مقالات پذیرفته شده در نهمین کنگره بین المللی زیست پزشکی
The Role of Artificial Intelligence Algorithms in Endometrial Cancer Prediction: A Systematic Review
The Role of Artificial Intelligence Algorithms in Endometrial Cancer Prediction: A Systematic Review
Reyhane Norouzi Aval,1,*Khalil Kimiafar,2
1. Department of Health Information Technology, School of Paramedical and Rehabilitation Sciences, Mashhad University of Medical Sciences, Mashhad, Iran 2. Department of Health Information Technology, School of Paramedical and Rehabilitation Sciences, Mashhad University of Medical Sciences, Mashhad, Iran
Introduction: Endometrial cancer is among the most common gynecological malignancies, and early detection is crucial for improving patient outcomes. Artificial intelligence (AI), particularly machine learning (ML) algorithms, has shown promise in enhancing diagnostic accuracy and predictive analytics in oncology. This systematic review aims to evaluate the various AI methodologies employed in the prediction and diagnosis of endometrial cancer.
Methods: A comprehensive literature search was conducted across PubMed, Web of Science, and Scopus databases up to February 1, 2025. The search terms included “artificial intelligence,” “machine learning,” “deep learning,” and “endometrial cancer.” Studies were selected based on predefined inclusion criteria focusing on AI applications in endometrial cancer prediction and diagnosis. The PRISMA guidelines were followed, resulting in the inclusion of 35 studies after screening and eligibility assessment.
Results: The analysis revealed that various AI algorithms have been applied in the context of endometrial cancer:
Machine Learning (ML) Techniques: Supervised learning models, such as support vector machines (SVM), random forests (RF), and logistic regression, have been utilized to analyze clinical and histopathological data. These models aim to identify patterns and risk factors associated with endometrial cancer development. Deep Learning (DL) Approaches: Convolutional neural networks (CNNs) have been employed to interpret medical imaging data, including histopathological slides and radiological images. CNNs have demonstrated high accuracy in distinguishing between malignant and benign endometrial tissues. Natural Language Processing (NLP): AI models incorporating NLP have been used to extract relevant information from electronic health records (EHRs), aiding in the identification of potential risk factors and early signs of endometrial cancer.
Conclusion: AI algorithms, encompassing ML, DL, NLP, and hybrid models, have shown significant potential in the prediction and diagnosis of endometrial cancer. These technologies can assist clinicians in early detection and personalized treatment planning. However, further research with larger datasets and standardized methodologies is necessary to validate these findings and facilitate the integration of AI-based models into clinical practice.