Scholarlink

Purpose <p>Artificial Intelligence (AI) is revolutionizing the prevention and control of breast cancer by improving risk assessment, prevention, and early diagnosis.&#xa0;Considering an emphasis on AI applications across the women’s breast cancer spectrum, this review summarizes developments, existing applications, and future potential prospects.</p> Methods <p>We conducted an in-depth review of the literature on AI applications in breast cancer risk prediction, prevention, and early detection from 2000 to 2025, with particular emphasis on Explainable AI (XAI), deep learning (DL), and machine learning (ML). We examined algorithmic fairness, model transparency, dataset representation, and clinical performance indicators.</p> Findings <p>As compared to traditional methods, AI-based models continuously enhanced risk categorization, screening sensitivity, and early detection (AUCs ranging from 0.65 to 0.975). However, challenges remain in algorithmic bias, underrepresentation of minority populations, and limited external validation. Remarkably, 58% of public datasets focused on mammography, leaving gaps in modalities such as tomosynthesis and histopathology.</p> Conclusions <p>AI technologies have an enormous number of opportunities for enhancing the diagnosis and treatment of breast cancer. However, transparent models, inclusive datasets, and standardized frameworks for explainability and external validation should be given the greatest attention in subsequent studies to ensure equitable and effective implementation.</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

Evolution and integration of artificial intelligence across the cancer continuum in women: advances in risk assessment, prevention, and early detection

  • 期刊论文

  • 发表日期：

    2025年8月20日

查看全文
  • Mitali Desai,
  • Binita Desai

摘要

Purpose

Artificial Intelligence (AI) is revolutionizing the prevention and control of breast cancer by improving risk assessment, prevention, and early diagnosis. Considering an emphasis on AI applications across the women’s breast cancer spectrum, this review summarizes developments, existing applications, and future potential prospects.

Methods

We conducted an in-depth review of the literature on AI applications in breast cancer risk prediction, prevention, and early detection from 2000 to 2025, with particular emphasis on Explainable AI (XAI), deep learning (DL), and machine learning (ML). We examined algorithmic fairness, model transparency, dataset representation, and clinical performance indicators.

Findings

As compared to traditional methods, AI-based models continuously enhanced risk categorization, screening sensitivity, and early detection (AUCs ranging from 0.65 to 0.975). However, challenges remain in algorithmic bias, underrepresentation of minority populations, and limited external validation. Remarkably, 58% of public datasets focused on mammography, leaving gaps in modalities such as tomosynthesis and histopathology.

Conclusions

AI technologies have an enormous number of opportunities for enhancing the diagnosis and treatment of breast cancer. However, transparent models, inclusive datasets, and standardized frameworks for explainability and external validation should be given the greatest attention in subsequent studies to ensure equitable and effective implementation.