AI-driven Lung Cancer Detection for Rapid Analysis of Medical Imaging Data

The modern healthcare industry has been focusing on developing AI solutions for diagnosis and clinical decision support to understand and tackle challenges in various therapeutical areas using different datasets and sources. The proposed methodology involves developing AI-based solutions for lung cancer diagnosis that provide solutions to accelerate and improve the analysis of screening tests and medical imaging reports to easily process large amounts of data and immediately extract insights or detect patterns from laboratory operators and analysts. The X-ray dataset is trained to extract features that are relevant to identifying valuable insights regarding a patient’s condition, the development of a disease, or even the molecular structure of cells. The AI models used for image captioning and processing are MobileNet, DenseNet, ResNet, and Inception, and these models are trained in Google Colab using Python. Prediction captions and graphical analysis such as the epochs versus loss plot help to compare the loss parameter and accuracy of four AI models. Results of the data augmentation and accuracy plot are presented for better visualization of the affected area of the lungs.