Diagnostic Analysis of Diseases With Key Symptoms of Bad Breath: Integration of Naïve Bayes for Early Detection

The study aims to develop an expert system that integrates Naïve Bayes’ algorithm for the early detection of various diseases, including Diabetes, Liver Disorders, Kidney Disorders, GERD, and Sinus Infections, based on the analysis of the main symptoms of bad breath. The study involved 1964 individuals, with 1473 of them being patients who had been diagnosed with one or more diseases and another 491 being healthy control subjects. Clinical data and samples of bad breath were collected and analyzed to develop a predictive model. The accuracy of this expert system is then compared to the diagnosis given by the medical expert. The results of testing on 200 test data showed that the system successfully matched the expert’s diagnosis with a success rate of 95.5%. These findings confirm that bad breath is reliable as an effective non-invasive diagnostic biomarker. The implementation of the Naïve Bayes algorithm demonstrates its outstanding ability to process and analyze diagnostic data for accurate predictions. This study shows the great potential of the integration of predictive technology in clinical medical practice, offering an efficient and non-invasive method for early detection of diseases, which can improve clinical outcomes and patients’ quality of life.