Ioana Slabu

Semantic segmentation is an important imaging analysis method enabling the identification of tissue structures. Histological image segmentation is particularly challenging, having large structural information while providing only limited training data. Additionally, labeling these structures to generate training data is time consuming. Here, we demonstrate the feasibility of a semantic segmentation using U-Net with a novel sparse labeling technique. The basic U-Net architecture was extended by attention gates, residual and recurrent links, and dropout regularization. To overcome the high class imbalance, which is intrinsic to histological data, under- and oversampling and data augmentation were used. In an ablation study, various architectures were evaluated, and the best performing model was identified. This model contains attention gates, residual links, and a dropout regularization of 0.125. The segmented images show accurate delineations of the vascular structures (with a precision of 0.9088 and an AUC-ROC score of 0.9717), and the segmentation algorithm is robust to images containing staining variations and damaged tissue. These results demonstrate the feasibility of sparse labeling in combination with the modified U-Net architecture.

The interest in mesenchymal stromal cells as a therapy option is increasing rapidly. To improve their implementation, location, and distribution, the properties of these must be investigated. Therefore, cells can be labeled with nanoparticles as a dual contrast agent for fluorescence and magnetic resonance imaging (MRI). In this study, a more efficient protocol for an easy synthesis of rose bengal–dextran-coated gadolinium oxide (Gd2O3-dex-RB) nanoparticles within only 4 h was established. Nanoparticles were characterized by zeta potential measurements, photometric measurements, fluorescence and transmission electron microscopy, and MRI. In vitro cell experiments with SK-MEL-28 and primary adipose-derived mesenchymal stromal cells (ASC), nanoparticle internalization, fluorescence and MRI properties, and cell proliferation were performed. The synthesis of Gd2O3-dex-RB nanoparticles was successful, and they were proven to show adequate signaling in fluorescence microscopy and MRI. Nanoparticles were internalized into SK-MEL-28 and ASC via endocytosis. Labeled cells showed sufficient fluorescence and MRI signal. Labeling concentrations of up to 4 mM and 8 mM for ASC and SK-MEL-28, respectively, did not interfere with cell viability and proliferation. Gd2O3-dex-RB nanoparticles are a feasible contrast agent to track cells via fluorescence microscopy and MRI. Fluorescence microscopy is a suitable method to track cells in in vitro experiments with smaller samples.