Gwinnett Technical College

Abstract - Healthcare fraud detection is an important challenge in modern healthcare systems due to the large volume of medical claims and the highly imbalanced nature of fraud datasets. Fraudulent claims represent only a small portion of the total claims, which makes it difficult for traditional machine learning models to accurately identify fraudulent activities. Conventional techniques such as Random Oversampling (ROS), Synthetic Minority Oversampling Technique (SMOTE), and Random Undersampling (RUS) are commonly used to address class imbalance, but they often lead to issues such as overfitting, noise generation, or loss of important information. In this study, a machine learning-based approach is proposed to improve Medicare fraud detection using the Medicare Part B dataset. The proposed framework applies a hybrid resampling technique called SMOTE-ENN, which combines SMOTE for generating synthetic minority samples with Edited Nearest Neighbors (ENN) to remove noisy and irrelevant data instances. Logistic Regression is used as the classification algorithm to detect fraudulent healthcare claims. The model performance is evaluated using multiple metrics, including accuracy, precision, recall, F1-score, AUC-ROC, and Area Under the Precision-Recall Curve (AUPRC). Experimental results demonstrate that the proposed approach achieves an accuracy of 98%, indicating its effectiveness in handling imbalanced datasets and improving fraud detection in healthcare systems. Key Words: Healthcare Fraud Detection, Machine Learning, SMOTE-ENN, Logistic Regression, Imbalanced Data, Medicare Claims.

BackgroundChronic diseases impose a sustained burden on healthcare systems through progressive deterioration and long-term costs. Although adherence-enhancing interventions are widely promoted, their return on investment (ROI) remains uncertain, particularly under heterogeneous patient behavior and socioeconomic variation.MethodsWe developed a simulation-based framework integrating disease progression, time-varying adherence, and policy timing. Cumulative healthcare costs were modeled over a 10-year horizon using continuous-time stochastic formulations calibrated with Medical Expenditure Panel Survey (MEPS) data stratified by income. ROI was estimated across adherence gains (delta) and policy costs (gamma).ResultsEarly and adaptive interventions yielded the highest ROI by sustaining adherence and slowing progression. ROI exceeded 20% when delta ge 0.20 and gamma le 1.5, whereas low-impact or high-cost policies failed to break even. Subgroup analyses showed a 32% ROI gap between the lowest and highest income strata, with projected savings of $312 per patient versus baseline. Sensitivity tests confirmed robustness under stochastic adherence and inflation variability.ConclusionsThe framework offers a transparent, adaptable tool for evaluating cost-effective adherence strategies. By linking behavioral effectiveness with fiscal feasibility, it supports design of robust and equitable chronic disease policies. Reported ROI values represent conservative lower bounds, and extensions incorporating DALYs and QALYs illustrate scalability toward health outcome integration.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12889-025-25279-3.

Internet of Things (IoTs) are becoming one of the integral parts of our lives, as all of the modern devices including pervasive systems use internet for its connectivity with the rest of the world. The Radio Frequency IDentification (RFID) provides unique identification and nonline of sight capabilities, therefore plays a very important role in development of IoTs. However, the RFID systems incorporate wireless channel for communication, therefore have some allied risks to the system from threat agents. In order to prevent the system from malicious activities in a cost effective way, numerous Ultralightweight Mutual Authentication Protocols (UMAPs) have been proposed since last decade. These UMAPs mainly involve simple bitwise logical operators such as XOR, AND, OR, etc., in their designs and can be implemented with extremely low cost RFID tags. However, most of the UMAP designers didn’t provide the proper hardware approximations of their UMAPs and presented only theoretical results which mostly mislead the reader. In this paper, we have addressed this problem by reporting our experiences with FPGA and ASIC-based implementation of UMAP named psuedo Kasami code-based Mutual Authentication Protocol (KMAP[Formula: see text]. Further, we have also improved the structure of the KMAP protocol to overcome the previously highlighted attack model. The hardware implementation results show that KMAP[Formula: see text] successfully conform to EPC-C1G2 tags and can be implemented using less than 4[Formula: see text]K GE (for 32-bit word length).

Whether you are new to the world of game development or a veteran of the field, you have most likely heard of, perhaps even taken part in, a game jam. An issue many teams face is "how do I accomplish everything I want with so little time?" This paper delves into the techniques and methods used by my team, Null Interactive, during Global Game Jam 2017. Furthermore, we discuss the product we created, strategies for using the limited time we were given, and the progression of our game jam experience.

ASCE has been developing and releasing periodic reports on the status of the nation's infrastructure since 1998. Over the years periodic reports on infrastructure status have also been developed for several states and local jurisdictions. Infrastructure reporting is helpful for sending political decision makers and their constituencies messages about the relative urgency of investing in civil infrastructure in the context of other national priorities, such as education, health care, and security. Periodic infrastructure reporting is also a formal exercise that states and local agencies can use to demonstrate their accountability as infrastructure stewards. This paper presents a decision support tool that was developed to enhance the infrastructure report card process in the state of Georgia. The Infrastructure Rating Tool (IRT) is based on multiple attribute decision making (MADM) and helps the evaluator to make more explicit the scoring criteria used to evaluate the different infrastructure categories, the weights assigned to these criteria to develop an aggregate score, and the scale used in determining the final grade. The tool was developed based on the 2009 ASCE Report Card methodology and first applied in the release of the 2009 ASCE Georgia Infrastructure Report Card. Given that there are subjective elements in the infrastructure reporting process, decision support tools such as the IRT can help to make the process more transparent and objective, and result in the development of more credible decision support information.

A laser demonstration has been developed to introduce students to the wonder of science. The demonstration builds on the natural wonder and adds electronics, audio amplification and modulation, music transmission, stroboscopy, motors, linear integrated circuits and plasma generators to the list of topics associated with the demonstration.

Die vorliegende Notiz behandelt das Problem der Spannungskonzentration an einem kreisformigen Loch in der Theorie der mikropolaren Elastizitat. Es wird gezeigt, dass der Konzentrationsfaktor im Gegensatz zum klassischen Fall nicht konstant ist, sondern von drei dimensionslosen Parametern abhangt, welche mit den Materialkonstanten verknupft sind. Bei passender Wahl dieses Parameters stellt sich heraus, dass der Konzentrationsfaktor um wenig vom klassischen Wert abweicht.