부산대학교 도서관

Predicting and detecting human health conditions in the early stages is a crucial thing to be considered in the medical field. Lung tumour has been one of the most often common malignancies in the past few decades, as well as the high incidence of melanoma-related fatality globally. As a cost-effective alternative, computer-aided diagnostic systems can combine complicated features and identify a patient's probability of getting a lung tumour, reducing the need for unneeded and costly medical treatments. The computing profession is completely digitizing everything, and the healthcare profession is following suit with image identification and data analytics. As a modern solution to predict the presence of tumours in scan images aConvolutional Neural Network (CNN),Deep Learning (DL) model is proposed in this study. The dataset required for this study is collected from the Kaggle website and pre-processed using various pre-processing methods to make it compatible with the DL model. After pre-processing the data, the dataset is divided into a training set, testing set and validation set. The CNN model is trained using the training set, validated using the validation set and tested for efficiency using the testing set. After training and testing the CNN model a user-friendly webpage is created to make this CNN model easily accessible to the individuals. The webpage is designed in such a way that when the patients upload their lung CT scan image and enter the necessary contact details requested by the webpage, the scan is analysed and evaluated by the trained CNN model and the result is sent as an email to the user.