Predicting Time to Event Outcomes in Lung Cancer Detection Using Deep CNN
Inayathulla Mohammed, Narendra Sai Vaddhi, Manasa Sandra, Manoj Kumar Jetti, Keerthana Binde, Ajay Nallabothula
Lung cancer remains one of the most lethal malignancies worldwide, with early detection and accurate
prognosis playing crucial roles in improving survival rates. Predicting time to event outcomes, like the
progression of cancer or patient survival, is essential for personalizing treatment strategies. It is a deep
learning-based method using convolutional neural network CNN for predicting time to event outcomes
in lung cancer diagnosis. SVM is chosen for its robust performance in high dimensional spaces and
effective handling of small sample sizes. The prediction accuracy of the SVM model is compared with
that of a deep CNN to evaluate the strengths and weaknesses of each method. The dataset comprises
clinical, genetic, and imaging data from lung cancer patients, preprocessed for model training and
validation. Performance metrics such as accuracy, precision, recall, and the area under the ROC curve
(AUC) are used to access both models.
