Using wavelet transform and hybrid CNN - LSTM models on VOC & ultrasound IoT sensor data for non-visual maize disease detection.

Early detection of plant diseases is crucial for safeguarding crop yield, especially in regions vulnerable to food insecurity, such as Sub-Saharan Africa. One of the significant contributors to maize crop yield loss is the Northern Leaf Blight (NLB), which traditionally takes 14-21 days to visually manifest on maize. This study introduces a novel approach for detecting NLB as early as 4-5 days using Internet of Things (IoT) sensors, which can identify the disease before any visual symptoms appear. Utilizing Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM) models, nonvisual measurements of Total Volatile Organic Compounds (VOCs) and ultrasound emissions from maize plants were captured and analyzed. A controlled experiment was conducted on four maize varieties, and the data obtained were used to develop and validate a hybrid CNN-LSTM model for VOC classification and an LSTM model for ultrasound anomaly detection. The hybrid CNN-LSTM model, enhanced with wavelet data preprocessing, achieved an F1 score of 0.96 and an Area under the ROC Curve (AUC) of 1.00. In contrast, the LSTM model exhibited an impressive 99.98% accuracy in identifying anomalies in ultrasound emissions. Our findings underscore the potential of IoT sensors in early disease detection, paving the way for innovative disease prevention strategies in agriculture. Future work will focus on optimizing the models for IoT device deployment, incorporating chatbot technology, and more sensor data will be incorporated for improved accuracy and evaluation of the models in a field environment.

Prototype of Monitoring Transportation Pollution Spikes through the Internet of Things Edge Networks.

Air pollution is a critical problem in densely populated urban areas, with traffic significantly contributing. To mitigate the adverse effects of air pollution on public health and the environment, there is a growing need for the real-time monitoring and detection of pollution spikes in transportation. This paper presents a novel approach to using Internet of Things (IoT) edge networks for the real-time detection of air pollution peaks in transportation, specifically designed for innovative city applications. The proposed system uses IoT sensors in buses, cabs, and private cars. These sensors are equipped with air quality monitoring capabilities, including the measurement of pollutants such as particulate matter (PM2.5 and PM10), nitrogen dioxide (NO2), ozone (O3), sulfur dioxide (SO2), and carbon dioxide (CO2). The sensors continuously collect air quality data and transmit them to edge devices within the transportation infrastructure. The data collected by these sensors are analyzed, and alerts are generated when pollution levels exceed predefined thresholds. By deploying this system within IoT edge networks, transportation authorities can promptly respond to pollution spikes, improving air quality, public health, and environmental sustainability. This paper details the sensor technology, data analysis methods, and the practical implementation of this innovative system, shedding light on its potential for addressing the pressing issue of transportation-related pollution. The proposed IoT edge network for real-time air pollution spike detection in transportation offers significant advantages, including low-latency data processing, scalability, and cost-effectiveness. By leveraging the power of edge computing and IoT technologies, smart cities can proactively monitor and manage air pollution, leading to healthier and more sustainable urban environments.

A User-Centered design and usability testing of a web-based medication reconciliation application integrated in an eHealth network.

BACKGROUND: Medication discrepancies, which are a threat to patient safety, can be reduced by medication reconciliation (MedRec). MedRec is a complex process that can be supported by the use of information technology and patient engagement. Therefore, the SEAMPAT project aims to develop a MedRec IT platform based on two applications. The application for the professionals is called: the "MedRec app". OBJECTIVE: In the present study, we aimed to describe the development and usability testing of the MedRec app, reporting results of a three iterations user-centered usability evaluation. METHODS: We used a three phase iterative user-centered study spread over 16 months. At each phase, the usability evaluation included several methods (observations, questionnaires, and follow-up discussions with participants) to collect quantitative and qualitative data in order to improve the current prototype and evolve to the next prototype. RESULTS: In total, 48 healthcare professionals (25 general practitioners and 23 hospital clinicians) participated to the MedRec app evaluation. There were 14, 32 and 5 participants for phases 1, 2 and 3 respectively. At each phase, many design modifications were done to strengthen usability. Concerning usability, participants considered the prototypes as an acceptable interface with a median System Usability Score of 73 at phase 2 and 75 at phase 3. Participants emphasized the need for improvements concerning workflow integration, usefulness and interoperability. CONCLUSION: The MedRec app was perceived as being useful, usable and satisfying. However, further improvements are required in several usability aspects. Our study demonstrates the importance of conducting usability assessments before investing time and resources in a large study evaluating the effect of an eMedRec approach on clinical outcomes. Our findings may also increase the chances of acceptability and sustained use over time by clinicians.

A web application to involve patients in the medication reconciliation process: a user-centered usability and usefulness study.

Objective: Medication reconciliation (MedRec) can improve patient safety by resolving medication discrepancies. Because information technology (IT) and patient engagement are promising approaches to optimizing MedRec, the SEAMPAT project aims to develop a MedRec IT platform based on two applications: the "patient app" and the "MedRec app." This study evaluates three dimensions of the usability (efficiency, satisfaction, and effectiveness) and usefulness of the patient app. Methods: We performed a four-month user-centered observational study. Quantitative and qualitative data were collected. Participants completed the system usability scale (SUS) questionnaire and a second questionnaire on usefulness. Effectiveness was assessed by measuring the completeness of the medication list generated by the patient application and its correctness (ie medication discrepancies between the patient list and the best possible medication history). Qualitative data were collected from semi-structured interviews, observations and comments, and questions raised by patients. Results: Forty-two patients completed the study. Sixty-nine percent of patients considered the patient app to be acceptable (SUS Score ≥ 70) and usefulness was high. The medication list was complete for a quarter of the patients (7/28) and there was a discrepancy for 21.7% of medications (21/97). The qualitative data enabled the identification of several barriers (related to functional and non-functional aspects) to the optimization of usability and usefulness. Conclusions: Our findings highlight the importance and value of user-centered usability testing of a patient application implemented in "real-world" conditions. To achieve adoption and sustained use by patients, the app should meet patients' needs while also efficiently improving the quality of MedRec.