Proactively detecting falls and preventing injuries are among the primary keys to a healthy life for the elderly. Near-fall remote monitoring in daily life could provide key information to prevent future falls and obtain quantitative rehabilitation status for patients with weak balance ability. In this study, we developed a deep learning-based novel classification algorithm to precisely categorize three classes (falls, near-falls, and activities of daily living (ADLs)) using a single inertial measurement unit (IMU) device attached to the waist. A total of 34 young participants were included in this study. An IMU containing accelerometer and gyroscope sensors was fabricated to acquire acceleration and angular velocity signals. A comprehensive experiment including thirty-six types of activities (10 types of falls, 10 types of near-falls, and 16 types of ADLs) was designed based on previous studies. A modified directed acyclic graph-convolution neural network (DAG-CNN) architecture with hyperparameter optimization was proposed to predict fall, near-fall, and ADLs. Prediction results of the modified DAG-CNN structure were found to be approximately 7% more accurate than the traditional CNN structure. For the case of near-falls, the modified DAG-CNN demonstrated excellent prediction performance with accuracy of over 98% by combining gyroscope and accelerometer features. Additionally, by combining acceleration and angular velocity the trained model showed better performance than each model of acceleration and angular velocity. It is believed that information to preemptively handle the risk of falls and quantitatively evaluate the rehabilitation status of the elderly with weak balance will be provided by monitoring near-falls.

Maintaining and monitoring the quality of eggs is a major concern during cold chain storage and transportation due to the variation of external environments, such as temperature or humidity. In this study, we proposed a deep learning-based Haugh unit (HU) prediction model which is a universal parameter to determine egg freshness using a non-destructively measured weight loss by transfer learning technique. The temperature and weight loss of eggs from a laboratory and real-time cold chain environment conditions are collected from ten different types of room temperature conditions. The data augmentation technique is applied to increase the number of the collected dataset. The convolutional neural network (CNN) and long short-term memory (LSTM) algorithm are stacked to make one deep learning model with hyperparameter optimization to increase HU value prediction performance. In addition, the general machine learning algorithms are applied to compare HU prediction results with the CNN-LSTM model. The source and target model for stacked CNN-LSTM used temperature and weight loss data, respectively. Predicting HU using only weight loss data, the target transfer learning CNN-LSTM showed RMSE value decreased from 6.62 to 2.02 compared to a random forest regressor, respectively. In addition, the MAE of HU prediction results for the target model decreased when the data augmentation technique was applied from 3.16 to 1.39. It is believed that monitoring egg freshness by predicting HU in a real-time cold chain environment can be implemented in real-life by using non-destructive weight loss parameters along with deep learning.

Long-lasting, drug-induced adaptations within the nucleus accumbens (NAc) have been proposed to contribute to drug-mediated addictive behaviors. Here we have used an optogenetic approach to examine the role of NAc medium spiny neurons (MSNs) expressing dopamine D2 receptors (D2R) in cocaine-induced behavioral sensitization. Adeno-associated viral vectors coding channelrhodopsin-2 (ChR2) were delivered into the NAc of D2R-Cre transgenic mice. This allowed us to selectively photostimulate D2R-MSNs in NAc. D2RMSNs form local inhibitory circuits, because photostimulation of D2R-MSN evoked inhibitory postsynaptic currents in neighboring MSNs. Photostimulation of NAc D2R-MSN in vivo affected neither the initiation nor the expression of cocaine-induced behavioral sensitization. However, photostimulation during the drug withdrawal period attenuated expression of cocaine-induced behavioral sensitization. These results show that D2R-MSNs of NAc play a key role in withdrawal-induced plasticity and may contribute to relapse after cessation of drug abuse.

Pre-impact fall detection can detect a fall before a body segment hits the ground. When it is integrated with a protective system, it can directly prevent an injury due to hitting the ground. An impact acceleration peak magnitude is one of key measurement factors that can affect the severity of an injury. It can be used as a design parameter for wearable protective devices to prevent injuries. In our study, a novel method is proposed to predict an impact acceleration magnitude after loss of balance using a single inertial measurement unit (IMU) sensor and a sequential-based deep learning model. Twenty-four healthy participants participated in this study for fall experiments. Each participant worn a single IMU sensor on the waist to collect tri-axial accelerometer and angular velocity data. A deep learning method, bi-directional long short-term memory (LSTM) regression, is applied to predict a fall’s impact acceleration magnitude prior to fall impact (a fall in five directions). To improve prediction performance, a data augmentation technique with increment of dataset is applied. Our proposed model showed a mean absolute percentage error (MAPE) of 6.69 ± 0.33% with r value of 0.93 when all three different types of data augmentation techniques are applied. Additionally, there was a significant reduction of MAPE by 45.2% when the number of training datasets was increased by 4-fold. These results show that impact acceleration magnitude can be used as an activation parameter for fall prevention such as in a wearable airbag system by optimizing deployment process to minimize fall injury in real time.

Due to an increase in interest towards functional and health-related foods, Panax ginseng sprout has been in the spotlight since it contains a significant amount of saponins which have anti-cancer, -stress, and -diabetic effects. To increase the amount of production as well as decrease the cultivation period, sprouted ginseng is being studied to ascertain its optimal cultivation environment in hydroponics. Although there are studies on functional components, there is a lack of research on early disease prediction along with productivity improvement. In this study, the ginseng sprouts were cultivated in four different hydroponic conditions: control treatment, hydrogen-mineral treatment, Bioblock treatment, and highly concentrated nitrogen treatment. Physical properties were measured, and environmental data were acquired using sensors. Using three algorithms (artificial neural networks, support vector machines, random forest) for germination and rottenness classification, and leaf number and length of stem prediction models, we propose a hierarchical machine learning model that predicts the growth outcome of ginseng sprouts after a week. Based on the results, a regression model predicts the number of leaves and stem length during the growth process. The results of the classifier models showed an F1-score of germination classification of about 99% every week. The rottenness classification model showed an increase from an average of 83.5% to 98.9%. Predicted leaf numbers for week 1 showed an average nRMSE value of 0.27, which decreased by about 33% by week 3. The results for predicting stem length showed a higher performance compared to the regression model for predicting leaf number. These results showed that the proposed hierarchical machine learning algorithm can predict germination and rottenness in ginseng sprout using physical properties.

Accurately measuring the lower extremities and L5/S1 moments is important since L5/S1 moments are the principal parameters that measure the risk of musculoskeletal diseases during lifting. In this study, protocol that predicts lower extremities and L5/S1 moments with an insole sensor was proposed to replace the prior methods that have spatial constraints. The protocol is hierarchically composed of a classification model and a regression model to predict joint moments. Additionally, a single LSTM model was developed to compare with proposed protocol. To optimize hyperparameters of the machine learning model and input feature, Bayesian optimization method was adopted. As a result, the proposed protocol showed a relative root mean square error (rRMSE) of 8.06~13.88% while the single LSTM showed 9.30~18.66% rRMSE. This protocol in this research is expected to be a starting point for developing a system for estimating the lower extremity and L5/S1 moment during lifting that can replace the complex prior method and adopted to workplace environments. This novel study has the potential to precisely design a feedback iterative control system of an exoskeleton for the appropriate generation of an actuator torque.