Repurposing deep learning models from computer vision to classify drug-resistant cells using autofluorescence lifetime imaging data

<p dir="ltr">An invasive breast ductal carcinoma cell line is used to explore whether metabolic profiling can be used to classify cell populations that are responsive (wild-type) or resistant to antibody-drug conjugate (ADC) treatment. The metabolic profile is determined by 2-photon autofluorescence lifetime imaging (FLIM) of the live cells. This multi-dimensional data contains both intensity and lifetime information corresponding to the location and interaction status of AD(P)H, a molecule that is a key component of cell metabolism, and its average fluorescence lifetime is a means to probe important indicators of cancer. </p><p dir="ltr"><br></p><p dir="ltr">Poster presented as part of the Crick BioImage Analysis Symposium 2025.</p><p dir="ltr">Permission has been given by authors to upload to Crick Figshare. Copyright remains with the original authors.</p>