A Nomogram Approach with Logistic Regression and Naïve Bayes to Identify Chronic Kidney Disease on Diabetes Type 2

A novel Nomogram model, integrating both logistic regression and naïve bayes methodologies, exhibits a promising capability in deciphering the molecular indicators of chronic kidney disease triggered by diabetes type 2 symptoms. This study employs the nomogram technique to pinpoint the manifestation of chronic kidney disease linked to elevated blood sugar levels. We meticulously examined molecular imaging of rat kidney tissues, identifying the presence of GLUT9 and SGLT2. Their presence is crucial as they instigate an increase in glucose re-absorption in the kidneys, fast-tracking kidney deterioration. The Nomogram model stands out by efficiently computing class probabilities based on distinct attributes. These attributes are derived from quantified observations of Tubules Proximal and Tubules Distal in rat kidneys, all stemming from meticulous wet lab experiments. Utilizing domain expert insights, we identified pivotal molecular traces for our investigation. To ascertain the prevalence of chronic disease in rats, the gathered dataset underwent classification through logistic regression and naïve bayes algorithms. The subsequent representation through the Nomogram model delivered an interactive and detailed analysis. Our results underscore the Nomogram’s superior performance in detecting symptoms relying on the applied classifiers.