Lameness is a serious problem in dairy farm management systems and has a significant impact on dairy product reduction. Image signal processing techniques have been developed to automatically detect lame cows in dairy farms. In this paper, we propose a method of artificial intelligence fusion in digital transformation techniques for detecting lameness scores in dairy cattle. Specifically, we investigate the use of Fourier Transformation and Moving Average Transformation in conjunction with Artificial Intelligence (AI) models to analyze image depth signals taken on individual cows as they walk from the milking station to resting areas. Our proposed method involves developing frequency signal variation measures of the collected image depth signals, using spectral analysis and Moving Average Transformation, and analyzing the frequency variation measures with AI models to detect cattle lameness scores. We present partial experimental results using self-collected real-life data, which showed that applying Fourier analysis to the backbone depth data for individual cows resulted in increased mean and standard deviations when the lame scores increased but decreased skewness and kurtosis. We also used Poincare graph representations to visualize and quantify the correlation between consecutive data points in a time series and found that wider intervals between peaks were associated with larger scores. These results demonstrate the potential of our proposed method for lameness detection in dairy cattle and highlight the importance of further research in this area.

抄録全体を表示

The introduction of home mechanical ventilation (HMV) or home oxygen therapy (HOT) for chronic obstructive pulmonary disease (COPD) has been shown to improve patient outcomes and shorten hospital stays. Despite this, COPD deaths continue to increase worldwide, and there is room for improvement in the treatment of COPD at home. Maintaining fraction of inspiratory oxygen (FiO₂) may increase the effectiveness of treatment. However, previous studies have shown when HOT was used in combination with HMV, FiO₂ decreased due to leakage from the gap between the patient’s face and non-invasive positive pressure ventilation (NPPV) mask. The latest HMV model, VOCSN (Ventec Life Systems, Inc.), is now being used in clinical practice as the newest ventilator model with HOT in HMV. In this study, experiments were conducted using the two types of ventilator circuits used in the VOCSN, but no statistically significant differences were found between them. We report on basic experiments conducted to investigate whether the VOCSN can maintain FiO₂ without being affected by leaks or ventilation volume compared to previous studies.

抄録全体を表示

To realize a super-smart society, it is necessary to aim for advanced integration of cyberspace and physical space (real society). Artificial Intelligence (AI) analysis of big data will bring effective information that meets the needs of individuals and companies to real society more quickly. On the other hand, to build a safe and secure society, it is important to develop AI methods that protect the privacy of big data in cyberspace. However, there is a little-known method that satisfies both data confidentiality and utilization of the learning method. Therefore, the authors proposed a learning method for secure distributed processing using decomposition data. This method has higher confidentiality and utilization than the conventional method, but the increase in computational complexity due to distributed processing is a problem. In the previous paper, the authors proposed the Back Propagation (BP) method to solve this problem. In this paper, we apply this method to the Neural Gas (NG) and k-means methods for secure distributed processing, which are unsupervised learning, and show its effectiveness.

抄録全体を表示