ABSTRACT
The resolution and contrast of microscope imaging is often affected by aberrations introduced by imperfect optical systems and inhomogeneous refractive structures in specimens. Adaptive optics (AO) compensates these aberrations and restores diffraction limited performance. A wide range of AO solutions have been introduced, often tailored to a specific microscope type or application. Until now, a universal AO solution - one that can be readily transferred between microscope modalities - has not been deployed. We propose versatile and fast aberration correction using a physics-based machine learning assisted wavefront-sensorless AO control (MLAO) method. Unlike previous ML methods, we used a specially constructed neural network (NN) architecture, designed using physical understanding of the general microscope image formation, that was embedded in the control loop of different microscope systems. The approach means that not only is the resulting NN orders of magnitude simpler than previous NN methods, but the concept is translatable across microscope modalities. We demonstrated the method on a two-photon, a three-photon and a widefield three-dimensional (3D) structured illumination microscope. Results showed that the method outperformed commonly-used modal-based sensorless AO methods. We also showed that our ML-based method was robust in a range of challenging imaging conditions, such as 3D sample structures, specimen motion, low signal to noise ratio and activity-induced fluorescence fluctuations. Moreover, as the bespoke architecture encapsulated physical understanding of the imaging process, the internal NN configuration was no-longer a "black box", but provided physical insights on internal workings, which could influence future designs.
ABSTRACT
A major challenge in cell and developmental biology is the automated identification and quantitation of cells in complex multilayered tissues. We developed CytoCensus: an easily deployed implementation of supervised machine learning that extends convenient 2D 'point-and-click' user training to 3D detection of cells in challenging datasets with ill-defined cell boundaries. In tests on such datasets, CytoCensus outperforms other freely available image analysis software in accuracy and speed of cell detection. We used CytoCensus to count stem cells and their progeny, and to quantify individual cell divisions from time-lapse movies of explanted Drosophila larval brains, comparing wild-type and mutant phenotypes. We further illustrate the general utility and future potential of CytoCensus by analysing the 3D organisation of multiple cell classes in Zebrafish retinal organoids and cell distributions in mouse embryos. CytoCensus opens the possibility of straightforward and robust automated analysis of developmental phenotypes in complex tissues.
There are around 200 billion cells in the human brain that are generated by a small pool of rapidly dividing stem cells. For the brain to develop correctly, these stem cells must produce an appropriate number of each type of cell in the right place, at the right time. However, it remains unclear how individual stem cells in the brain know when and where to divide. To answer this question, Hailstone et al. studied the larvae of fruit flies, which use similar genes and mechanisms as humans to control brain development. This involved devising a new method for extracting the brains of developing fruit flies and keeping the intact tissue alive for up to 24 hours while continuously imaging individual cells in three dimensions. Manually tracking the division of each cell across multiple frames of a time-lapse is extremely time consuming. To tackle this problem, Hailstone et al. created a tool called CytoCensus, which uses machine learning to automatically identify stem cells from three-dimensional images and track their rate of division over time. Using the CytoCensus tool, Hailstone et al. identified a gene that controls the diverse rates at whichstem cells divide in the brain. Earlier this year some of the same researchers also published a study showing that this gene regulates a well-known cancer-related protein using an unconventional mechanism. CytoCensus was also able to detect cells in other developing tissues, including the embryos of mice. In the future, this tool could aid research into diseases that affect complex tissues, such as neurodegenerative disorders and cancer.
