KEYWORDS: Deep learning, Tissues, In vivo imaging, Endoscopy, Education and training, Diseases and disorders, Biopsy, Biological samples, Tissue optics, Neural networks
Conventional imaging techniques target this problem by using specific antibody markers. Although such markers allow decent specificity, they are often limited in the field of application, especially for in vivo use, which limits the potential for clinical translations. In contrast to that, label-free optical technologies, like multiphoton microscopy (MPM), can generate highly resolved 3D images from unstained samples, by exploiting natural optical contrast. Label-free MPM can show epithelial damage and immune infiltration in unstained colon samples. Here, we imaged a mixture of T cells and neutrophils with label-free MPM. In order to obtain ground-truth images, we simultaneously recorded images of a Cd4+ specific fluorescent marker for T cells. A deep neural network was then trained for the segmentation of T cells and neutrophils based on such label-free MPM images. Upon training, this model can then be used to detect both cell types without relying on specific fluorescent markers, that were used to obtain ground truth. In the future, the augmentation of label-free MPM by such computational specificity could have great potential for in vivo endomicroscopy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.