Patient scheduling is the process of scheduling and sequencing the patients for various multiple resources in health care domain. The multiple constraints and multiple goals to be achieved in minimal time makes this problem highly complex. Computational complexity is high in using exact methods for solving optimization problem. Motivated by the real needs in hospital environments, this paper focuses on finding an optimal schedule using the meta-heuristic technique Particle Swarm Optimization (PSO) and coordinating the hospital environment using multi-agents. Agents have been proved to be an effective approach to resource allocation because of its coordination and social abilities. For solving, various agents like Patient Agent (PA), Resource Agent (RA) and Common Agent (CA) are used. In addition to these agents a PSO Agent is used to perform the PSO optimization. On arrival of the patients, this PSO agent is called to perform PSO optimization dynamically and an optimized schedule is generated. The objective is to reduce the patient waiting time in the hospital. This agent based approach is implemented in JADE (Java Agent Development Environment) and tested for different data sets. The results are compared with the traditional dispatching rules like First Come First Serve(FCFS), Minimum Slack(MS),Shortest Processing Time(SPT) and Longest Processing Time (LPT). The performance improvement of total weighted waiting time in Agent-based Patient Scheduling using PSO (APS-PSO) for 10 resources and 50 patients is 4.13% when compared to the best performing dispatching rule MS and 52% with respect to LPT. Similarly, there is an average of 8.69% improvement in terms of total weighted completion time. When the number of late patient is considered, the performance of the PSO based approach increased by 17%

Diabetes mellitus is considered one of the main causes of death worldwide. If diabetes fails to be treated and diagnosed earlier, it can cause several other health problems, such as kidney disease, nerve disease, vision problems, and brain issues. Early detection of diabetes reduces healthcare costs and minimizes the chance of serious complications. In this work, we propose an e-diagnostic model for diabetes classification via a machine learning algorithm that can be executed on the Internet of Medical Things (IoMT). The study uses and analyses two benchmarking datasets, the PIMA Indian Diabetes Dataset (PIDD) and the Behavioral Risk Factor Surveillance System (BRFSS) diabetes dataset, to classify diabetes. The proposed model consists of the random oversampling method to balance the range of classes, the interquartile range technique-based outlier detection to eliminate outlier data, and the Boruta algorithm for selecting the optimal features from the datasets. The proposed approach considers ML algorithms such as random forest, gradient boosting models, light gradient boosting classifiers, and decision trees, as they are widely used classification algorithms for diabetes prediction. We evaluated all four ML algorithms via performance indicators such as accuracy, F1 score, recall, precision, and AUC-ROC. Comparative analysis of this model suggests that the random forest algorithm outperforms all the remaining classifiers, with the greatest accuracy of 92% on the BRFSS diabetes dataset and 94% accuracy on the PIDD dataset, which is greater than the 3% accuracy reported in existing research. This research is helpful for assisting diabetologists in developing accurate treatment regimens for patients who are diabetic.

Software Defined Networking (SDN) is a new approach in network architecture which decouples the control plane from data plane. The SDN controller has the global view of the underlying network, enabling enormous research works in the area of traffic engineering (TE). Existing congestion control solutions for SDN TE are either reactive or proactive in nature. In reactive schemes when congestion event occurs the source is informed which then reacts by rate adjustment and in proactive scheme pre computation of protection path demands additional resources. We propose a proactive flow monitoring method to reduce congestion and packet loss in SDN. This technique reduces packet loss by automatically re-routing congested link's traffic through a dynamically computed alternate path. Simulation result shows that proposed scheme reduces average packet loss leading to overall increase in network efficiency.