Adding new functions to the standard optical microscope, researchers from the University of Illinois at Urbana-Champaign are hoping to move cancer diagnostics in the digital age.
Combining infrared measurements with optical images of high resolution and machine learning algorithms, the researchers created a digital biopsy, which are associated with conventional methods of pathology and surpassed modern infrared microscopes. Under the leadership of Rohit Bhargava, a professor of bioengineering and director of the Cancer Center in Illinois, the group published the results in the Proceedings of the National Academy of Sciences.
"The advantage is that it is possible to measure how the organization of cells and their chemical composition. The measurement of the chemical composition of the tumor cells and their microenvironment lead to better cancer diagnosis and understanding of disease ", - says Bhargava.
Standard tissue pathology is to add dyes to pathologists can see the shape and structure of cells under a microscope. However, it is sometimes difficult to distinguish cancer from healthy tissue and to determine the boundaries of the tumor, so in many cases, the diagnosis is subjective.
"For more than a century, we are adding dyes in a biopsy of human tissue for diagnosis of tumors. However, the shape and color, caused by dyes give very limited information about the underlying molecular changes that cause cancer ", - says Bhargava.
Technologies such as infrared microscopy, can measure the molecular composition of tissue, providing a quantitative measure which may differentiate cell types. Unfortunately, the infrared microscopes are expensive, and require special sample preparation and processing, making them impractical for most clinical and research conditions. Bhargava Group has developed a hybrid microscope, adding an infrared laser and a special microscope lens, called an interference lens to the optical chamber. Infrared measures the optical hybrid as infrared data and the optical image with high resolution using a light microscope. This type is widespread in hospitals and laboratories.
Scientists have created a hybrid microscope of ready-made components, upgrade it. It has a high resolution, large field of view and the availability of an optical microscope. Furthermore, infrared data may be analyzed computationally, without the addition of any dyes that can damage tissue. The software can recreate a variety of spots or even block them, to create a more complete, fully digital picture of what is in the fabric.
Researchers checked by microscope imaging tissue samples of breast cancer and compared the results with the results of hybrid microscope conventional staining techniques. They found that the infrared-optical hybrid of superior modern technology in infrared microscopes in several ways: it has a 10 times greater coverage, and four times higher resolution, allowing you to acquire images of large size infrared in unprecedented detail.
"Infrared optical microscopy hybrid is compatible with conventional microscopy in biomedical applications - the researchers say. - We combine the ease of use and universal accessibility of optical microscopy with a wide palette of infrared molecular contrast and machine learning. And thus we hope to change the way we usually understand the microscopic structure of tissue. "
The researchers plan to continue improving the computational tools used for the analysis of hybrid images. They are working on the optimization of machine learning programs that can measure multiple wavelengths of infrared light, creating images that are easy to distinguish between different types of cells, and integrate this data with detailed optical images for precise mapping of cancer in a sample. They also plan to explore other areas of application of hybrid microscopy, such as forensic science, the science of polymers and other biomedical applications.