Tuberculosis Prediction using KNN Algorithm

Abstract

In this paper, a machine learning model is used to develop a model that is used for tuberculosis prediction. Tuberculosis is known to be one of the top reasons for death from an infectious agent that affects the lungs and continues to threaten the human population on a wider basis. According to WHO, tuberculosis is a serious threat to the human population after HIV/AIDS. It is estimated by the World Health Organization (WHO) that 1/3^rd of the global population is infected with TB and that seven to eight million new cases of TB occur each year across the globe Because the disease is difficult to differentiate between the common cold, it takes a long time to decide the patient is affected by the disease. So we use the detection of tuberculosis by utilizing the K-NN algorithm method for classification and HOG as feature extraction. K-NN abbreviated as K-Nearest Neighbour is one of the simplest Machine Learning algorithms based on the Supervised Learning technique.

The data provided K-NN model should be labeled one. Then these datasets are given to a training model where the training process of the model is being undergone. Once the training is completed, the next step is to predict the output. For this process, we have to provide new data that may or may not belong to the dataset, so that the model can predict the output of it. If the prediction is wrong, again the training is done until we get the actual output matching with the desired output given by the designer for verification purposes. This is the basic working process under the K-NN algorithm. The data that is used for this separation is a Tuberculosis dataset that contains various information about the different symptoms that are helpful in detecting tuberculosis effectively. Here it is used in the early detection of tuberculosis which helps save millions of people which might otherwise lead to death because of lack of detection. ML model helps to improve the efficiency in detecting by considering various symptoms. ML models are more accurate at differentiating even the slightest difference that deviates from the data that was used to train the model. Unlike the manpower we fail to detect the slightest as we notice the symptoms only after they become more severe. The accuracy of this model was found to be 98%. The following model uses a dataset consisting of data that contrasts between males and females and the various symptoms are shown in them. It also contrasts the severity of these two.