Computational imaging modalities for multi-focal whole-slide imaging systems
Leon Van Der Graaff (TU Delft - ImPhys/Computational Imaging)
Geert J.L.H. Van Leenders (Erasmus MC)
Fanny Boyaval (Leiden University Medical Center)
Sjoerd Stallinga (TU Delft - ImPhys/Imaging Physics)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Whole-slide imaging systems can generate full-color image data of tissue slides efficiently, which are needed for digital pathology applications. This paper focuses on a scanner architecture that is based on a multi-line image sensor that is tilted with respect to the optical axis, such that every line of the sensor scans the tissue slide at a different focus level. This scanner platform is designed for imaging with continuous autofocus and inherent color registration at a throughput of the order of 400 MPx/s. Here, single-scan multi-focal whole-slide imaging, enabled by this platform, is explored. In particular, two computational imaging modalities based on multi-focal image data are studied. First, 3D imaging of thick absorption stained slides (∼60 μm) is demonstrated in combination with deconvolution to ameliorate the inherently weak contrast in thick-Tissue imaging. Second, quantitative phase tomography is demonstrated on unstained tissue slides and on fluorescently stained slides, revealing morphological features com-plementary to features made visible with conventional absorption or fluorescence stains. For both computational approaches simplified algorithms are proposed, targeted for straightforward parallel processing implementation at ∼GPx=s throughputs.