Enhancing Live-Cell Segmentation in Ptychograpgy Imaging: A Multi-Stage Framework for Error Correction

<p dir="ltr">Poster presented as part of the Crick BioImage Analysis Symposium 2023.</p><p dir="ltr">In bioimage analysis, precise visualization of live cellular processes is reliant on accurate segmentations. Ptychography, a label-free technique, provides high-resolution images with minimal interference, yet accurate live-cell segmentation remains difficult due to noise and morphology variations. </p><p dir="ltr">Our research introduces a multi-stage framework using machine learning for error-correcting live-cell segmentation in ptychography imaging. By combining classification and semantic segmentation, our approach improves accuracy, offering deeper insights into live-cell behaviour. This poster outlines the innocative framework and its promising results in cell lineage tracking.</p><p dir="ltr"><br></p><p dir="ltr"><i>Permission has been given by authors to upload to Crick Figshare. Copyright remains with the original authors.</i></p>