Search
Menu

AI, Lasers Automate Rapid Sorting and Analysis of Live Cells

Facebook X LinkedIn Email
To realize the potential of personalized medicine and overcome crisis situations like the recent pandemic, labs need an efficient way to isolate living cells for analysis and testing. The ability to isolate specific cell types without impairing cell vitality is also necessary to advance pharmaceutical research.

In response to this need, the Fraunhofer Institutes of Laser Technology (ILT) and Production Technology (IPT) developed an AI-assisted tool that automatically sorts and isolates living cells from samples using a high-throughput process. The technology, called LIFTOSCOPE, combines high-speed microscopy, AI-based analysis, and localization of living cells and cell clusters with laser-induced forward transfer (LIFT).
The MIR LIFT process transfers living cells to microtiter plates one after the other at a rate of up to 100 Hz. To integrate the highly efficient process into microscopy platforms regardless of the manufacturer, Fraunhofer ILT has developed, among other things, the holder shown here for a six-well microtiter plate. Courtesy of Fraunhofer ILT.
The MIR LIFT process transfers living cells to microtiter plates one after the other at a rate of up to 100 Hz. To integrate the highly efficient process into microscopy platforms regardless of the manufacturer, Fraunhofer ILT has developed, among other things, the holder (shown) for a six-well microtiter plate. Courtesy of Fraunhofer ILT.

The LIFTOSCOPE can localize, identify, and analyze dozens of living cells per second. It can complete a single cell transfer within 200 μs and can activate and transfer 10,000 cells to microtiter plates within 100 s.

To create LIFTOSCOPE, the project team integrated an AI-assisted, high-throughput process into a commercially available inverted microscope and equipped the microscope with a high-speed camera and a flash light source. It integrated the mid-infrared (MIR) LIFT process for cell transfer directly into the beam path of the microscope.

The camera system that is connected to the microscope delivers 100 high-resolution images per second. The AI technology uses semantic segmentation to identify the desired cell types in the image data. The AI also determines the cells’ exact position and center of gravity. The AI system can be trained to recognize pluripotent stem cells, high-producer cells, and immune cells.

After the cells are sorted and analyzed, they are transferred to microtiter plates by using the MIR LIFT process. The high-precision, optically monitored, laser-based process used in LIFTOSCOPE allows it to accurately control and transfer cells. The survival rate for transferred cells is over 90%. “Depending on the cell type, up to 100% of all cells survive this procedure,” Nadine Nottrodt, head of the Biofabrication Group at Fraunhofer ILT, said.

The cells are transferred at a high-frequency rate of up to 100 hertz (Hz). A holder for a six-well microtiter plate, developed by the team, makes it easier to integrate the technology into any microscopy platform, regardless of the manufacturer.

The LIFT process is fast and simple, the researchers said. Using a 9-ns laser pulse that requires just a few microjoules (µJ) of pulse energy, LIFT stimulates a liquid medium located directly under the targeted cell to form a vapor bubble. The cell, previously released from its bond, is briefly lifted by the bubble. As soon as the bubble collapses, suction is formed that flushes the cell into the culture vessel of the microtiter plate.


“The cells are randomly distributed in the samples. For this reason, our system follows a predefined grid and transfers cells that are within a radius of 50 micrometers of the focal point,” project manager Richard Lensing said.

Although fluorescent markers can be used to identify specific cells, the LIFT process performs well even without additives. The AI technology provides precise localization to ensure that the cells are captured by the jet and transported into the microtiter plates. The LIFT process uses an MIR 2940-nm wavelength laser to directly excite the liquid medium. This wavelength is hardly absorbed by the polymers in the sample carriers. 
Actin-stained 3T3 fibroblast spheroid for LIFT transfer, cultivated in laser-fabricated microwells. Courtesy of Fraunhofer ILT, Aachen, Germany.
Actin-stained 3T3 fibroblast spheroid for LIFT transfer, cultivated in laser-fabricated microwells. Courtesy of Fraunhofer ILT, Aachen, Germany.

The team developed two strategies for moving the cell culture — a stop-and-go approach and a continuous process. The stop-and-go approach provides the ability to sort samples with many different cells, which makes sample preparation easier. However, it also reduces efficiency. “In stop-and-go operation, a short rest phase must be inserted before and after the cell LIFT, because each stop triggers hydrodynamic currents in the sample, which must first settle before the next cell can be transferred,” Nottrodt said.

When the continuous process is used, the LIFTOSCOPE scans the sample carriers in a grid of up to 1600 lines with 50-μm spacing, and transfers every cell that comes into focus during this continuous movement. The more cells that are transferred, the more time that is saved by using this approach. When 10,000 cells are transferred, the continuous process is more than twice as fast as the stop-and-go strategy. With a transfer of 100,000 cells, it is 20 times faster than stop-and-go operation.

The team aims to stabilize the high throughput of the automated cell recognition and LIFT process, limiting the time required to complete one microtiter plate to 10 minutes. High-precision actuators for imaging and positioning the laser focus in the process cycle will ensure that the LIFT process has the resolution required for AI-supported cell detection and measurement. This will also ensure that the laser focus is positioned directly below the cell to within 25 μm.

This AI-based, laser-based process could fully automate the task of isolating living cells. According to Nottrodt, based on the progress made by the project team, it is feasible that cell LIFT could be synchronized with the image frequency of the high-speed camera, which would enable a single-cell sorting rate of 100 cells per second. The next step for the team will be to attain market maturity with a prototype of the process.

The Fraunhofer ILT and IPT will present the LIFTOSCOPE device at analytica 2024, to be held April 9-12 in Munich.

Published: March 2024
Glossary
positioning
Positioning generally refers to the determination or identification of the location or placement of an object, person, or entity in a specific space or relative to a reference point. The term is used in various contexts, and the methods for positioning can vary depending on the application. Key aspects of positioning include: Spatial coordinates: Positioning often involves expressing the location of an object in terms of spatial coordinates. These coordinates may include dimensions such as...
artificial intelligence
The ability of a machine to perform certain complex functions normally associated with human intelligence, such as judgment, pattern recognition, understanding, learning, planning, and problem solving.
machine learning
Machine learning (ML) is a subset of artificial intelligence (AI) that focuses on the development of algorithms and statistical models that enable computers to improve their performance on a specific task through experience or training. Instead of being explicitly programmed to perform a task, a machine learning system learns from data and examples. The primary goal of machine learning is to develop models that can generalize patterns from data and make predictions or decisions without being...
Research & TechnologyeducationEuropeFraunhofer Institute of Laser TechnologyImagingLasersLight SourcesOpticspositioningSensors & Detectorspulsed laserscamerasMicroscopyBiophotonicscoronaviruscancermedicalpharmaceuticalpersonalized medicineartificial intelligencemachine learningLIFTOSCOPEMIR LIFTBioScan

We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.
Vision Spectra Conference 2024LIVE NOW: 3D Imaging of Transparency: The Next Frontier in Automated Item Picking X