Search

Abstract Urinary exfoliated tumor cells have emerged as promising biomarkers for predicting, diagnosing, and guiding therapy in bladder cancer. Several methodologies based on biological and physical differences between normal cells and malignant tumor cells have been developed over the past few years. However, these methods still did not have sufficient sensitivity or specificity. In this study, a remote analysis protocol was devised utilizing motion microscopy. This technique amplifies vibrations within a recorded video by re-rendering motions, thereby generating highly magnified visuals. This approach aims to detect dynamic motions that may not be perceptible to the human eye under normal observation. Remarkably, motion microscopy unveiled discernible fluctuations surrounding bladder malignant tumor cells, which we referred to herein as cellular trail. The cellular trails were predominantly evident at around 1 Hz in amplified video images, with a velocity of 22 μm/s. Moreover, cellular trails were observed regardless of whether they were in a non-Newtonian or Newtonian fluid environment. Significantly, this phenomenon was distinguishable even in urine samples. In conclusion, we suggest motion microscopy as an innovative approach for detecting urinary malignant tumors with potential clinical utility as a complementary tool to cytology.

Abstract It is necessary to verify whether nail fold capillaroscopy can be utilized for the early detection of breast cancer. To establish this technology, an animal model was developed, utilizing mice for nail fold observations. Nail fold capillaroscopy revealed a human-like anatomical pattern and facilitated the observation of cellular movement within blood vessels. Injection of MCF-7 or mammary fibroblasts in mice allowed the observation of cellular vibrations using motion microscopy from nail fold. We have named this technology ‘capillary cell motion microscopy.’ Intriguingly, we were able to identify distinct cellular vibrations in the MCF-7 group. Moreover, evaluating its effectiveness in mice with chemically induced cancer revealed higher sensitivity (81%-85%) compared to conventional methods (45%-68%). Capillary cell motion microscopy, operating at 0.5–1.5 Hz, provided clear distinction of tumor cells and demonstrated potential applicability in human subjects. While condition adjustments may be necessary, this method holds promise for noninvasive breast cancer detection through nail fold observations.

Abstract Owing to global warming and pollution concerns, reducing the environmental footprint of the textile and fashion industry has received considerable attention. Within this industry, the dyeing and finishing processes contribute significantly to greenhouse gas emissions and water pollution. This study introduces an innovative approach to address these challenges by leveraging digital twin technology to optimize the textile dyeing process. A smart analysis module was developed to continuously monitor and analyze the dyeing parameters in real time to implement control actions to automatically reduce the process duration. Integrated with this module, a digital twin of the dyeing machine enabled the real-time monitoring of energy consumption and process parameters. A case study comparing the traditional dyeing process with the optimized process was conducted. The results showed that dyeing time was reduced by ~ 17.5% without compromising dyeing quality. Energy consumption and greenhouse gas emissions were also reduced by ~ 12.1% when using the optimized process. This study offers a practical and sustainable option for textile dyeing, particularly for small and medium-sized enterprises.

Abstract Solute carrier family (SLC) transporters are expressed in the digestive system and play important roles in maintaining physiological functions in the body. In addition, SLC transporters act as oncoproteins or tumor-suppressor proteins during the development, progression, and metastasis of various digestive system cancers. SLC22A18, a member of the SLC22 gene family, is an orphan transporter with an unknown endogenous substrate. Previous study revealed that SLC22A18 is downregulated in colorectal cancer tissues and that it acts as a suppressor in colorectal cancer, although the effects of SLC22A18 variants on colon cancer cell proliferation, migration, and invasion are unknown. Therefore, in this study, we identified SLC22A18 variants found in multiple populations by searching public databases and determined the in vitro effects of these missense variations on transporter expression and cancer progression. Our results indicated that three missense SLC22A18 variants—p.Ala6Thr, p.Arg12Gln, and p.Arg86His—had significantly lower cell expression than the wild type, possibly owing to intracellular degradation. Furthermore, these three variants caused significantly higher proliferation, migration, and invasion of colon cancer cells than the wild type. Our findings suggest that missense variants of SLC22A18 can potentially serve as biomarkers or prognostic tools that enable clinicians to predict colorectal cancer progression.

Anomaly detection systems based on artificial intelligence (AI) have demonstrated high performance and efficiency in a wide range of applications such as power plants and smart factories. However, due to the inherent reliance of AI systems on the quality of training data, they still demonstrate poor performance in certain environments. Especially in hazardous facilities with constrained data collection, deploying these systems remains a challenge. In this paper, we propose Generative Anomaly Detection using Prototypical Networks (GAD-PN) designed to detect anomalies using only a limited number of normal samples. GAD-PN is a structure that integrates CycleGAN with Prototypical Networks (PNs), learning from metadata similar to the target environment. This approach enables the collection of data that are difficult to gather in real-world environments by using simulation or demonstration models, thus providing opportunities to learn a variety of environmental parameters under ideal and normal conditions. During the inference phase, PNs can classify normal and leak samples using only a small number of normal data from the target environment by prototypes that represent normal and abnormal features. We also complement the challenge of collecting anomaly data by generating anomaly data from normal data using CycleGAN trained on anomaly features. It can also be adapted to various environments that have similar anomalous scenarios, regardless of differences in environmental parameters. To validate the proposed structure, data were collected specifically targeting pipe leakage scenarios, which are significant problems in environments such as power plants. In addition, acoustic ultrasound signals were collected from the pipe nozzles in three different environments. As a result, the proposed model achieved a leak detection accuracy of over 90% in all environments, even with only a small number of normal data. This performance shows an average improvement of approximately 30% compared with traditional unsupervised learning models trained with a limited dataset.