Abstract
Lung infections continue to pose a serious challenge to global public health, often progressing into life-threatening respiratory illnesses if not diagnosed and managed in a timely manner. Conventional diagnostic techniques, although widely used, can sometimes fall short in providing early and accurate predictions due to the variability and uncertainty present in clinical data. This study introduces an intelligent prediction model based on fuzzy logic systems aimed at addressing these limitations. Fuzzy logic offers a unique capability to handle imprecise and ambiguous information, closely resembling the reasoning patterns of medical professionals. The proposed system integrates a range of clinical parameter including symptoms, vital signs, and other patient-specific data to evaluate the likelihood of lung infection. The model is designed to interpret overlapping and non-binary data inputs, producing reliable infection risk assessments. Using a curated dataset of anonymized patient records, the fuzzy logic system was trained and tested to measure its prediction accuracy. Results reveal that the model demonstrates strong performance in detecting potential lung infections at early stages, offering a promising supplementary tool for clinicians. By enhancing the speed and precision of infection prediction, this approach can lead to faster interventions, reduced hospitalization rates, and better overall healthcare outcomes. The study highlights the practical benefits of fuzzy logic in medical diagnostics and its potential for broader clinical applications.
Keywords: Accuracy, Fuzzy Logic, Lung Infection Levels, Memberships Functions, Precision, Recall.