City zones have become increasingly overcrowded due to the extensive population widening ratio and the swift relocation of people from villages. The traffic monitoring process is a significant issue in these areas due to the massive traffic flow on the roads. This research proposed a cluster-based improved authentication and communication protocol for an Intelligent Transportation System in Vehicular AdHoc Networks (VANETs). Our primary objective is to optimize resource sharing in vehicular communication. We enhanced the reliability, scalability, and stability of fast-moving VANETs by introducing cluster-based routing protocols for Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communications. We used a third-party certification authority for vehicle authentication to address security and privacy concerns. Our proposed protocol support minimizes the end-to-end (E2E) delay and route request and reduces link failure. Our protocol’s leading yield includes throughput enhancement, TCP Socket Initialization time minimization, TCP handshake response speedup, and DNS lookup improvement. The protocols are centered on short-range peer-to-peer (P2P) wireless communication in a 400 m radius cluster. They include innovative P2P wireless communications on VANET using minimized resources. The proposed protocols deliver a secure authentication mechanism with a securely generated vehicle authentication key provided by a certification authority. Furthermore, we have developed RESTful APIs in vehicular communication for implementation purposes and also offered and implemented algorithms for resource sharing regarding V2V and V2I communication. Ultimately, we evaluated the performance of our experiments.

Vehicular ad-hoc networks (VANETs) facilitate vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication for intelligent transportation systems (ITS). However, security remains a major concern for VANETs, especially with the emergence of 5G connectivity. Traditional security models like trust-based schemes are inadequate to protect against potential cyber threats. Moreover, delay in message delivery is another major issue caused by fast-moving vehicles in VANETs. Zero trust has become a new paradigm to strengthen security by eliminating implicit trust. This research presents a lightweight zero-trust framework for securing V2V and V2I communications in 5G VANETs using two algorithms. The framework enforces continuous authentication and access control using a centralized certification server (CCS) to authorize vehicle communication. Microsegmentation principles (visibility, granular security, and dynamic adaptation) divide the network into secure zones (clusters) and minimize lateral movement after a breach. The proposed framework integrates authorization, encryption, and anomaly detection to realize the zero trust vision. The framework is implemented using secure web services, and performance is measured using a third-party tool. The results of experimental analysis demonstrate that the framework's performance can effectively strengthen VANET security by preventing impersonation attacks and ensuring legitimacy of vehicular communication. The lightweight nature of the algorithms allows efficient realization of zero trust without excessive overheads. This proposed research highlights the promise of zero-trust principles for advancing security in 5G VANET ecosystems.

City zones have become increasingly overcrowded as a result of the extensive population widening ratio and the swift relocation of people from villages. The traffic monitoring process is a major issue in these areas due to the massive traffic flow on the roads. This research proposed a cluster-based improved authentication and communication scheme for an Intelligent Transportation System in Vehicular AdHoc Networks (VANETs). Our primary objective is to optimize resource sharing in vehicular communication. We enhanced the reliability, scalability, and stability of fast-moving VANETs by introducing cluster-based routing schemes for Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communications. We used a third-party certification authority for vehicle authentication to address security and privacy concerns. Our proposed schemes support minimizing the end-to-end (E2E) delay and route request, in addition to reducing link failure. Our schemes’ leading yield includes throughput enhancement, TCP Socket Initialization time minimization, TCP handshake response speedup, and DNS lookup improvement. The schemes are centred on short-range peer-to-peer (P2P) wireless communication in a 400-meter radius cluster. This includes innovative P2P wireless communications on VANET using minimized resources. The proposed schemes deliver a secure authentication mechanism with a securely generated vehicle authentication key provided by a certification authority. Furthermore, we have developed RESTful APIs in vehicular communication for implementation purposes, and also offered and implemented algorithms for resource sharing regarding V2V and V2I communication. Ultimately, we evaluated the performance of our experiments.

Accurate estimation of meal composition from food images can support safer and more reliable insulin bolus decision-making for individuals with Type 1 diabetes. Existing food recognition and nutrition estimation systems are often designed for general dietary logging and do not directly integrate food analysis with personalized insulin therapy parameters. This study presents an image-based nutrition estimation and insulin decision-support module developed within the AI-assisted Diabetes Care (AIDCARE) platform. The proposed system uses a convolutional neural network (CNN) to classify food items from a single meal image and retrieves reference nutritional values from a food composition database. A separate multimodal large language model (MLLM)-based estimation component is then used to estimate portion size, allowing carbohydrate and nutrient values to be scaled according to the observed serving. A curated food image dataset containing 40 food categories was used to evaluate three CNN architectures: ResNet50, Inception V3, and EfficientNet-B0. EfficientNet-B0 achieved the best classification performance, with 94.91% validation accuracy, 95.55% precision, 94.87% recall, and 94.90% F1-score. The portion-estimation component achieved an MAE of 12.27 g and an RMSE of 15.11 g. The estimated carbohydrate value is combined with user-specific clinical parameters, including the insulin-to-carbohydrate ratio and insulin sensitivity factor, to generate advisory bolus guidance. To support safety, the system requires user confirmation or correction of the recognized food category and estimated portion before insulin guidance is displayed. The proposed system is intended for advisory decision support only and is not designed to replace clinical judgment or autonomous insulin delivery systems.

Background/Objectives: Accurate estimation of meal composition from food images can support safer and more reliable insulin bolus decision-making for individuals with Type 1 diabetes. Existing food recognition and nutrition estimation systems are often designed for general dietary logging and do not directly integrate food analysis with personalized insulin therapy parameters. Methods: This study presents an image-based nutrition estimation and insulin decision-support module developed within the AI-assisted Diabetes Care (AIDCARE) platform. The proposed system uses a convolutional neural network (CNN) to classify food items from a single meal image, and retrieves reference nutritional values from a food composition database. A separate multimodal large language model (MLLM)-based estimation component is then used to estimate portion size, allowing carbohydrate and nutrient values to be scaled according to the observed serving. Results: A curated food image dataset containing 40 food categories was used to evaluate three CNN architectures: ResNet50, Inception V3, and EfficientNet-B0. EfficientNet-B0 achieved the best classification performance, with 94.91% validation accuracy, 95.55% precision, 94.87% recall, and 94.90% F1-score. The portion-estimation component achieved an MAE of 12.27 g and an RMSE of 15.11 g. The estimated carbohydrate value is combined with user-specific clinical parameters, including the insulin-to-carbohydrate ratio and insulin sensitivity factor, to generate advisory bolus guidance. To support safety, the system requires user confirmation or correction of the recognized food category and estimated portion before insulin guidance is displayed. Conclusions: The proposed system is intended for advisory decision support only and is not designed to replace clinical judgment or autonomous insulin delivery systems.

Background/Objectives: The objective evaluation of regional dietary intake remains a core challenge in personalized health management due to complex plate presentations and a lack of culturally specific dataset benchmarks. Methods: This study introduces a confidence-aware hybrid vision–language framework engineered for traditional Turkish food recognition and structured nutritional assessment. Results: We curate a balanced dataset containing 14,711 verified images spanning 40 representative Turkish culinary classes to train and evaluate seven deep learning architectures. Among the visual models, EfficientNet V2-L achieved the highest standalone performance with an accuracy of 93.47%, 0.92 macro-precision, 0.92 macro-recall, and a 0.92 F1 score. To overcome visual ambiguity and automate content analysis, a confidence-aware routing strategy escalates uncertain predictions ($\tau <0.70$) or user-rejected classifications to the Google Gemini 2.5 Flash multimodal large language model (MLLM). Conclusions: This hybrid paradigm yields a combined classification accuracy of 95.50% while validating portion weight estimations within a mean absolute error (MAE) of 18.42 g and total energy within 36.75 kcal. Fully realized as a cross-platform Flutter mobile application, the end-to-end pipeline demonstrates localized plate detection, adaptive portion analysis, and structured nutrient tracking, providing a scalable design for consumer-facing digital nutrition platforms.