Scholarlink

<p>Gynecological cancers (GCs) represent an important and evolving field in oncology that requires a thorough understanding of their biology and treatment pathways, warranting ongoing research to ensure appropriate healthcare. On the technological front, in recent years, artificial intelligence (AI) has demonstrated promising capabilities to enhance the accuracy of diagnosis, classification, and prediction across various fields, particularly in healthcare and medicine, including GCs. In light of these data, this review provides a comprehensive overview of AI applications and methods used to diagnose, detect, and classify GCs; identifies research trends over the last decade (2016–2025); determines research methodologies and researchers’ focuses; and identifies gaps in the current literature. The review covered publications retrieved from two major databases, Scopus and Web of Science, using predefined search queries. Based on this, 1,239 research papers were retrieved from the databases, of which 162 studies met the inclusion criteria after duplicates were removed, sequential screening was conducted, and eligibility was filtered. The selected studies were categorized based on the AI techniques used, including machine learning (ML), deep learning (DL), ensemble learning, metaheuristic optimization, and explainable AI (XAI). The results reveal that although cervical cancer has been extensively studied using ML and DL methods, ovarian cancer remains underrepresented, with a notable scarcity of publicly available datasets, especially for optimization and interpretation research. It is impossible to discuss research and identify gaps without addressing the datasets available for research purposes. In addition, the review included 33 benchmark datasets and direct links to them, comprising 15 datasets for cervical cancer, 11 for ovarian cancer, 1 for both, 1 for uterine cancer, 4 for endometrial cancers, and 1 for gynecologic surgical procedures. In addition, this work provides a comparative assessment with recent reviews in the same field to identify research trends, methodological gaps, and unexplored directions. Despite medical advances, there are still challenges facing the treatment of these cancers, most notably the limited possibilities and relatively difficult and expensive methods of obtaining samples and studying women’s different responses to treatment. All of these challenges are obstacles to providing comprehensive healthcare for women, which is why early detection of GCs is the cornerstone of all studies in this field.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial Intelligence Techniques for Diagnosis of Gynecological Malignancies: A Systematic Review

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月4日

查看全文
  • Marwa M. Emam,
  • Doaa S. Ibrahim,
  • Essam H. Houssein

摘要

Gynecological cancers (GCs) represent an important and evolving field in oncology that requires a thorough understanding of their biology and treatment pathways, warranting ongoing research to ensure appropriate healthcare. On the technological front, in recent years, artificial intelligence (AI) has demonstrated promising capabilities to enhance the accuracy of diagnosis, classification, and prediction across various fields, particularly in healthcare and medicine, including GCs. In light of these data, this review provides a comprehensive overview of AI applications and methods used to diagnose, detect, and classify GCs; identifies research trends over the last decade (2016–2025); determines research methodologies and researchers’ focuses; and identifies gaps in the current literature. The review covered publications retrieved from two major databases, Scopus and Web of Science, using predefined search queries. Based on this, 1,239 research papers were retrieved from the databases, of which 162 studies met the inclusion criteria after duplicates were removed, sequential screening was conducted, and eligibility was filtered. The selected studies were categorized based on the AI techniques used, including machine learning (ML), deep learning (DL), ensemble learning, metaheuristic optimization, and explainable AI (XAI). The results reveal that although cervical cancer has been extensively studied using ML and DL methods, ovarian cancer remains underrepresented, with a notable scarcity of publicly available datasets, especially for optimization and interpretation research. It is impossible to discuss research and identify gaps without addressing the datasets available for research purposes. In addition, the review included 33 benchmark datasets and direct links to them, comprising 15 datasets for cervical cancer, 11 for ovarian cancer, 1 for both, 1 for uterine cancer, 4 for endometrial cancers, and 1 for gynecologic surgical procedures. In addition, this work provides a comparative assessment with recent reviews in the same field to identify research trends, methodological gaps, and unexplored directions. Despite medical advances, there are still challenges facing the treatment of these cancers, most notably the limited possibilities and relatively difficult and expensive methods of obtaining samples and studying women’s different responses to treatment. All of these challenges are obstacles to providing comprehensive healthcare for women, which is why early detection of GCs is the cornerstone of all studies in this field.