. From this, DeepDOF learned how to choose the best phase mask to image a specific sample, and how to remove blur from the image captured from the sample to focus cells of different depths.
Richards-Kortum, Rice’s Malcolm Gillis University Professor, professor of bioengineering and director of the Rice 360° Institute for Global Health said, “A clinical study is needed to find out whether DeepDOF can be used as proposed for margin assessment during surgery. We hope to begin clinical validation in the coming year.”