Zenvgg1X: An Advance Hybrid Framework based on Gene Sequencing for Analysis and Prediction of Lung Cancers using KNN-GNN and Custom VGG19

Lung cancer affects about 238,340 people annually, of which men account for 117,550 cases whereas women have 120,790. Lung cancer causes roughly 127,070 deaths every year. The survival rate for lung cancer is only 56 percent if it can be detected within the lungs; however, it drops to just 5 percentage where there are distant tumors. As few as 16 percentage of cases can be diagnosed if the cancer is identified early. The study created a new lung cancer detection framework, Zenvgg1X a diagnosis and predictive framework combining VGG19, which is a deep convolutional neural network (CNN), and K-Nearest Neighbors - Graph Neural Network (KNN-GNN) hybrid. The combination of these two allows analyzing both images and texts, thus increasing the success rate of lung cancer detection. By working with medical images for identification purposes only the VGG19 model that is well recognized in terms of image classification analysis for lung cancer while KNN-GNN determine patterns from text related to diseases-which if you ask me, the combinations sometimes provide statistical outputs that are mindbogglingly high in accuracy such as 87.5 percent using only 30 percent of the data or more impressively reaching up to an astonishing figure around 98 percent with larger dataset. Further inclusion of Tp53 gene mutation data can enable early-stage prediction of lung cancer which may lead to improved survival rates. It goes without saying that this hybrid model represents a giant stride in medical diagnostics because it offers hope for what could eventually become a simple routine test for early diagnosis and treatment of lung cancer.