Journal of Ambient Intelligence and Smart Environments

Hyperbilirubinemia or jaundice occurs in 60% of healthy babies and 80% of preterm infants because of an increase in unconjugated bilirubin in red blood cells. It is subjective to determine the severity of jaundice by visual assessment of the skin color of a newborn, and clinical judgement is dependent on the doctor’s knowledge. The paper explains the development of a non-invasive bilirubin detection technique called CliNicS, to check the bilirubin level of premature babies and report premature births and deaths to the health organization via an IOT network. CliNicS provides a noninvasive, transcutaneous bilirubin monitoring system using LED having a wavelength of 410 nm to 460 nm, and it also provides the treatment automatically by using LCT (LED Controlled Therapy) method. The level of bilirubin will be detected by using the photo detector, and the bilirubin measurement will be displayed on the LCD display. The bilirubin levels will be transmitted to doctors and health organizations via the IOT network. The proposed method helps to detect neonatal jaundice earlier, which reduces the risk of hyperbilirubinemia in newborns and makes it easier to measure total serum bilirubin levels than ever before.

In order to solve the problem of household garbage classification accurately and efficiently, convolutional neural network classifier is an effective method. In this study, a garbage classification device was designed, and the image dataset Wit-Garbage for garbage classification was constructed based on the device by collecting garbage images under different light intensity and weather environment. The performances of the five network models VGG16, ResNet50, DenseNet121, MobileNet V2, Inception V3 on this dataset were compared by transfer learning. Then, the lightweight convolutional neural network MobileNet V2 was optimized by fine-tuning the hyperparameters, such as the type of optimizer, learning rate, Dropout parameter and number of freezing layers, respectively, and the training accuracy and efficiency were discussed in detail. Finally, the optimized model MobileNet V2 was deployed to the self-made garbage classification device for verification. The results show that the MobileNet V2 network model is superior to other networks in terms of training accuracy and efficiency on the proposed dataset, when the image input size was 224 ∗ 224 pixels, the Adamax optimizer was adopted, the learning rate was 0.0001, the Dropout was less than 0.5, and the number of frozen layers is less than 30. The actual verification results show that the average accuracy of the optimized network model trained on the proposed dataset for MSW classification was up to 98.75%, and compared with the model before optimization, the average accuracy was improved by 2.83%, and the average detection time was reduced by 69%.

Smart devices, such as smart phones, voice assistants and social robots, provide users with a range of input modalities, e.g., speech, touch, gestures, and vision. In recent years, advancements in processing of these input channels enable more natural interaction (e.g., automated speech, face, and gesture recognition, dialog generation, emotion expression etc.) experiences for users. However, there are several important challenges that need to be addressed to create these user experiences. One challenge is that most smart devices do not have sufficient computing resources to execute the Artificial Intelligence (AI) techniques locally. Another challenge is that users expect responses in near real-time when they interact with these devices. Moreover, users also want to be able to seamlessly switch between devices and services any time and from anywhere and expect personalized and privacy-aware services. To address these challenges, we design and develop a cloud-based middleware (CMI) which helps to develop multi-modal interaction applications and easily integrate applications to AI services. In this middleware, services developed by different producers with different protocols and smart devices with different capabilities and protocols can be integrated easily. In CMI, applications stream data from devices to cloud services for processing and consume the results. It supports data streaming from multiple devices to multiple services (and vice versa). CMI provides an integration framework for decoupling the services and devices and enabling application developers to concentrate on “interaction” instead of AI techniques. We provide simple examples to illustrate the conceptual ideas incorporated in CMI.

Elderly people requiring care the entire day usually depend on the availability of their family members to give assistance. However, the family members might not provide appropriate help especially in an emergent situation. The application of Internet of Things (IoT) technology with a variety of interconnected devices provides the solution. We propose an IoT-based smart healthcare system comprising wearable devices, which integrates a variety of contact sensors with location-based mesh networks (LBMN) such as Wi-Fi and Bluetooth Low Energy (BLE) connections to continuously sense various parameters of aging people. The BLE-connected devices such as wearable sensors, fixed sensors, seat cushions, pedal mats, magnetic reed switches, and mobile devices are all involved in collecting, processing, and transmitting physiological data and their locations to the cloud. Through the utilization of convenient interfaces such as software applications on smartphones and web pages on computers, it provides real time monitoring of the elderly in terms of localization, activity pattern, and health status. Thus the system enables early detection of health risks to the elderly. We used Platform as a service (PaaS) to receive and store the health data generated from the interconnected devices and to perform analysis. The essential feature of this LBMN is to generate a complete 6W(Who, What,When,Where,Why and How)big data for policy, feed it to the PaaS analysis to easily and quickly obtain more accurate data, and then develop possible health strategy or preventive measures. The proposed healthcare system detected that, out of the 20 participants recruited, 2 persons (10%) were often restless. It was also able to detect abnormal daily activity patterns with more tag positioning and the historical data from the devices. More importantly, it can help to prevent potential physical and neuropsychiatric disorders based on the real-time monitoring information and analyzed historical data for the aging people.