Abstract
Breast cancer stands as the predominant malignancy among women globally, with a particularly pronounced impact in low- and middle-income countries (LMICs) like South Africa (SA). Given the significant burden of this disease, the imperative for advancing detection technologies is evident. Aim and Objectives: This study endeavours to establish a point-of-care risk-stratified breast cancer screening program in the Gauteng region by comparing the efficacy of clinical breast examination (CBE) and artificial intelligence (AI) breast ultrasound screening techniques, alongside assessing incidence rates. The first objective involves prospectively evaluating breast cancer incidence rates using CBE exclusively at urban clinics in Gauteng for risk-stratified screening. The second objective extends this assessment by incorporating AI breast ultrasound software alongside CBE at similar urban clinics. Lastly, the study aims to compare patient referral pathways and breast cancer incidence rates between CBE and AI methods. Methodology: A quantitative comparative descriptive study spanning a 6-month period is employed, encompassing women aged 25 to 85, with or without a clinical history of breast cancer or related symptoms. Current Results: From July 31st to February 28th, 2024, 530 patients underwent screening at Daspoort Policlinic. Concurrently, 1200 women received breast cancer self-examination education at Melusi settlement, supported by weekly transportation to Daspoort Poli Clinic, ensuring comprehensive patient referral pathways. Among these women, 530 presented with concerning signs, leading to eight referrals for short-term follow-up due to BIRADS-2-3 findings, reaffirming the study's assertion regarding the limited need for immediate surgical intervention. Notably, the patient referral pathway blueprint™ has expedited clinic appointments, facilitating timely interventions. Conclusion: The statistical outcomes of this research offer promising implications for clinical downstaging and rural screening efforts. Compared to recent studies, the developed diagnostic algorithm demonstrates superior efficacy, underscoring its validity and accuracy in clinical application.