Search

The older population faces a high probability of experiencing age-related problems, such as osteoporosis, immobility, gait disturbances, stroke, Parkinson’s disease, and cognitive behavioral functional difficulties. Such problems negatively affect their lives. Thus, access to long-term care is a critical issue for older adults. In response to the aforementioned serious health issues, society must strive to provide a supportive and effective rehabilitation environment for older adults. This study designed an intelligent active and passive limb rehabilitation system to track and quantify the effectiveness of joint movements in patients automatically. The proposed method uses a camera and PoseNet to capture key feature information regarding human skeleton nodes and identify rehabilitation actions through joint movements. In addition, to solve the problem of joint occlusion during joint angle measurement, the designed system is equipped with a self-designed inertial measurement unit with GY-85 nine-axis sensors, which are mounted on the occluding part of the joints. A fuzzy inference system was developed to provide scores, suggestions, and encouragement for each rehabilitation session. This system also provides an interactive interface for users to monitor each rehabilitation session and examine whether rehabilitation is performed accurately.

Background/AimTo automatically detect and classify the early stages of retinopathy of prematurity (ROP) using a deep convolutional neural network (CNN).MethodsThis retrospective cross-sectional study was conducted in a referral medical centre in Taiwan. Only premature infants with no ROP, stage 1 ROP or stage 2 ROP were enrolled. Overall, 11 372 retinal fundus images were compiled and split into 10 235 images (90%) for training, 1137 (10%) for validation and 244 for testing. A deep CNN was implemented to classify images according to the ROP stage. Data were collected from December 17, 2013 to May 24, 2019 and analysed from December 2018 to January 2020. The metrics of sensitivity, specificity and area under the receiver operating characteristic curve were adopted to evaluate the performance of the algorithm relative to the reference standard diagnosis.ResultsThe model was trained using fivefold cross-validation, yielding an average accuracy of 99.93%±0.03 during training and 92.23%±1.39 during testing. The sensitivity and specificity scores of the model were 96.14%±0.87 and 95.95%±0.48, 91.82%±2.03 and 94.50%±0.71, and 89.81%±1.82 and 98.99%±0.40 when predicting no ROP versus ROP, stage 1 ROP versus no ROP and stage 2 ROP, and stage 2 ROP versus no ROP and stage 1 ROP, respectively.ConclusionsThe proposed system can accurately differentiate among ROP early stages and has the potential to help ophthalmologists classify ROP at an early stage.

Manual inspection and harvesting of ripening tomatoes is time consuming and labor intensive. Smart agriculture can emphasize the use of digital horticultural resources for farming and can increase farm sustainability; to that end, we proposed a fuzzy Mask R-CNN model to automatically identify the ripeness levels of cherry tomatoes. First, to annotate the images automatically, a fuzzy c-means model was used to maintain the spatial information of various foreground and background elements of the image. Then, a Hough transform method was applied to locate the specific geometric edge positions of the tomatoes. Each data point of the image space was annotated to a JavaScript Object Notation file. Second, annotated images were trained with Mask R-CNN to identify each tomato precisely. Finally, to prevent preharvest abscission of tomatoes, a hue-saturation-value color model and fuzzy inference rules were used to predict the ripeness of the tomatoes. A trigonometric function with Euclidian distance was calculated from the origin of calyx and stem to the bottom of the tomato to obtain the position of the pedicle head and dissect the fruit in a timely manner. For detection of 100 tomato images, Mask R-CNN achieved an accuracy of 98.00%. The ripeness classification of tomatoes achieved overall weighted precision and recall rates of 0.9614 and 0.9591, respectively. Thus, automatic tomato harvesting applications can empower farmers to make better decisions and enhance overall production efficiency and yield.

Diabetic retinopathy (DR) is a chronic and progressive sight-threatening complication of retinal microvasculature associated with diabetes mellitus. These morphological subtle variations in the retina can result in visual impairment or eventual vision loss if not treated in time. The early detection and diagnosis of DR is mandatory to safeguard the patient’s vision. To investigate and intervene at an earlier stage, we aim to systematically extract and analyze different DR factors, such as retinal condition of eyes, presence of microaneurysms (MAs), MAs with exudate, diameter of optical disk (OD), and Euclidian distance between macula and center of the OD. Considering the DR attributes, first we find out the important attributes of DR by applying fuzzy analytical network process to rank attributes from the most to the least related factors to DR. Then, a transformed fuzzy neural network is created to enhance the classification accuracy. Finally, we extract association rules among the selected attributes of DR to reveal their importance and degree of severity. Findings of this study reveal a new perspective on the treatment of DR at the early stage and prevention from any complications to improve the quality of life of all people.

Birdwatching is a common hobby but to identify their species requires the assistance of bird books. To provide birdwatchers a handy tool to admire the beauty of birds, we developed a deep learning platform to assist users in recognizing 27 species of birds endemic to Taiwan using a mobile app named the Internet of Birds (IoB). Bird images were learned by a convolutional neural network (CNN) to localize prominent features in the images. First, we established and generated a bounded region of interest to refine the shapes and colors of the object granularities and subsequently balanced the distribution of bird species. Then, a skip connection method was used to linearly combine the outputs of the previous and current layers to improve feature extraction. Finally, we applied the softmax function to obtain a probability distribution of bird features. The learned parameters of bird features were used to identify pictures uploaded by mobile users. The proposed CNN model with skip connections achieved higher accuracy of 99.00 % compared with the 93.98% from a CNN and 89.00% from the SVM for the training images. As for the test dataset, the average sensitivity, specificity, and accuracy were 93.79%, 96.11%, and 95.37%, respectively.

Various technological developments in home-care systems have allowed elderly people to live independently without compromising their safety. A pilot study employing deep learning algorithm was conducted to study the daily routines of elderly people. We monitored unsupervised, diverse daily activities of elderly people such as household chores, sleeping, cooking, cleaning, using the bathroom, watching television, and meditating. The activities were monitored to track human-environment interactions by using motion sensors, actuators, and surveillance systems that were mounted inside living rooms, bedrooms, and kitchens and on bathroom doorways to detect safety hazards in the environment for elderly people. Such collected data were used in deep belief networks to ascertain and identify activities that are related to various health and self-care problems. Simulation results show that the proposed system outperforms the support vector machines in terms of F1 score and accuracy in identifying daily activities.

The ever-escalating rise in numbers of the aging population has preempted a revolutionary change in the healthcare sector and serves as a major counterpoint to modern life in the 21st century. Increasing demand being placed on the health sector is almost certainly an inevitable process. However, providing appropriate healthcare services is requisite for senior citizens who suffer from various health issues and conditions. To minimize these health risks, we derived an intuitive technique for determining the incongruity of health symptoms by using a symptom checker, which is embedded into a versatile mobile app named Help-to-You (H2U). The designed app helps the users and carers to determine and identify conceivable reasons for elderly ailments and to assist users in deciding when to counsel a health practitioner. The intention of this empirical study was to further analyze and foresee certain variations of infections based on the symptoms accounted for by the patient. The recommended solution consolidated conceptual design with multi-criteria decision analysis (MCDA) technique and an analytic hierarchy process (AHP) with fuzzy weights to deal with the uncertainty of imprecision and ambiguity resulting from various disease factors. Experimental results verified the effectiveness of the proposed model, subsequently providing a variety of life assistance services.

To elevate the imperativeness and reinforce the fitness of elders, assisting a home care system can be an admittance that provides comprehensive nursing and monitoring them in the regular interim. In order to provide an interactive service management platform to the elders a smart environment of numerous sensors are clubbed together to establish an intuitive platform that can control the home appliances and gadgets within the living space of elders. The proposed system used voice and gesture to control the home appliances like turning on/off the light, closing/opening of curtains, TV, and fan or AC within the living spaces. The system also supports the real-time activity and monitor the healthcare system for the elderly citizens like heart rate and body temperature. In the case of emergency, for instance, anomalous behaviors like heart stroke occur, the proposed system set-up triggers an alarm and the emergency bulb will be on to alert to their kin. This smart environment can set the temperature and help control the living parameters of light based on the users' comfort and their health conditions. The whole design is to provide modest support systems for the elder to live healthily and safety in the independent living of elders.

The development on newer healthcare services for the elderly citizens has become an immediate necessity today. There have been distinctive health challenges focused in the society through technical innovations. Most of the elderly people today experiences loneliness and psychological depressions, either as a result of living alone/ abandonment or due to reduced connection with their children and relatives. To enhance the quality of services in the elderly healthcare system we have developed a ubiquitous intuitive IoT-based (Internet of Things) Help to You (H2U) healthcare system to integrate various technologies of wearable devices, biosensors and wireless sensor networks in order to provide an intensive service management platform. This method would support the real time activity and monitor the healthcare system for the elderly citizens. In this method the information collected by various wearable devices in real time are stored in the central database which thereby connects people, doctors, and practitioner at the time of an emergency for the right information. This way the system could increase accessibility, efficiency, and also lower the health expenses to improve the comfort and safety as well as management of daily routines of an elderly life.