A Comprehensive Review on Accurate Mesothelioma Classification Using Medical Imaging and AI Techniques

Abstract

Mesothelioma is a rare and aggressive malignancy primarily associated with asbestos exposure, and its early diagnosis remains challenging due to subtle visual differences and overlapping clinical features. Medical imaging modalities, particularly computed tomography (CT), play a vital role in identifying pleural thickening, effusions, and tumor spread, while histopathological images provide cellular-level evidence essential for definitive diagnosis. In recent years, artificial intelligence (AI) techniques have demonstrated significant potential in enhancing the accuracy and consistency of mesothelioma classification using both CT and histopathological data. This comprehensive review analyzes existing research on mesothelioma classification employing medical imaging and AI-based approaches. It covers conventional image processing methods, machine learning techniques, and advanced deep learning models applied to CT scans and histopathological images. The review highlights the importance of feature extraction, multimodal data fusion, and robust preprocessing strategies for improving diagnostic performance. Additionally, the role of explainable AI is discussed to address clinical interpretability and trust in automated systems. Comparative performance analysis, commonly used datasets, and evaluation metrics are summarized to identify current trends, limitations, and research gaps. Finally, challenges such as data scarcity, class imbalance, and real-world clinical deployment are discussed, providing directions for future research toward reliable AI-assisted mesothelioma diagnosis.