Deep Learning for Predicting Gamma-Ray Interaction Positions in LYSO Detector.

Positron Emission Tomography (PET) is among the most commonly used medical imaging modalities in clinical practice, especially for oncological applications. In contrast to conventional imaging modalities like X-ray Computed Tomography (CT) or Magnetic Resonance Imaging (MRI), PET retrieves in vivo information about biochemical processes rather than just anatomical structures. However, physical limitations and detector constraints lead to an order of magnitude lower spatial resolution in PET images. In recent years, the use of monolithic detector crystals has been investigated to overcome some of the factors limiting spatial resolution. The key to increasing PET systems' resolution is to estimate the gamma-ray interaction position in the detector as precisely as possible.In this work, we evaluate a Convolutional Neural Network (CNN) based reconstruction algorithm that predicts the gamma-ray interaction position using light patterns recorded with Silicon photomultipliers (SiPMs) on the crystal's surfaces. The algorithm is trained on data from a Monte Carlo Simulation (MCS) that models a gamma point source and a detector consisting of Lutetium-yttrium oxyorthosilicate (LYSO) crystals and SiPMs added to five surfaces. The final Mean Absolute Error (MAE) on the test dataset is 1.48 mm.

CRIPTO and its signaling partner GRP78 drive the metastatic phenotype in human osteotropic prostate cancer.

CRIPTO (CR-1, TDGF1) is a cell surface/secreted oncoprotein actively involved in development and cancer. Here, we report that high expression of CRIPTO correlates with poor survival in stratified risk groups of prostate cancer (PCa) patients. CRIPTO and its signaling partner glucose-regulated protein 78 (GRP78) are highly expressed in PCa metastases and display higher levels in the metastatic ALDHhigh sub-population of PC-3M-Pro4Luc2 PCa cells compared with non-metastatic ALDHlow. Coculture of the osteotropic PC-3M-Pro4Luc2 PCa cells with differentiated primary human osteoblasts induced CRIPTO and GRP78 expression in cancer cells and increases the size of the ALDHhigh sub-population. Additionally, CRIPTO or GRP78 knockdown decreases proliferation, migration, clonogenicity and the size of the metastasis-initiating ALDHhigh sub-population. CRIPTO knockdown reduces the invasion of PC-3M-Pro4Luc2 cells in zebrafish and inhibits bone metastasis in a preclinical mouse model. These results highlight a functional role for CRIPTO and GRP78 in PCa metastasis and suggest that targeting CRIPTO/GRP78 signaling may have significant therapeutic potential.

A zebrafish xenograft model for studying human cancer stem cells in distant metastasis and therapy response.

Lethal and incurable bone metastasis is one of the main causes of death in multiple types of cancer. A small subpopulation of cancer stem/progenitor-like cells (CSCs), also known as tumor-initiating cells from heterogenetic cancer is considered to mediate bone metastasis. Although over the past decades numerous studies have been performed in different types of cancer, it is still difficult to track small numbers of CSCs during the onset of metastasis. With use of noninvasive high-resolution imaging, transparent zebrafish embryos can be employed to dynamically visualize cancer progression and reciprocal interaction with stroma in a living organism. Recently we established a zebrafish CSC-xenograft model to visually and functionally analyze the role of CSCs and their interactions with the microenvironment at the onset of metastasis. Given the highly conserved human and zebrafish genome, transplanted human cancer cells are able to respond to zebrafish cytokines, modulate the zebrafish microenvironment, and take advantage of the zebrafish stroma during cancer progression. This chapter delineates the zebrafish CSC-xenograft model as a useful tool for both CSC biological study and anticancer drug screening.