Validation & Diagnosis of Cystic Fibrosis Prognosis using Gradient Boost Decision Trees

Cystic Fibrosis (CF) is a recessive disorder which damages lungs, digestive track and any other organs. It is an inherited disease caused by a defective gene called cystic fibrosis transmembrane conductance regulator(CTRF). Various techniques have been proposed to detect CF at an early stage which helps prevent serious, lifelong problems related to CF. The present research provides a study and complete analysis on the available approaches; machine learning, deep learning and others. The performance measures in terms of accuracy, sensitivity, specificity and others found using various approaches are: K- Nearest Neighbours (KNN), Bayesian Optimization, Random Forest, Convolutional Neural Network (CNN), Support Vector Machine (SVM), Logistic Regression have compared with our proposed model Gradient Boost Decision trees (GBDT). The results depicts that the proposed model is outperforming in measures of accuracy, error rate, specificity, sensitivity and others.