Machine Learning Based Lung Cancer Detection with Multiview Images

Lung cancer presents a significant global health challenge, underscoring the crucial need for early detection to improve patient outcomes. This study introduces an innovative approach by combining Retina Net, an advanced object detection model, with machine learning techniques to enhance the accuracy and efficiency of identifying lung nodules in computed tomography (CT) scans. The proposed system leverages Retina Net’s exceptional precision, achieved through training on a diverse dataset of annotated lung CT scans, enabling it to effectively recognize and categorize nodules, distinguishing between malignant and benign cases. The evaluation includes benchmarking against existing detection methods and assessing computational efficiency for clinical applicability. Moreover, the research introduces a systematic framework for CT scan analysis, informed by a comprehensive literature review. This framework integrates pre-processing, segmentation, feature extraction, and classification blocks to address imperceptible noise and improve image quality. Machine learning techniques are then applied for image classification based on extracted features, with a primary focus on detecting lung cancer and implementing preventive measures. The integration of Retina Net for nodule detection and the structured framework for CT scan analysis forms a comprehensive strategy. These combined models aim to increase precision, reduce false positives, and contribute to more effective tools for early lung cancer diagnosis. The research emphasizes the practical viability of these approaches in clinical settings, striving to alleviate the burden of this deadly disease and enhance overall patient outcomes.