Active focus stabilisation using astigmatism with universal objective lens compatibility and sub-10 nm precision

Session

Session 4 - New Technologies: Recent advances from Acquisition to Analysis

Authors

Amir Rahmani (1, 2), Tabitha Cox (1), Akhila Thamaravelil Abhimanue Achary (1), Aleks Ponjavic (1, 2, 3)

Affiliations

1. School of Physics and Astronomy, University of Leeds, Leeds, LS2 9JT
2. Bragg Centre for Materials Research, University of Leeds, Leeds, LS2 9JT
3. School of Nutrition and Food Science, University of Leeds, Leeds, LS2 9JT

Keywords

Focus stabilisation | Optical microscopy | Astigmatism | Autofocus

Abstract text

Super-resolution microscopy techniques can resolve biological samples with nanoscale resolution that surpasses the diffraction limit. This, however, requires focus stabilisation to correct for axial drift, which is particularly important for high-throughput automated imaging. Many solutions have been developed to tackle this problem. Marker-based methods are highly accurate, but the introduction of fiducials during sample preparation can give rise to additional challenges. With a beam-based approach, it is common to monitor the reflection of an infrared laser from the coverslip-sample interface. Nevertheless, there are limitations of current beam-based approaches that can be summarised as: (i) relatively low sampling rates; (ii): only being compatible with high numerical aperture (NA); (iii) electronic components requiring soldering and custom circuit boards and (iv): the need to put a focus stabilisation module close to the objective lens as the performance is limited by laser pointing stability. 

Here, we present a standalone, cost-effective, and fiducial-free focus stabilisation system that operates over a long axial range with nanoscale precision and can be implemented using off-the-shelf components. In our focus stabilisation system, an infrared laser beam is focused by the objective lens, back-reflected from the coverslip-sample interface, and then imaged onto a camera. Introducing a cylindrical lens in front of the camera creates an astigmatic point-spread-function. Real-time monitoring of variations in the shape of this astigmatic intensity profile as a response to focal drift allows for the transmission of a control signal to a piezo z-stage, consequently facilitating the stabilisation of the sample.  

We characterise the performance of our astigmatism-based drift correction system and find that it is much less sensitive to optical component stability, unlike TIR-based systems that need to be close to the coverslip. We achieve sub-10 nm axial drift estimation and correction. We then show the astigmatism leads to one of the advantages of our system – its ability to operate within a large axial range, extended over 20 μm by employing cylindrical lenses with different focal length. This trade-off between precision and axial range makes it possible to achieve 20 nm precision with low numerical aperture 10x objective lenses. We have implemented our solution on a Raspberry Pi platform that can perform stabilisation at 100 Hz, which is suitable for drift correction on most super-resolution methods. With compatibility across different objective lenses, we have created a straightforward optical system that can seamlessly integrate into most microscopy setups, offering advantages such as ease of implementation, universality, and robustness.