Scientists at the University of Sharjah have revealed in a new study that oncologists are making progress in the early detection of pancreatic cancer and understanding its progression through the use of artificial intelligence (AI). The findings, published in the Beni-Suef University Journal of Basic and Applied Sciences, offer fresh hope for the diagnosis and treatment of pancreatic cancer.

Pancreatic cancer has a high mortality rate, with 467,409 deaths and 510,992 new cases reported worldwide in 2022. Often called the "king of cancers," it is typically diagnosed at an advanced stage due to the absence of distinct molecular markers and clinical symptoms. Researchers emphasize that early detection and accurate staging are critical to improving treatment outcomes. In the study, they provide a concise overview of how AI is being applied to pancreatic cancer diagnosis, prognosis, and treatment, noting that AI-driven preliminary screening can significantly improve patient outcomes. AI-powered image analysis technologies have the potential to transform computer-assisted diagnostic systems and help physicians achieve more precise assessments.

Multi-omics integration is a key focus of the research, requiring the combination and analysis of diverse data types and expert insights to gain a deeper understanding of pancreatic cancer. The researchers highlight the major significance of AI in the multi-omics domain, stating that the healthcare sector is entering a new era thanks to scientific and technological advances and the integration of AI with healthcare. This progress is the result of collaborative efforts across multiple disciplines. Although computer systems have limitations, their powerful processing capabilities are expected to deliver substantial breakthroughs.

The authors have endowed AI models with the ability to detect pancreatic tumors at an early stage, helping physicians assess risk and formulate treatment plans. However, they also call for better mastery and control of these models, noting that their operation and interpretation are not straightforward, and the multitude of solutions complicates clinical adoption. Physicians need clear interpretability to evaluate the applicability and reliability of algorithmic outputs. Despite the complexity of AI applications, researchers are working to develop methods that make these tools usable by healthcare professionals and trustworthy for patients. They predict that advances in explainable AI will reduce the barriers to adoption and make the tools more clinically acceptable. Cancer researchers are already creating and applying explainable AI methods to interpret predictions. Additionally, Internet of Things (IoT) approaches are gaining attention among oncology researchers and have the potential to revolutionize the detection, prognosis, and treatment of pancreatic cancer. AI can also assist oncologists in formulating personalized treatment plans by integrating patient data and predicting responses to various therapies. The authors call for more AI-based pancreatic cancer research and the development of "semi-autonomous models that can reduce clinicians' workload, increase productivity, or operate fully autonomously."