Precalving behavior in dairy cattle with different calving times.

Predicting the calving time in dairy cattle can help in avoiding calving accidents and reducing burdens on animal caretakers. In this study, we analyzed the behavior of pregnant dairy cattle for 7 days prior to calving, to assess the feasibility of predicting the calving time. Eleven Holstein cows were divided into two groups based on their calving times, that is, in the morning (the Morning Parturition Group) or the evening (the Evening Parturition Group). Their behavior was recorded on video. An analysis was conducted of the daily occurrences of each type of behavior and the number of switches of behavior during the day and at night. A statistical analysis was conducted, using a two-way factorial analysis. The behavioral sequence was analyzed using an adjacency matrix. Hierarchical structure charts were created, using Interpretive Structural Modeling. The results suggest that feeding and exploratory behaviors are associated with the calving time period and thus can be useful when predicting that period. The hierarchical structure charts suggest that the Morning Parturition Group had no definite behavioral sequence pattern, unlike the Evening Parturition Group. The detection of an unstable behavioral sequence pattern might predict the calving time period.

Analysis of characteristics required for gait evaluation of patients with knee osteoarthritis using a wireless accelerometer.

BACKGROUND: Knee osteoarthritis (KOA) is associated with reduced quality of life due to knee pain and gait disturbance. However, the evaluation of KOA is mainly based on images and patient-reported outcome measures (PROMs), which are said to be insufficient for functional evaluation. Recently, gait analysis using an accelerometer has been used for functional evaluation of KOA patients. Nevertheless, evaluation of the entire body motion is insufficient. The aim of this study was to clarify the gait characteristics of KOA patients using the distribution of scalar products and the interval time of heel contact during spontaneous walking and to compare them with healthy subjects. METHODS: Participants wore a three-axis accelerometer sensor on the third lumbar vertebra and walked for 6 min on a flat path at a free walking speed. The sum of a composite vector (CV) scalar product and a histogram for distribution were used for body motion evaluation. The CV consisted of a synthesis of acceleration data from three axes. In addition to the summation of the CV, a histogram can be created to evaluate in detail the magnitude of the waves. The amount of variation was measured in the left-right and front-back directions. Variability was evaluated from the distribution of heel contact duration between both feet measured from the vertical acceleration. RESULTS: KOA patients showed a smaller sum of CV that converged to small acceleration in the distribution when compared with healthy subjects. In the KOA group, the amount of variation in the forward and backward directions was greater than that in the forward direction. The variability of heel-ground interval time was greater in the KOA group than in healthy subjects. CONCLUSION: KOA patients walked with less overall body movement, with limited movable range of the knee joint and pain-avoiding motion. The gait of the KOA group was considered unstable, with long time intervals between peaks. The increase in the amount of forward variation was thought to be due to the effect of trunk forward bending during walking. The clinical relevance of this study is that it was possible to evaluate KOA patients' gait quantitatively and qualitatively.

Dairy cattle behavior classifications based on decision tree learning using 3-axis neck-mounted accelerometers.

Demand has been increasing recently for an automated monitoring system of animal behavior as a tool for the management of livestock animals. This study investigated the association between the behavior of dairy cattle and the acceleration data collected using three-axis neck-mounted accelerometers, as well as the feasibility of improving the precision of behavior classifications through machine learning. In total 38 Holstein dairy cows were used, and kept in four different farms. A logger was mounted to each collar to obtain acceleration data for calculating the activity level and variations. At the same time the behavior of the cattle was observed visually. Characteristic acceleration waves were recorded for eating, rumination, and lying, respectively; and the activity level and variations were significantly different among these behaviors (p < 0.01). Decision tree learning was performed on the data set from Farm A and validated its precision; which proved to be 99.2% in cross-validation, and 100% in test data sets from Farms B to D. This study showed that highly precise classifications for eating, rumination, and lying is possible by using decision tree learning to calculate the activity level and variations of cattle based on the data obtained by three-axis accelerometers mounted to a collar.