Patient Prognosis based on Lung Nodule Detection: A Review and Prediction Using Machine Learning and Deep Learning Techniques

Abstract

Lung Cancer Integrating Pathological and Radiological Features is a strong and descriptive title that effectively communicates the core aspects. It emphasizes the fusion of different data sources using a neural network approach for the purpose of understanding and predicting lung cancer outcomes. It's clear and concise, which is ideal for a project title. In this study, we provide a summary of the latest CAD methodologies that use deep learning to pre-process, segment, classify, and retrieve lung nodule data from CT scans, in addition to reduce false positives. Up to November 2020, academic conferences and publications were the source of a selection of articles. We go over the history of deep learning, go over some key points about lung nodule CAD systems, and evaluate the effectiveness of the chosen research over a range of datasets. Researchers and radiologists are able to acquire a better understanding of computer-aided design machine learning as well as deep learning methods for the detection, segmentation, classification, and retrieval of pulmonary nodules by reading this review. We review the effectiveness of existing methods, discuss their drawbacks, and suggest future lines of investigation for high-impact research.