UCLA researchers have developed a new laser-based technology to rapidly screen blood samples for the presence of cancer cells. The label-free system measures 16 different physical characteristics of each cell and analyzes the data to identify whether the cell is cancerous. Not having to introduce any labeling chemicals and being gentle on the cells, the technique leaves the cells alive and available for further inspection using other means.
It relies on a photonic time stretch microscope and a computer that runs deep learning artificial intelligence algorithms. The microscope can take millions of images per second thanks to unusual optics that produce high quality shots even at this speed. The deep learning system can actually run a variety of algorithms and the researchers tested a few to see which are better than others at spotting cancer cells.
From the study abstract in Scientific Reports:
Our system captures quantitative optical phase and intensity images and extracts multiple biophysical features of individual cells. These biophysical measurements form a hyperdimensional feature space in which supervised learning is performed for cell classification. We compare various learning algorithms including artificial neural network, support vector machine, logistic regression, and a novel deep learning pipeline, which adopts global optimization of receiver operating characteristics. As a validation of the enhanced sensitivity and specificity of our system, we show classification of white blood T-cells against colon cancer cells, as well as lipid accumulating algal strains for biofuel production. This system opens up a new path to data-driven phenotypic diagnosis and better understanding of the heterogeneous gene expressions in cells.
Study in Scientific Reports: Deep Learning in Label-free Cell Classification…
Source: UCLA…