Publication Details
Abstract
Lung cancer is a widespread and life threatening disease that has international concerns, and requires innovative methods of identifying it at an early stage and classifying it. In the present paper, a new solution, utilizing the newest machine learning algorithms to enhance the effectiveness and accuracy of lung cancer diagnostics, is presented. The proposed solution will integrate the most recent image-processing algorithms and an effective classification model to analyze the medical imaging data as a whole. The motivation behind this study is that the literature review is done in a critical fashion and gaps that exist in the current literature are determined by this approach and opportunities that machine learning offers to medical imaging. It is this set of gaps that we are targeting, and this approach is a mixture of image preprocessing techniques, feature extraction techniques, along with a state of the art machine learning model. The implementation model is described with step-by-step processes, tools and technologies applied to train and validate on a mixed dataset. Experimental results verify our technique is effective, and the sensitivity, specificity, and the overall accuracy of the technique are better than the existing procedures. These discoveries can be significant due to the fact that they will present a useful and timely diagnostic instrument to the health professionals in order to attain better patient outcomes. In conclusion, the paper can be said to contribute to the field of lung cancer detection as well as predetermine the subsequent improvements and extensions of its application in medical imaging.