Volume 9, Issue 4

Aquaculture is one of the most important industries worldwide, and most marine products are produced by aquaculture. On the other hand, the aquaculture farmers are faced on the challenge of damage to marine products due to abnormal seawater temperatures. Research on seawater temperature prediction have been conducted, but many of them require a large amount of training data. Collecting seawater temperature data is not easy, and it takes an enormous time to introduce in new aquaculture farms. Therefore, the purpose of this study is to predict seawater temperature even with a small amount of training data for about one year. In this paper, we propose a seawater temperature prediction model using transition learning. The proposed model also attempts to improve the prediction accuracy by considering the difference in water depth between observation points. The results of the evaluation experiment showed that the prediction accuracy can be improved by transfer learning when learning with a small amount of data. In addition, we also confirmed that adding water depth values to the input layer may not lead to improved prediction accuracy for transfer learning.

This paper presents a method for automatic 3D segmentation of brain tumors in MRI using optical scanning holography. Automatic segmentation of tumors from 2D slices (coronal, sagittal and axial) enables efficient 3D reconstruction of the region of interest, eliminating the human errors of manual methods. The method uses enhanced optical scanning holography with a cylindrical lens, scanning line by line, and displays MRI images via a spatial light modulator. The outgoing phase component of the scanned data, collected digitally, reliably indicates the position of the tumor.The tumor position is fed into an active contour model (ACM), which speeds up segmentation of the seeding region. The tumor is then reconstructed in 3D from the segmented regions in each slice, enabling tumor volume to be calculated and cancer progression to be estimated. Experiments carried out on patient MRI datasets show satisfactory results. The proposed approach can be integrated into a computer-aided diagnosis (CAD) system, helping doctors to localize the tumor, estimate its volume and provide 3D information to improve treatment techniques such as radiosurgery, stereotactic surgery or chemotherapy administration. In short, this method offers a precise and reliable solution for the segmentation and 3D reconstruction of brain tumors, facilitating diagnosis and treatment.

RetePAIoT of Emilia-Romagna region is an IoT Public Network, financed by Emilia-Romagna Region and developed by Lepida Scpa, where citizens, private companies and Public Administrations can integrate free of charge their own sensors of any type and anywhere in the region. The main objective of the project is to provide a facility to implement the IoT paradigm, receiving data potentially from thousands of different sensors from the territory and to make them available to their owners and, in aggregate or anonymous form, to Public Administrations for their institutional purposes. In this context the interpretation of payloads sent by sensors, i.e. the extraction of the values measured by sensors, as well sharing them with all authorized subjects. are fundamental aspects that present a significant complexity due to the variety and unplannable context of the project. This paper illustrates the model and the architecture of a solution for the automatic extraction of values potentially coming from thousands of different sensors, which therefore requires a high level of flexibility, robustness and scalability as well as different methods for sharing them with third parties, depending on purposes and technical level required.

In healthcare, accurate communication is critical, and medical coding, especially coding using the ICD (International Classification of Diseases) standards, plays a vital role in achieving this accuracy. Traditionally, ICD coding has been a time-consuming manual process performed by trained professionals, involving the assignment of codes to patient records, such as doctor’s notes. In this paper, we present an automated ICD coding approach using deep learning models and demonstrate the feasibility and effectiveness of the approach across subsets of ICD codes. The proposed method employs a fine-grained approach that individually predicts the appropriate medical code for each diagnosis. In order to utilize sufficient evidence to enhance the classification capabilities of our deep leaning models, we integrate GPT-4 to extract semantically related sentences for each diagnosis from doctor’s notes. Furthermore, we introduce a hierarchical classifier to handle the large label space and complex classification inherent in the ICD coding task. This hierarchical approach decomposes the ICD coding task into smaller, more manageable subclassification tasks, thereby improving tractability and addressing the challenges posed by the high number of unique labels associated with ICD coding.

This study addresses the need for improved prognostics in switch-mode power supplies (SMPS) that incorporate electromagnetic interference (EMI) filters, with a focus on aluminum electrolytic capacitors, which are critical for the reliability of these systems. The primary aim is to develop a robust model-based approach that can accurately predict the degradation and operational lifetime of these capacitors under varying environmental conditions. To achieve this, the research employs a generalized state space averaging technique to simulate a population of impending degradation trajectories for the capacitors. Environmental and degradation effects are modeled comprehensively. Frequency-based test features are derived from the gain, control, and impedance transfer functions of the filter and SMPS. These features are fitted with b-spline functionals for resampling and subsequently analyzed using functional principal component analysis to project the data onto the principal modes of variation. The extracted features serve as inputs to a fuzzy multi-task functional fusion predictor, which estimates the state of health at critical frequencies. The effectiveness of this model-based approach is validated through extensive experimentation, demonstrating its potential to significantly enhance the predictive maintenance strategies for SMPS with EMI filters.

This paper presents a system that combines speech and pointing gestures along with four distinct hand gestures to precisely identify both the object of interest and parameters for robotic tasks. We utilized skeleton landmarks to detect pointing gestures and determine their direction, while a pre-trained model, trained on 21 hand landmarks from 2D images, was employed to interpret hand gestures. Furthermore, a dedicated model was trained to extract task information from verbal instructions. The framework integrates task parameters derived from verbal instructions with inferred gestures to detect and identify objects of interest (OOI) in the scene, essential for creating accurate final task configurations.

In this paper, we will discuss two algorithms to solve the so-called package design problem, by which a set of queries (referred to as a query log) is represented by a collection of bit strings with each indicating the favourite activities or items of customers. For such a query log, we are required to design a package of activities (or items) so that as many customers as possible can be satisfied. It is a typical problem of data mining. For this problem, the existing algorithm requires at least O(n2m) time, where m is the number of activities (or items) and n is the number of queries. We try to improve this time complexity. The main idea of our first algorithm is to use a new tree search strategy to explore the query log. Its average time complexity is bounded by O(nm2 + m2m/2). By our second algorithm, all query bit strings are organized into a graph, called a trie-like graph. Searching such a graph bottom-up, we can find a most popular package in O(n2m3(log2 nm)log2 nm) time. Both of them work much better than any existing strategy for this problem.

The Mind Monitoring System (MIMOSYS) is a novel voice analysis technique for mental health assessment that has been validated in some languages; however, no research has been conducted on the Vietnamese yet. This study aimed to examine the ability of the Vitality score extracted from the MIMOSYS system to assess depression status based on the standards of the Patient Health Questionnaire-9 items (PHQ-9) and Beck’s Depression Inventory (BDI) questionnaire using the Vietnamese language. In this cross-sectional study conducted between August 1, 2022 and September 4, 2022, students and staff from a university, and patients from a hospital were recruited. Participants were asked to complete the self-administered depression questionnaires (PHQ-9 and BDI), and their voice data was collected by reading designated sentences. MIMOSYS extracted the Vitality score from the participant’s voice data. One hundred and twenty-two participants with a mean age of 31 years were included in the study; 72.4% of them were female. After adjusting for age and sex, negative correlations between the Vitality score and psychological test scores were found. For discriminating individuals with a high risk of depression, using the BDI score as a standard, the area under the curve of the Vitality score was 0.72. Sensitivity, specificity, and accuracy evaluations also reported a moderate discrimination ability of the Vitality score on the risk of depression by the BDI. In conclusion, voice analysis can be a viable technique for depression assessment in Vietnamese; however, further investigations are necessary to confirm our findings.

Gas leakage caused by various causes poses significant risks to public safety. To address this problem, an intelligent model is proposed for the accurate monitoring of Liquefied Petroleum Gas (LPG) distribution based on the integration of Internet of Things (IoT) and Business Intelligence (BI) technologies. Through the use of sensors and actuators, it seeks to mitigate risks and prevent accidents by enabling automated control of devices and infrastructures. The PRISMA methodology was used to perform a systematic review and obtain general characteristics of the components. Then, the proposed model was evaluated according to Y. 4908 which addresses IoT network interoperability, usability and security, the evaluation with 30 IT professionals who examined the BI model. The results obtained by the professionals were encouraging and favorable. The proposal, which enables remote LPG monitoring, establishes service through a website, mobile app or SMS when it detects fluctuations in humidity, temperature and gas indicators, shuts off the flow of LPG and notifies immediately. The research led to the development of a model that combines an IoT component with a four-tier BI, demonstrating its effectiveness and acceptance in the professional arena. At the overall medium level, 49% strongly agree, 38% agree, 12% neither agree and 1% disagree. It is concluded that the model has an overall average level of approval of 87%.

SME big data maturity models will be reviewed in this study to identify systematic publications related to the subject. For SMEs to remain competitive, digitalization is essential. Due to limited resources, SMEs need to be more proactive in digitalization. Still, the benefits, such as operational efficiency, cost reduction, quality improvement, and innovative culture, make digitalization attractive and valuable to customers. In recent years, there has been an increase in the use of big data techniques in operations. The paper discusses big data applications in SMEs through the lens of a big data maturity model. This paper met two objectives. First, this paper summarizes the most commonly used maturity models in the existing literature. Second, existing Big Data maturity models have limitations. Moreover, this paper outlines key considerations for selecting a Big Data maturity model to support data-driven decisions. Based on the Big Data maturity dimensions, further work aims to develop a new Big Data maturity model.

This research analyses the challenges faced by ASEAN countries in managing its energy efficiencies and resources due to rapid economic growth, increasing energy demand, and diverse energy infrastructures across member states. This paper explores the energy management policies and strategies within the ASEAN region, focusing on the integration of energy efficiency measures, renewable energy initiatives, and cross-border energy trade. This paper analyse the region’s progress towards its sustainable energy goals, the role of policy frameworks, and the impact of regional collaboration. Key challenges such as energy security, affordability, and environmental sustainability are examined, alongside opportunities for innovation in energy technologies and policy reforms. The findings highlight the importance of a cohesive energy management strategy that balances the diverse needs of ASEAN member states while advancing the region’s transition towards a low-carbon future. This paper provides policy recommendations aimed at enhancing ASEAN’s energy resilience and supporting its sustainable development goals.

Augmented reality (AR) technology has been increasingly used in developing teaching materials with the aim of sparking more interest in technology (T) and engineering (E) among students in STEM education. In the proposed system, AR is integrated with an educational robot controlled by a KidBright microcontroller board, developed by the Educational Technology research team (EDT) at the National Electronics and Computer Technology Center in Thailand. Moreover, the KidBright program has been implemented over 2,200 Thai schools. To maximize the benefits of the KidBright program, the Assistive System for Collaborative Assembly Task using Augmented Reality (ASCAT-AR) was created with the objective of enabling students to learn and collaborate in assembling robots. Students will work in pairs to assemble robots using the system and learn about mechanics, sensors, and 3D-printed parts. The students were divided into two groups: Group A read the manual and assembled the robot independently, while Group B used the ASCAT-AR system. In addition, AR applications offer smooth graphic rendering at 44-60 frames per second. Evaluation result showed that Group B students had a higher average success rate than average success rate of Group A students. The results showed that users of the ASCAT-AR system were more motivated in learning and obtained more knowledge about robot technology and programming.