Measuring the mechanical properties of living tissue is a challenging task due to the small sizes and the fragility of the living organisms. A promising method, which works best on small scales, is the passive microrheology, which observes the motion of tracing beads within the sample. The video imaging method observes this motion by imaging the tracer particles with suitable optics (e.g. a microscope). As living tissue is a complex material, the viscoelastic properties are highly frequency dependent; therefore, a fast high-speed camera is needed to resolve the important frequencies in the 100 to 1000 Hz regime. As the data rate of high-speed cameras exceed the storage speed, only short burst of measurements can be carried out. This leads to a limited dynamic range of frequencies and missed measurement opportunities. It normally is not possible to track all the particles in real time to avoid the storage requirement of the video, as the tracking needs to be very precise and thus has a high computing demand.
In this presentation, a combination of a CMOS imaging sensor with an FPGA is presented, which, in combination, allows for virtually unlimited long high-speed tracking of up to eight particles at up to 10 kHz. First, the sensor and the FPGA combinations are laid out. Secondly, the used particle tracking algorithm and its implementation is explained and benchmarked with a known state-of-the-art algorithm. Finally, this integrated sensor solution is mounted on a standard microscope and hour long tracking experiments on living 3T3 fibroblasts are carried out, studying the impact of blebbistatin on the mobility of polystyrene beads within the cell.
|