A Novel Enhanced Prediction of Lung Cancer Histopathological Medical Image Processing Using Hybrid Deep Learning

Lung cancer is a serious disease that affects human organs. The type and location of cancer will mostly determine the course of medical treatment. Creating an automated method is crucial to the early detection of malignant conditions. On the other hand, conventional techniques are still employed to diagnose illnesses. Due to the accuracy and efficiency limits of traditional screening techniques like CT scans and X-rays, many cases of lung cancer are discovered later on, when treatment is already underway. There are few choices. In this research paper, we implement a neural network hybrid classifier to construct an extremely precise and effective method for lung cancer prediction. To begin with, we create a sizable collection of patient information and medical photos for training and assessment of the CNN and RNN Classifier. A range of imaging modalities, such as CT scans, X-rays, and other medical pictures, are included in the collection. Next, we create and train a CNN classifier to identify lung cancer using a hybrid deep-learning methodology. To achieve an earlier and more accurate diagnosis of lung cancer, the NN classifiers were improved for high accuracy and efficiency. Additionally, datasets for several CT scanning images, including axial, coronal, and sagittal, have been produced. However, compared to the traditional 3D-CNN and regular CNN we can’t get the exact results, but in our approach, we included the CNN model along with the RNN model, if we combined both models, we produced more aesthetically beautiful results and yielded more accurate results regarding diseases. CNN model, which analyzes historical data to detect lung cancer, was unable to yield reliable disease detection findings because they used only the CNN model. The resultant accuracy is obtained by RNN is $\mathbf{9 2 \%}$ and CNN is $\mathbf{9 6 \%}$.