Breast cancer is the leading cause of cancer related deaths in women globally and in South Africa. 2 261 419 Women were diagnosed with breast cancer in 2020 (Dlamini, et al, 2024). The key to reducing breast cancer related deaths is early detection and treatment. South Africa faces a severe challenge due to constrained resources in the public health sector, which serves 71% of the South African population. There are also significant disparities both ethnically and socioeconomically in the screening, treatment, and survival for breast cancer (Dlamini et al, 2024). South Africa is one of the most unequal countries in the world, it is paramount to use innovation to bridge the gap of inequality on all fronts. Artificial Intelligence (AI) is developing at an exceptional rate and expanding into various spaces. It has shown revolutionised health care by improving efficiency, accuracy, and access to populations at large. In the case of breast cancer screening and diagnosis, AI is being used to determine the risk, evaluate prognostication, and support clinicians with decision-making regarding treatment and management planning.
“Prevention is better than cure. This is what cancer screening tests are about. Used to identify and eliminate common cancers or precancerous conditions early on, so that more advanced cancers can be prevented, these tests can literally save your life “ .
– Discovery
In a South African context, this innovation can help address health inequality particularly disparities in screening and treatment. However, this can only be achieved if it is applied in public sector. Public sector serves 71% of the South African population and is funded by the State. Public health care facilities in South Africa often grapple with constrained resources, significantly impacting access to quality care for many citizens. These limitations manifest as shortages in human resources, essential medications, and critical equipment. Additionally, concerns around waste management and infrastructure maintenance further exacerbate the situation.
This lack of resources disproportionately affects individuals from lower socioeconomic backgrounds who rely on these facilities due to limited financial means to access health care from private facilities. This creates a situation of limited distributive justice, where access to essential health care services is not equitably distributed amongst the population. This highlights the urgency for interventions that address these disparities and ensure that vulnerable populations have access to the quality care they deserve. The potential of AI in breast cancer imaging to improve patient outcomes through earlier diagnoses, personalised treatment plans, and ultimately, a reduction in breast cancer mortality rates is significant.
However, the technology’s current development and prevalence within private organisations raises concerns about affordability and equitable access, particularly in resource-constrained settings like South Africa. The cost of procuring such technology will most likely be significant, which will be a barrier for State funded facilities.
“The greatest opportunity offered by AI is not reducing errors or workloads, or even curing cancer: it is the opportunity to restore the precious and time-honoured connection and trust – the human touch – between patients and doctors”
– Eric. J. Topol
While private facilities catering to a limited portion of the population, approximately 29%, may have the financial means to acquire this technology, a utilitarian perspective compels us to consider the potential for maximising overall benefit. In this case, the ethical principle of utilitarianism argues for prioritising broader accessibility to ensure the technology serves the greater good of the South African population. While private facilities may possess the financial means to acquire this technology, its true potential for good lies in serving the broader population. The ability to prevent deaths and improve countless lives through early detection far outweighs the benefits of a technology confined to the privileged few.
Should the technology be introduced in a South African context, and hopefully it will be, the conversation should be geared towards using this technology in the public sector to maximise its use. The technology promises accurate breast cancer image analysis with limited human assistance. This can be used in rural areas where there are no radiologists on site or to assist facilities burdened with large number caseloads and limited health care workers on site. The deployment of AI technology in public health facilities transcends mere economic considerations. It embodies the core principle of health care as a human right, enshrined in the South African constitution. Limiting access solely to those with financial means creates a stark ethical challenge, exacerbating existing health care disparities within a country already grappling with significant socioeconomic inequalities. By ensuring equitable access, we can harness the power of AI to create a more just and effective breast cancer screening system for all South Africans.
Furthermore, this necessitates exploring strategies to make AI-powered screening financially viable within the public health care system. This could involve public-private partnerships, exploring cost-effective implementation models, and potentially leveraging international collaborations to make this life-saving technology more readily available to all. While the potential of AI for breast cancer screening is undeniable, we must acknowledge the ethical challenges it presents.
One of the most concerning issues is bias. AI algorithms are trained on vast datasets, and if these datasets lack sufficient representation of African populations, it can lead to biased decision-making and poor clinical outcomes. This has significant implications for accuracy. AI trained primarily on European or Western data may struggle to interpret mammograms or ultrasounds from individuals with different skin tones or breast tissue densities. Inaccurate readings could lead to missed diagnoses or unnecessary biopsies, posing a real health risk.
It is paramount to advocate for responsible development and implementation of AI in health care. This requires inclusive dataset to ensure the technology can be applicable to all individuals across the globe.
Written by: Nomfundo Maseko
References:
Dlamini, Z., Molefi, T., Khanyile, R., Mkhabele, M., Damane, B., Kokoua, A., Bida, M., Saini, K.S., Chauke-Malinga, N., Luvhengo, T.E. and Hull, R., 2023. From Incidence to Intervention: A Comprehensive Look at Breast Cancer in South Africa. Oncology and Therapy, pp.1-11.
Image: https://www.biomedcentral.com/collections/spot-breast-cancer
