Scholarlink

Breast cancer incidence is rapidly rising in sub-Saharan Africa (SSA), with projections indicating a doubling by 2040. Black African women experience the highest mortality rates globally, primarily due to late-stage diagnosis and limited access to diagnostic services. There is an urgent need for innovative, low-cost, and accessible solutions to improve early detection. We integrated community-based participatory research in a prospective, multi-site imaging study conducted in Nigeria and Uganda. The study aims to develop a machine learning (ML) approach for early detection of breast lesions on point-of-care ultrasound (POCUS). Over 1000 women have participated in community outreaches and received at-home breast screening training and breast cancer prevention (nutrition, exercise, etc.) resources. Currently, a total of 407 women from the community events are enrolled in the study and have received POCUS screening onsite. Breast lesions were identified in 31 women (8.6%), of whom 14 (45%) completed follow-up imaging and clinical evaluations. The POCUS images are being preprocessed and prepared for an imaging challenge to enable development of robust state-of-the-art ML solutions. While recruitment is still underway, we outline our approach to inclusion of underserved community in ML methods development. Through this study, we aim to demonstrate the feasibility of POCUS for breast cancer screening in underserved African settings. The resulting dataset—the first of its kind—will facilitate ML-driven screening tools tailored for Black African women, supporting earlier detection and aligning with global targets to decrease breast cancer mortality and morbidity.

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

Bridging the Gap: A Community Driven and AI-Enabled Approach to Early Breast Cancer Detection in Black African Women

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Adaobi Emegoakor,
  • Confidence Raymond,
  • Yewande Gbadamosi,
  • Richard Malumba,
  • Charity Umoren,
  • Diana Spence Betancourt,
  • Aondona M. Iorumbur,
  • Chinasa Kalaiwo,
  • Abbas Rabiu Muhammad,
  • Dennis Musinguzi,
  • Patience Atukunda,
  • Peter Makhoul,
  • Rosta Asiimwe,
  • Alfred Jatho,
  • Amaka Nnamani,
  • Franca Eze,
  • Oluyemisi Toyobo,
  • Abiodun Fatade,
  • Udunna C. Anazodo,
  • Farouk Dako,
  • Michael Kawooya,
  • Maruf Adewole

摘要

Breast cancer incidence is rapidly rising in sub-Saharan Africa (SSA), with projections indicating a doubling by 2040. Black African women experience the highest mortality rates globally, primarily due to late-stage diagnosis and limited access to diagnostic services. There is an urgent need for innovative, low-cost, and accessible solutions to improve early detection. We integrated community-based participatory research in a prospective, multi-site imaging study conducted in Nigeria and Uganda. The study aims to develop a machine learning (ML) approach for early detection of breast lesions on point-of-care ultrasound (POCUS). Over 1000 women have participated in community outreaches and received at-home breast screening training and breast cancer prevention (nutrition, exercise, etc.) resources. Currently, a total of 407 women from the community events are enrolled in the study and have received POCUS screening onsite. Breast lesions were identified in 31 women (8.6%), of whom 14 (45%) completed follow-up imaging and clinical evaluations. The POCUS images are being preprocessed and prepared for an imaging challenge to enable development of robust state-of-the-art ML solutions. While recruitment is still underway, we outline our approach to inclusion of underserved community in ML methods development. Through this study, we aim to demonstrate the feasibility of POCUS for breast cancer screening in underserved African settings. The resulting dataset—the first of its kind—will facilitate ML-driven screening tools tailored for Black African women, supporting earlier detection and aligning with global targets to decrease breast cancer mortality and morbidity.