Infertility caused by ovarian or tubal problems can be treated using In Vitro Fertilization and Embryo Transfer (IVF-ET); however, this is not possible for women with uterine loss and malformations that require uterine reconstruction for the treatment of their infertility. In this study, we are the first to report the usefulness of decellularized matrices as a scaffold for uterine reconstruction. Uterine tissues were extracted from Sprague Dawley (SD) rats and decellularized using either sodium dodecyl sulfate (SDS) or high hydrostatic pressure (HHP) at optimized conditions. Histological staining and quantitative analysis showed that both SDS and HHP methods effectively removed cells from the tissues with, specifically, a significant reduction of DNA contents for HHP constructs. HHP constructs highly retained the collagen content, the main component of extracellular matrices in uterine tissue, compared to SDS constructs and had similar content levels of collagen to the native tissue. The mechanical strength of the HHP constructs was similar to that of the native tissue, while that of the SDS constructs was significantly elevated. Transmission electron microscopy (TEM) revealed no apparent denaturation of collagen fibers in the HHP constructs compared to the SDS constructs. Transplantation of the decellularized tissues into rat uteri revealed the successful regeneration of the uterine tissues with a 3-layer structure 30 days after the transplantation. Moreover, a lot of epithelial gland tissue and Ki67 positive cells were detected. Immunohistochemical analyses showed that the regenerated tissues have a normal response to ovarian hormone for pregnancy. The subsequent pregnancy test after 30 days transplantation revealed successful pregnancy for both the SDS and HHP groups. These findings indicate that the decellularized matrix from the uterine tissue can be a potential scaffold for uterine regeneration.

Excavation, which is one of the most frequently performed tasks during construction often poses danger to human operators. To reduce potential risks and address the problem of workforce shortage, automation of excavation is essential. Although previous studies have yielded promising results based on the use of reinforcement learning (RL) for automated excavation, the properties of excavation task in the context of RL have not been sufficiently investigated. In this study, we investigate Qt-Opt, which is a variant of Q-learning algorithms for continuous action space, for learning the excavation task using depth images. Inspired by virtual adversarial training in supervised learning, we propose a regularization method that uses virtual adversarial samples to reduce overestimation of Q-values in a Q-learning algorithm. Our results reveal that Qt-Opt is more sample-efficient than state-of-the-art actor-critic methods in our problem setting, and we verify that the proposed method further improves the sample efficiency of Qt-Opt. Our results demonstrate that multiple optimal actions often exist within the process of excavation and the choice of policy representation is crucial for satisfactory performance.

No abstract

In recent times, research on highly efficient construction has gained increasing popularity. During construction with hydraulic excavators, the sequence of excavation motions significantly affects the duration and precision of the work; therefore, it should be as efficient as possible. Because automation is still difficult, the ability improvement support system of nonskilled operators is required. However, currently, there exists no effective method for evaluating or planning the excavation motion sequence. This study proposes a method for evaluating excavation motion sequences based on a pattern recognition technique. First, the data array of hydraulic excavator motions is divided into unit excavation motions, characterized using a clustering method. Second, the patterns are divided into excavation phases with similar excavation motion patterns. Finally, the excavation motion sequence is evaluated by calculating the transition frequency between excavation phases. The result shows that the proposed approach can classify the skill level of the operator with accuracy of 0.8. Therefore, the frequency of excavation phase transitions can be used as a useful indicator for evaluating excavation motion sequences.

No abstract