Cancer Genetics and Epigenetics 2024, Vol.12, No.4, 210-222 http://medscipublisher.com/index.php/cge 210 Feature Review Open Access Multi-Modal Data Fusion Using AI for Colon Cancer Prediction Jiyun Zhao, Peishen Yu,Yan Zhang Faculty of Life Sciences and Medicine, Harbin Institute of Technology, Harbin, 150001, Heilongjian, China Corresponding author: zhangtyo@hit.edu.cn Cancer Genetics and Epigenetics, 2024, Vol.12, No.4 doi: 10.5376/cge.2024.12.0022 Received: 05 Jul., 2024 Accepted: 09 Aug., 2024 Published: 20 Aug., 2024 Copyright © 2024 Zhao et al., This is an open access article published under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Preferred citation for this article: Zhao J.Y., Yu P.S., and Zhang Y., 2024, Multi-Modal data fusion using ai for colon cancer prediction, Cancer Genetics and Epigenetics, 12(4): 210-222 (doi: 10.5376/cge.2024.12.0022) Abstract The integration of multi-modal data using artificial intelligence (AI) has shown significant promise in the prediction and management of colon cancer. This study explores the current advancements in AI-driven multi-modal data fusion techniques, which combine histopathological images, genomic data, and clinical information to enhance the accuracy of colon cancer prediction. Recent studies have demonstrated that AI models, particularly those employing deep learning and machine learning algorithms, can outperform traditional diagnostic methods by leveraging the complementary strengths of different data modalities. For instance, AI-based systems have been successfully applied to colorectal cancer pathology image analysis, lymph node staging, and the detection of liver metastases, showcasing improved diagnostic accuracy and prognostic capabilities. Furthermore, innovative frameworks such as Pathomic Fusion and multi-task learning approaches have been developed to integrate histopathology and genomic features, leading to better survival outcome predictions and personalized treatment plans. Despite these advancements, challenges such as data heterogeneity, the need for large-scale datasets, and ethical considerations remain. This study underscores the potential of AI in revolutionizing colon cancer prediction and highlights the necessity for continued research and collaboration across disciplines to fully realize its benefits. Keywords Multi-modal data fusion; Artificial intelligence; Colon cancer prediction; Deep learning; Genomic data integration 1 Introduction Colon cancer, also known as colorectal cancer (CRC), is a significant health concern worldwide, being the third leading cause of cancer-related deaths in North America (Kapelanski-Lamoureux et al., 2023). The high mortality rate is primarily due to the late-stage diagnosis and the development of metastases, particularly in the liver, which occurs in over 50% of CRC patients (Rompianesi et al., 2022; Kapelanski-Lamoureux et al., 2023). Early and accurate prediction of colon cancer can significantly improve patient outcomes by enabling timely and targeted interventions. Traditional diagnostic methods often fall short in providing the necessary precision, leading to a pressing need for advanced predictive tools. Artificial Intelligence (AI) has emerged as a promising solution in the field of oncology, offering enhanced capabilities in data analysis and pattern recognition. AI-powered tools have shown potential in various aspects of cancer management, including diagnosis, prognosis, and treatment response prediction (Huang et al., 2020; Lim et al., 2023; Mansur et al., 2023). For instance, AI algorithms can analyze complex multi-omic data to identify biomarkers and predict treatment responses, thereby facilitating personalized medicine approaches (Rompianesi et al., 2022; Kapelanski-Lamoureux et al., 2023). The integration of AI in medical imaging and histopathological analysis has also improved the accuracy and efficiency of cancer detection and characterization (Chiu et al., 2022; Lim et al., 2023). This study provide a comprehensive overview of the current state and future directions of multi-modal data fusion using AI for colon cancer prediction. It explore the various AI methodologies employed in integrating diverse data types, such as genomic, proteomic, and imaging data, to enhance the predictive accuracy for colon cancer. The study also discuss the clinical implications of these AI-driven approaches, highlighting their potential to revolutionize colon cancer management by enabling early detection, precise prognostication, and personalized treatment strategies. By synthesizing findings from recent studies, this study underscore the transformative impact of AI in the realm of colon cancer prediction and to identify areas for future research and development.
RkJQdWJsaXNoZXIy MjQ4ODYzNQ==