Background: In the modern world, mobile apps are essential for human advancement, and pandemic control is no exception. The use of mobile apps and technology for the detection and diagnosis of COVID-19 has been the subject of numerous investigations, although no thorough analysis of COVID-19 pandemic prevention has been conducted using mobile apps, creating a gap. Objective: With the intention of helping software companies and clinical researchers, this study provides comprehensive information regarding the different fields in which mobile apps were used to diagnose COVID-19 during the pandemic. Methods: In this systematic review, 535 studies were found after searching 5 major research databases (ScienceDirect, Scopus, PubMed, Web of Science, and IEEE). Of these, only 42 (7.9%) studies concerned with diagnosing and detecting COVID-19 were chosen after applying inclusion and exclusion criteria using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) protocol. Results: Mobile apps were categorized into 6 areas based on the content of these 42 studies: contact tracing, data gathering, data visualization, artificial intelligence (AI)–based diagnosis, rule- and guideline-based diagnosis, and data transformation. Patients with COVID-19 were identified via mobile apps using a variety of clinical, geographic, demographic, radiological, serological, and laboratory data. Most studies concentrated on using AI methods to identify people who might have COVID-19. Additionally, symptoms, cough sounds, and radiological images were used more frequently compared to other data types. Deep learning techniques, such as convolutional neural networks, performed comparatively better in the processing of health care data than other types of AI techniques, which improved the diagnosis of COVID-19. Conclusions: Mobile apps could soon play a significant role as a powerful tool for data collection, epidemic health data analysis, and the early identification of suspected cases. These technologies can work with the internet of things, cloud storage, 5th-generation technology, and cloud computing. Processing pipelines can be moved to mobile device processing cores using new deep learning methods, such as lightweight neural networks. In the event of future pandemics, mobile apps will play a critical role in rapid diagnosis using various image data and clinical symptoms. Consequently, the rapid diagnosis of these diseases can improve the management of their effects and obtain excellent results in treating patients.

This study presents a novel workflow for automated Digital Terrain Model (DTM) extraction from Airborne LiDAR point clouds based on a convolutional neural network (CNN), considering a transfer learning approach. The workflow consists of three parts: feature image generation, transfer learning using ResNet, and interpolation. First, each point is transformed into a featured image based on its elevation differences with neighboring points. Then, the feature images are classified into ground and non-ground using ImageNet pretrained ResNet models. The ground points are extracted by remapping each feature image to its corresponding points. Last, the extracted ground points are interpolated to generate a continuous elevation surface. We compared the proposed workflow with two traditional filters, namely the Progressive Morphological Filter (PMF) and the Progressive Triangulated Irregular Network Densification (PTD). Our results show that the proposed workflow establishes an advantageous DTM extraction accuracy with yields of only 0.52%, 4.84%, and 2.43% for Type I, Type II, and the total error, respectively. In comparison, Type I, Type II, and the total error for PMF are 7.82%, 11.60%, and 9.48% and for PTD 1.55%, 5.37%, and 3.22%, respectively. The root means square error (RMSE) for the 1 m resolution interpolated DTM is only 7.3 cm. Moreover, we conducted a qualitative analysis to investigate the reliability and limitations of the proposed workflow.

Incorporating social vulnerability (SoV) into vulnerability assessments for protected areas provides critical insights for enhancing the resilience and adaptability of these regions to environmental hazards. This study aimed to develop and apply an integrated social-ecological vulnerability assessment framework for a forest-protected area, focusing on scales relevant to management decisions. We collected data from 70 villages within the protected area through structured surveys, interviews, and reviews of secondary socio-economic and environmental datasets. A total of 24 socio-economic and environmental indicators were selected, based on expert consultation and literature review, to quantify the indices of exposure, sensitivity, adaptive capacity, and SoV. These indicators were weighted using the Fuzzy Analytical Hierarchy Process (FAHP) to ensure objective and expert-driven prioritization. Subsequently, the SoV index, derived from socio-economic metrics, was incorporated as a key variable in the exposure assessment. This integration involved 18 additional ecological indicators measured across a spatial grid, enabling a fine-scale analysis of the protected area's exposure, sensitivity, adaptive capacity, and overall vulnerability. Integrating social and ecological dimensions allowed for a spatially explicit evaluation of vulnerability at the grid-cell level, providing granular insights for localized management interventions. The results revealed that SoV significantly influenced the levels of exposure and overall vulnerability across the study area. High SoV values corresponded to areas with elevated overall vulnerability, emphasizing the interplay between social conditions and ecological risks. To enhance resilience in the medium and long term, the study recommends shifting investments toward active community engagement in conservation planning, increasing environmental awareness through education programs, and improving the socio-economic well-being of local communities to mitigate social vulnerability. This approach highlights the importance of integrating social dimensions into ecological assessments to inform more equitable and effective management strategies.

Internet of Things (IoT) aims at connecting all objects through the Internet. The IoT-based smart city is the concept of employing IoT devices to manage cities more easily, quickly and effectively. Recently, a new paradigm has emerged in the networking world, called software-defined networking (SDN), which separates the control data plane and the data plane. On the other hand, IoT devices embedded in a smart city often detect sensitive data. It is of paramount importance to prevent leaking this data unintentionally, known as privacy preserving. Although there are some methods to preserve data privacy in smart cities, they are either expensive or cannot preserve privacy efficiently. Intending to provide an efficient privacy preserving in a cost-effective smart city, we propose a novel method called differential privacy (DP)-preserving Smart City (DPSmartCity). It maintains the privacy of IoT devices’ sensitive data by equipping the environment with SDN technologies and leveraging a customized DP technique while its distribution changes frequently. We then mathematically investigate the DP aspect of the given algorithms by providing several proofs. The evaluation results show the effectiveness of the proposed method for diffident parameters, such as overload amount and latency points. Based on our threat model, it is 25.89% more robust against unintentional disclosure of sensitive data from a penetration point of view.

[This corrects the article DOI: 10.2196/44406.].