A Feline-Inspired Optimizer Enhanced through Self-Improvement, Coupled with Machine Learning, for the Identification of Lung Cancer in CT Scans

• Published in: 2024 Second International Conference on Advances in Information Technology (ICAIT) Vol. 1; pp. 1 - 6
• Main Authors: Arunachalam, P, Geetha, S., Nessariose Jose, Naduvathezhath, G, Vivekanandan, Shrutika, H S, Sowmya shree
• Format: Conference Proceeding
• Language: English
• Published: IEEE 24-07-2024
• Subjects: Artificial Intelligence in Healthcare; Cat Mouse Optimizer Algorithm; Classification algorithms; Clinical Decision Support Systems; Collaboration; Computed tomography; Computed Tomography (CT) Imaging; Deep Learning; Early Detection; Healthcare Technology; Image Classification; Lung cancer; Lung Cancer Diagnosis; Machine Learning; Machine learning algorithms; Medical diagnostic imaging; Medical Image Analysis; Pattern Recognition; Precision Medicine; Radiology; Self- Upgrading Algorithms; Technological innovation; Time factors; Training; Transforms
• DOI: 10.1109/ICAIT61638.2024.10690312

The study introduces an innovative breakthrough in the analysis of medical images, particularly focusing on the timely identification and categorization of lung cancer using computed tomography (CT) scans. Employing the pioneering Cat Mouse Optimizer algorithm, which has undergone enhancements via self-improving mechanisms, we've crafted an intricate framework. This framework seamlessly integrates cutting-edge machine learning algorithms to streamline the process of diagnosing lung cancer from CT scans with unparalleled precision and efficiency. Conventional methods for diagnosing lung cancer often hinge on manual interpretation of CT images by radiologists, a process susceptible to time constraints and human error. Our methodology tackles these challenges by capitalizing on the capabilities of artificial intelligence and deep learning methodologies. Through extensive training on extensive datasets of annotated CT scans, our model becomes adept at identifying nuanced patterns and indicators of lung cancer, leading to robust classification outcomes. The hallmark attributes of our system encompass its flexibility and scalability, enabling continual enhancement and adaptation over time. The self-enhancing features of the Cat Mouse Optimizer algorithm empower the system to autonomously fine-tune its algorithms and parameters, informed by feedback from new data and user interactions. This guarantees that the system remains abreast of the latest advancements in medical imaging and lung cancer research. Moreover, our framework underscores the importance of interpretability and transparency, furnishing clinicians with invaluable insights into the decision-making process of the AI model. This cultivates trust and collaboration between human experts and AI systems, culminating in more precise diagnoses and enhanced patient outcomes. Through rigorous experimentation and validation across varied datasets, we showcase the effectiveness and dependability of our methodology in real-world clinical scenarios. Our system exhibits promising outcomes in terms of sensitivity and specificity, surpassing existing approaches to lung cancer classification. In essence, our project marks a significant stride forward in the domain of medical image analysis, presenting a potent tool for the early detection and diagnosis of lung cancer. By amalgamating the strengths of advanced machine learning techniques and self-improving algorithms, we aspire to transform the landscape of lung cancer detection and management, ultimately contributing to saving lives and advancing healthcare outcomes.