Collagen fiber centerline tracking in fibrotic tissue via deep neural networks with variational autoencoder-based synthetic training data generation.

The role of fibrillar collagen in the tissue microenvironment is critical in disease contexts ranging from cancers to chronic inflammations, as evidenced by many studies. Quantifying fibrillar collagen organization has become a powerful approach for characterizing the topology of collagen fibers and studying the role of collagen fibers in disease progression. We present a deep learning-based pipeline to quantify collagen fibers' topological properties in microscopy-based collagen images from pathological tissue samples. Our method leverages deep neural networks to extract collagen fiber centerlines and deep generative models to create synthetic training data, addressing the current shortage of large-scale annotations. As a part of this effort, we have created and annotated a collagen fiber centerline dataset, with the hope of facilitating further research in this field. Quantitative measurements such as fiber orientation, alignment, density, and length can be derived based on the centerline extraction results. Our pipeline comprises three stages. Initially, a variational autoencoder is trained to generate synthetic centerlines possessing controllable topological properties. Subsequently, a conditional generative adversarial network synthesizes realistic collagen fiber images from the synthetic centerlines, yielding a synthetic training set of image-centerline pairs. Finally, we train a collagen fiber centerline extraction network using both the original and synthetic data. Evaluation using collagen fiber images from pancreas, liver, and breast cancer samples collected via second-harmonic generation microscopy demonstrates our pipeline's superiority over several popular fiber centerline extraction tools. Incorporating synthetic data into training further enhances the network's generalizability. Our code is available at https://github.com/uw-loci/collagen-fiber-metrics.

Evaluation of the annual effective dose due to the external irradiation induced by using NORM added consumer products.

This study aims to evaluate the annual effective dose from a sleeping mattress containing naturally occurring radioactive material (NORM). In this study, the dose rate was measured using two different portable radiation detectors, namely the Geiger Müller (GM) tube and portable high-purity germanium (HPGe) detector; the annual effective dose was calculated using annualized usage of the products, and the equivalent does was evaluated via Monte Carlo (MC) simulation and using the model of the human body, which is known as a computational human phantom. The dose rate of the product, excluding background radiation at the shielded room, was measured as 0.22 and 0.13 µSv/h in the GM-tube and portable HPGe, respectively. Assuming that the sleeping mattress was used for an average sleeping of 8 h/day, the annual effective dose was calculated as 0.64 and 0.38 mSv/y using the GM-tube and portable HPGe detectors, respectively. Also, the annual effective dose calculated using MC simulation and radioactivity values from the nuclides analysis was 0.13 mSv/y. The annual effective dose calculated using the two different portable detectors and MC simulation is less than the annual effective dose limit for the general public, which is set at 1 mSv/y. This technique could be used not only for the safety regulation for products containing NORM but also for the accurate evaluation of the effective dose for radiation workers in the diverse radiation field.