In this work, a novel platform for surface-enhanced Raman spectroscopy (SERS)-based chemical sensors utilizing three-dimensional microporous graphene foam (GF) decorated with silver nanoparticles (AgNPs) is developed and applied for methylene blue (MB) detection. The results demonstrate that silver nanoparticles significantly enhance cascaded amplification of SERS effect on multilayer graphene foam (GF). The enhancement factor of AgNPs/GF sensor is found to be four orders of magnitude larger than that of AgNPs/Si substrate. In addition, the sensitivity of the sensor could be tuned by controlling the size of silver nanoparticles. The highest SERS enhancement factor of ∼5 × 10⁴ is achieved at the optimal nanoparticle size of 50 nm. Moreover, the sensor is capable of detecting MB over broad concentration ranges from 1 nM to 100 μM. Therefore, AgNPs/GF is a highly promising SERS substrate for detection of chemical substances with ultra-low concentrations.

Non-invasive and accurate method for continuous blood glucose monitoring, the self-testing of blood glucose is in quest for better diagnosis, control and the management of diabetes mellitus (DM). Therefore, this study reports a multiple photonic band near-infrared (mbNIR) sensor augmented with personalized medical features (PMF) in Shallow Dense Neural Networks (SDNN) for the precise, inexpensive and pain free blood glucose determination. Datasets collected from 401 blood samples were randomized and trained with ten-fold validation. Additionally, a cohort of 234 individuals not included in the model training set were investigated to evaluate the performance of the model. The model achieved the accuracy of 97.8% along with 96.0% precision, 94.8% sensitivity and 98.7% specificity for DM classification based on a diagnosis threshold of 126 mg/dL for diabetes in fasting blood glucose. For non-invasive real-time blood glucose monitoring, the model exhibited ± 15% error with 95% confidence interval and the detection limit of 60–400 mg/dL, as validated with the standard hexokinase enzymatic method for glucose estimation. In conclusion, this proposed mbNIR based SDNN model with PMF is highly accurate and computationally cheaper compared to similar previous works using complex neural network. Some groups proposed using complicated mixed types of sensors to improve noninvasive glucose prediction accuracy; however, the accuracy gain over the complexity and costs of the systems harvested is still in questioned (Geng et al. in Sci Rep 7:12650, 2017). None of previous works report on accuracy enhancement of NIR/NN using PMF. Therefore, the proposed SDNN over PMF/mbNIR is an extremely promising candidate for the non-invasive real-time blood glucose monitoring with less complexity and pain-free.

Background: The prevalence of peritoneal dialysis (PD) in Thailand is increasing rapidly in part because of Thailand’s Peritoneal Dialysis First policy. PD is a home-based renal replacement therapy in which patients with chronic kidney disease perform up to 4 exchanges of dialysate fluid per day in the peritoneal cavity. Overhydration is one of the most common complications in patients on PD and is associated with increased morbidity and mortality. To monitor hydration status, patients collect hydration metrics, including body weight, blood pressure, urine output, and ultrafiltration volume, from each dialysis cycle and enter this information into a PD logbook. This information is reviewed bimonthly at PD clinic appointments. The chronic kidney disease-PD (CKD-PD) app with near-field communication (NFC) and optical character recognition (OCR) was developed to automate hydration metric collection. The information was displayed in the app for self-monitoring and uploaded to a database for real-time monitoring by the PD clinic staff. Early detection and treatment of overhydration could potentially reduce the morbidity and mortality related to overhydration. Objective: This study aims to identify usability issues and technology adoption barriers for the CKD-PD app with NFC and OCR and a monitoring system and to use this information to make rapid cycle improvements. Methods: A multidisciplinary team of nephrologists, PD clinic nurses, computer programmers, and engineers trained and observed 2 groups of 5 participants in the use of the CKD-PD app with NFC and OCR and a monitoring system. The participants were observed using technology in their homes in 3 phases. The data collected included the Unified Theory of Acceptance and Use of Technology questionnaire, think-aloud observation, user ratings, completion of hydration metrics, and upload of hydration metrics to the central database. These results were used by the team between phases to improve the functionality and usefulness of the app. Results: The CKD-PD app with NFC and OCR and a monitoring system underwent 3 rapid improvement cycles. Issues were identified regarding the usability of the NFC and OCR data collection, app stability, user interface, hydration metric calculation, and display. NFC and OCR improved hydration metric capture; however, issues remained with their usability. App stability and user interface issues were corrected, and hydration metrics were successfully uploaded by the end of phase 3. Participants’ scores on technology adoption decreased but were still high, and there was enthusiasm for the self-monitoring and clinical communication features. Conclusions: Our rapid cycle process improvement methodology identified and resolved key barriers and usability issues for the CKD-PD app with NFC and OCR and a monitoring system. We believe that this methodology can be accomplished with limited training in data collection, statistical analysis, and funding.

The COVID-19 pandemic has caused significant changes in many aspects of daily life, including learning, working, and communicating. As countries aim to recover their economies, there is an increasing need for smart city solutions, such as crowd monitoring systems, to ensure public safety both during and after the pandemic. This paper presents the design and implementation of a real-time crowd monitoring system using existing public Wi-Fi infrastructure. The proposed system employs a three-tiered architecture, including the sensing domain for data acquisition, the communication domain for data transfer, and the computing domain for data processing, visualization, and analysis. Wi-Fi access points were used as sensors that continuously monitored the crowd and uploaded data to the server. To protect the privacy of the data, encryption algorithms were employed during data transmission. The system was implemented in the Sri Chiang Mai Smart City, where nine Wi-Fi access points were installed in nine different locations along the Mekong River. The system provides real-time crowd density visualizations. Historical data were also collected for the analysis and understanding of urban behaviors. A quantitative evaluation was not feasible due to the uncontrolled environment in public open spaces, but the system was visually evaluated in real-world conditions to assess crowd density, rather than represent the entire population. Overall, the study demonstrates the potential of leveraging existing public Wi-Fi infrastructure for crowd monitoring in uncontrolled, real-world environments. The monitoring system is readily accessible and does not require additional hardware investment or maintenance. The collected dataset is also available for download. In addition to COVID-19 pandemic management, this technology can also assist government policymakers in optimizing the use of public space and urban planning. Real-time crowd density data provided by the system can assist route planners or recommend points of interest, while information on the popularity of tourist destinations enables targeted marketing.

This study reports on the application of an extreme learning machine (ELM) in near-real-time kidney monitoring via urine neutrophil gelatinase-associated lipocalin (NGAL) detection with a 3D graphene electrode. This integration marks the first instance of combining a graphene-based electrode with machine learning to enhance the NGAL detection accuracy, building on our group's 2020 research. The methodology involves two key components: a graphene electrode functionalized with a lipocalin-2 antibody for NGAL detection and the ELM application for improved prediction accuracy by using urine analysis data. The results show a significant 15% increase in the area under the curve (AUC) for NGAL determination, with error reduction from ±6 to 0.54 ng/mL within a linear range of 2.7-140 ng/mL. The ELM also lowered the detection limit from 14.8 to 0.89 ng/mL and increased accuracy, precision, sensitivity, specificity, and F1 score for AKI prediction by 8.89, 30.69, 6.78, 9.94, and 19.07%, respectively. These findings underscore the efficacy of simple neural networks in enhancing graphene-based electrochemical sensors for AKI biomarkers. ELM was chosen for its optimal performance-resource balance, with a comparative analysis of ELM, support vector machines, multilayer perceptron, and random forest algorithms also included. This research suggests the potential for miniaturizing AI-enhanced sensors for practical applications.

The fish farming industry is advancing by adopting technologies designed to enhance efficiency, productivity, and sustainability. This study investigates integrating a Retrieval-Augmented Generation Large Language Model (RAG-LLM) with a Deep Q-Network (DQN) in autonomous aquaculture. It compares their performance to traditional expert-led methods and other AI-based systems. The developed autonomous system employs ensemble learning of RAG-LLM and DQN, incorporating IoT devices to thoroughly monitor feeding schedules, disease management, growth, and water quality parameters. This integration allows the system to generate optimal policies through majority voting, leveraging pre-trained LLM knowledge to improve initialization conditions and accelerate learning convergence. The hybrid approach of RAG-LLM and DQN demonstrates superior growth rates and rapid stabilization of automation policies. This highlights its potential to enable non-experts to manage fish farms and efficiently scale production for global food sustainability.