Pulmoset Classification Using Machine Learning Algorithms

This work aims to develop an automated lung cancer detection that uses Machine Learning techniques and addresses the weaknesses/drawbacks of the existing diagnostic practices. Unfortunately, lung cancer still ranks among the deadliest forms of cancer mainly because it is detectable only at an advanced stage. Effective screening improves chances of survival considerably, but the current avenues, for example, manual scanning of images, take too long, are too subjective, and are often accompanied by misreading’s in the human being. Machine learning and more specifically algorithms such as SVM, Random Forest and Decision Trees has been identified as suitable algorithms for the analysis of large datasets even when it comes to recognition of patterns that are likely to be overlooked by the expert –As the models and the techniques developed in the project will facilitate lung cancer recognition, its diagnosis will be relatively more accurate and efficient. The objective is to design a viable system that overcome important aspects related to the manual interpretation of images such as, time taken in the procedure and human bias in interpreting the results. Such advancement will enhance early diagnosis and may eventually lead to better lung cancer detection. This work embodies the significance of the need for integrated AI solutions into the healthcare ecosystem for diseases that need early intervention. If successfully carried out, lung cancer diagnosis could undergo significant transformation in terms of its universality and ease of implementation.