Counting Cells in Seconds
A lensless imaging system finds and recognizes the shadows of T cells and bacteria.
Clinical tests for identifying and counting normal and bacterial cells in blood and other samples can tell doctors the source of a bacterial infection or help them monitor the immune health of people with HIV. But conventional cell counting is costly and time-consuming. A simple, lensless imaging system being developed by researchers at the University of California, Los Angeles, uses a chip like the one found in a digital camera to count and distinguish different types of cells in blood and drinking water, and simple algorithms to identify and count the cells. The imager could be carried in a device the size of a cell phone and used to monitor water quality and to provide cheap diagnostics in rural and underdeveloped areas.
The imager can find small numbers of cells in a large, unprocessed sample. A water or blood sample is loaded onto a glass slide above a light-sensing chip identical to those used in consumer digital cameras; then it’s illuminated from above. “What we record is not an image but a diffraction signature,” says Aydogan Ozcan, an assistant professor of electrical engineering at UCLA who’s developing the cell counter. Unlike conventional microscopes, which take detailed pictures of very small samples, Ozcan’s diffraction technique is rapid and inexpensive. The blurred, pixellated images created by his cell counter are of such low quality that Ozcan doesn’t call the system a microscope. But these images contain just enough information to identify and count cells, which is all that’s needed for many clinical diagnostic applications.
Cell counting is usually done using machines called flow cytometers, which cost up to hundreds of thousands of dollars. The technique must be performed in the lab and requires multiple steps. Conventional microscopes can also be used to find and count cells, but microscopes are costly and the process is complex. “If you wanted to screen for a few bacterial cells in a few milliliters of water, you’d need to do hundreds of tests with a regular microscope,” says Ozcan.
In Ozcan’s method, as light passes through a given type of cell, the light diffracts or bends in a characteristic way. Each cell type has a unique diffraction signature that depends on its size, shape, and an optical quality called refractive index. Ozcan has compiled a library of characteristic diffraction signatures for different cell types. After his cell counter takes an image, it quickly consults his library to determine the number of cells of each type in the sample. These calculations don’t require much processing power and could be done in a mobile device such as a cell phone, says Ozcan.
The counter has high throughput–while it’s capable of detecting small numbers of cells, it can image as many as 100,000 cells in a 20-centimeter-squared field of view in one second. The counter can, for example, determine the concentration of red blood cells in an unprocessed blood sample with 90 percent accuracy. Red blood cell count can be used to diagnose anemia, to monitor malaria, and to monitor patients’ responses to chemotherapy.
“What [Ozcan] is doing has potential for hand-held devices that work in the field,” says Alexander Revzin, an assistant professor of biomedical engineering at the University of California, Davis. Rezvin has begun a collaboration with Ozcan to develop a cheap, diffraction-based test for counting T cells in HIV patients–a measure of the health of the immune system that’s used to determine when to start drug treatment and whether it’s working. “Obviously a poor-resource setting is one target, but it doesn’t just need to be used in Africa if this is a robust technology,” says Rezvin.
“This is a very practical technique,” says Mehmet Fatih Yanik, an assistant professor in the Research Laboratory of Electronics at MIT. “Ozcan’s work can significantly reduce the cost and effort required for cell counting, allowing its commonplace use even in Third World countries for a variety of medical applications.”
So far, Ozcan’s group has developed protype cell counters on the lab benchtop. Next, he says, he’ll convert a cell phone into a mobile diagnostic lab by taking out the camera lens and putting in the imaging chip and a mechanical system to load microscope slides.
AI and robotics are changing the future of work. Are you ready? Join us at EmTech Next 2019.Register now