The abstract is read at the publisher; the summary is JClub's.
Volumetric refractive index patterns from holotomography enable label-free classification of CD57-defined human memory CD8+ T-cell subsets, reaching 94.44% accuracy.
- Why it matters: Closely related T-cell subsets overlap in biophysical measurements, making label-free discrimination difficult for studying immune differentiation.
- What they did: Primary human T cells were imaged by holotomography and classified using a FishNet-inspired 3D convolutional neural network, with FACS defining CD4+, naive CD8+, and CD57- and CD57+ memory CD8+ populations.
- The result: A dedicated binary model distinguished CD57- versus CD57+ memory CD8+ cells with 94.44% accuracy, indicating intrinsic optical phenotypes can be extracted from RI tomograms.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.