Kyphosis Disease Prediction with help of RandomizedSearchCV and AdaBoosting

This Excessive inward arching of the upper back is known as kyphosis. Kyphosis in older persons is frequently brought on by the spinal bones' weakening, which leads to their compression or cracking. Other forms of kyphosis may manifest in children or adolescents as a result of spinal deformity or gradual twisting of the spinal bones. As a result, we applied multiple machine learning techniques in this research to forecast the kyphosis illness using biological data. The main focus of this study is the categorization of kyphosis illness in children who have already had spinal fusion surgery. This dataset, which comprises 81 rows and 4 columns, has 3 inputs and single output. In this work, our main objective is to apply several machine learning methods, such as Random Forest, Decision Tree, Logistic Regression, and Gradient Classifiers, to biological data and to assess the accuracy of the aforementioned algorithms. The features are scaled using the Power Transform and Quantile Transformer outliers. Adaboost classifier, hyperparameter adjustment, and Polynomial Feature Selection Method produced the finest outcomes. This demonstrates that ML should be accepted and used as an essential technique for resolving biological issues generally.