Title: |
Automated Disease Detection: A Diabetes Case Study Using Advanced Machine Learning Algorithms |
Authors: |
Sohaib Najat Hasan |
Source: |
International Journal of Latest Engineering Research and Applications, pp 01 - 12, Vol 11 - No. 09, 2026 |
Abstract: |
In the world as a whole, diabetes mellitus ranks among the most prevalent chronic diseases and is a major public health issue due to the various diseases that may result from its complications, including cardiovascular diseases, kidney failure, neuropathy and vision impairment. This is why the early detection of those who are at risk of developing diabetes is key to ensuring the appropriate clinical intervention, and minimize long-term health effects. We suggest a framework for automating the disease detection process as case study of diabetes detection, and compare the performances of state-of-the-art machine learning algorithms for diabetes infection of the clinical and demographic data. A number of supervised learning algorithms including logistic regression, support vector machine, KNN, decision tree, random forest, XGBoost and ANN are created and benchmarked. This involves using techniques such as handling missing values and data normalization, feature selection, and class balancing methods, in order to make models more reliable and more revealing. The accuracy, precision, recall, F1-score, sensitivity, specificity and area under the receiver operating characteristic curve are used to assess the proposed models. Special attention is given to the ability to find models with high sensitivity and at the same time have an acceptable false-positive rate, especially for medical screening applications. The experimental design shows the promise of machine learning in helping to automate and informatise diabetes detection, and in helping healthcare professionals to find high-risk patients earlier. The study also offers a comparative basis to choose appropriate machine learning methods for the future intelligent clinical decision-support systems. |
Kaywords: |
Diabetes detection, machine learning, disease prediction, artificial intelligence, classification, clinical decision support, ensemble learning, healthcare analytics. |
 |
Download Full Article |
Title: |
Exploring the Synergies of Dedicated Metering, Energy Management Systems, and Time-of-Use Rates for Electric Vehicles: A Sustainable Governance Framework Towards Urban Microgrids and Circular Economy |
Authors: |
Ming-Fong Jhang, Chun-Shuo Hsiao |
Source: |
International Journal of Latest Engineering Research and Applications, pp 13 - 19, Vol 11 - No. 09, 2026 |
Abstract: |
Amid the global transition toward net-zero emissions, installing electric vehicle (EV) charging infrastructure in existing residential complexes faces capacity limits, perceived risks, and severe collective action problems. This study develops a governance-oriented framework integrating the "Dedicated Meter Power Supply" policy, Energy Management Systems (EMS), and Time-of-Use (TOU) tariffs to address these structural barriers. Taking Community S in Taichung City (264 units, 306 parking spaces) as an empirical case, the research analyzes how management committees can utilize financial engineering to overcome special resolution thresholds. Results demonstrate that an innovative "early-bird dynamic refund" fundraising model triggered significant economies of scale, slashing per-unit infrastructure costs by 59.1% and achieving a 4.9-fold oversubscription rate. Furthermore, EMS dynamic load control perfectly aligned with Taipower’s 18-hour off-peak window achieves optimal peak shaving, reducing user energy charges by over 60%. By integrating the Technology Acceptance Model (TAM), Innovation Diffusion Theory, and Markov Chain market forecasting, this study theoretically confirms that extreme peak-to-off-peak price spreads effectively internalize the external costs of grid congestion. This framework not only advances SDG 12 (Responsible Consumption) but also accelerates the market transition from internal combustion engine (ICE) vehicles to battery electric vehicles (BEVs), maximizing carbon reduction in the Tank-to-Wheel (TTW) phase and laying a scalable foundation for future Vehicle-to-Grid (V2G) microgrids. |
Kaywords: |
Circular economy, Collective action problem, Dedicated metering, Energy management system (EMS), Technology acceptance model (TAM), Time-of-use (TOU) rates |
|
Download Full Article |
DOI: |
10.56581/IJLERA.11.09.13-19 |