Abstract: The optical microscope has been invaluable for scientific discovery. To visualize biological structures, contrast agents need to be added to samples and images are often interpreted using human visual cues.
Here we describe the evolution and availability of a new tool – chemical imaging, in which dyes and stains are not needed as the chemical composition of tissue is directly recorded via infrared spectroscopy and provides image contrast. Artificial intelligence algorithms subsequently parse the vast data to provide informative views.
We first describe the progress in this technology and current capabilities, showing how a convergence of theoretical advances, new components in the mid-infrared and new optical designs are transforming measurement science. Next, we show how modern machine learning techniques are not only improving information extraction but promise to make measurements faster and cheaper as well. We provide examples of how this technology can help understand composition, molecular changes and impact of physio-chemical organization in cancer samples.
One popular avenue for the application of AI in medicine has been digital pathology. We show how chemical imaging enables new opportunities for all-digital molecular pathology as well as providing key data for conventional pathology as well. Finally, we describe a new model for translating technologies such as the one described here via the Cancer Center at Illinois. The CCIL was established to accelerate diagnostics and therapeutics by a convergence of engineering and cancer.
Bio: Rohit Bhargava is Founder Professor in Bioengineering and serves as the Director of the Cancer Center at Illinois of the University of Illinois at Urbana-Champaign. He graduated with a dual-degree B.Tech. (1996) from the Indian Institute of Technology, New Delhi, and received a doctoral degree from Case Western Reserve University (2000).