Advancing early breast cancer detection with artificial intelligence in low-resource healthcare systems: a narrative review

INTRODUCTION

Breast cancer continues to be a major global health challenge, with almost 2.3 million new cases diagnosed each year.¹ Early detection is crucial for improving survival rates and alleviating the burden of the disease. However, this is especially difficult in low-resource settings, where access to standard screening methods is limited.² Barriers such as inadequate infrastructure, high costs, and a shortage of skilled healthcare professionals hinder timely diagnosis.^3,4

Advancements in artificial intelligence (AI) have the potential to transform breast cancer screening by addressing various challenges in the field. AI technologies can analyze complex datasets and provide accurate diagnostic insights, making them cost-effective and scalable solutions, especially in low-resource settings.⁵ By enhancing mammography interpretation and powering portable ultrasound devices and smartphone applications, AI has demonstrated significant promise in improving both access to and the efficiency of breast cancer detection.⁶

This review examines the role of AI in breast cancer screening, with an emphasis on its use in low-resource settings. It discusses the challenges associated with traditional screening methods and highlights innovative AI-driven technologies that are transforming early detection efforts. Additionally, the review addresses the obstacles to integrating AI into underserved healthcare systems. Through case studies and success stories, the feasibility of these solutions is illustrated, and a discussion on effective implementation strategies provides a roadmap for scaling AI-powered tools across various settings.

By focusing on the specific needs of low-resource communities, AI has the potential to bridge the gap in breast cancer screening and ensure equitable access to healthcare. This article aims to provide a thorough overview of the current applications of AI in breast cancer screening, as well as highlight future directions for maximizing its potential in global health.

BREAST CANCER SCREENING: CHALLENGES IN LOW-RESOURCE SETTINGS

Limited access to breast cancer screening facilities is a significant issue. Many underserved areas do not have enough mammography units, and the shortage of skilled personnel makes the situation even worse. Geographic barriers, particularly in rural regions, hinder timely access to essential diagnostic services, leaving many women without the chance for early detection. The financial implications of conventional screening methods exacerbate the problem.⁷ Mammography machines, along with their maintenance and the need for trained professionals, lead to high operational costs. These expenses often make traditional breast cancer screening inaccessible for low-income populations, thereby perpetuating disparities in healthcare access.8 Additionally, many governments and healthcare systems in low-resource settings face financial constraints, which hinder the establishment of widespread and sustainable screening programs.⁹

The absence of adequate screening infrastructure leads to delayed diagnoses and poor outcomes. When routine screening is not conducted, breast cancer is often diagnosed at advanced stages, where treatment options are limited and less effective.¹⁰ These late-stage diagnoses significantly contribute to higher mortality rates, highlighting the urgent need for solutions to improve early detection.¹¹ Addressing these challenges requires innovative approaches to overcome financial, geographic, and infrastructural barriers. Integrating cost-effective and scalable technologies, such as artificial intelligence and alternative screening methods, holds promise for improving accessibility and reducing disparities in breast cancer detection.

THE ROLE OF ARTIFICIAL INTELLIGENCE IN BREAST CANCER SCREENING

Artificial intelligence (AI) has become a transformative technology in healthcare, providing innovative solutions to enhance breast cancer screening. By utilizing advanced algorithms and machine learning, AI improves the accuracy, efficiency, and accessibility of screening methods. This is especially valuable in tackling the challenges faced by low-resource settings.¹²

Automated image analysis

One of the most significant contributions of AI is its ability to analyze medical images with remarkable precision. In the context of breast cancer screening, AI-powered algorithms can detect abnormalities in mammograms, ultrasounds, and other imaging modalities with accuracy that is comparable to, or even exceeds, that of experienced radiologists.⁵ Deep learning models, which are trained on large datasets, can identify subtle patterns that indicate the presence of malignancy, thereby reducing both false positives and false negatives. This capability is especially beneficial in low-resource settings, where the shortage of radiologists can hinder timely and accurate interpretation of imaging results.¹³

Expanding screening modalities

AI technologies enhance the effectiveness of affordable and accessible screening methods, such as handheld ultrasound devices and thermography. For example, AI-powered portable ultrasound systems automate the interpretation of images, reducing the dependence on highly skilled operators.¹⁴ Additionally, integrating AI into thermography, which identifies heat patterns linked to cancer, improves its diagnostic accuracy, making it a viable option in places where mammography is not available.¹⁵

Personalized risk assessment

Artificial intelligence (AI) enhances personalized screening strategies by incorporating patient data such as genetic predisposition, medical history, and lifestyle factors. This allows for a more accurate assessment of an individual's risk of developing breast cancer.¹⁵ By adopting this personalized approach, healthcare resources can be allocated more efficiently, prioritizing high-risk individuals for further screening and follow-up care.

Streamlining processes and lowering expenses

AI enhances diagnostic accuracy and streamlines the screening process by automating tasks like image interpretation and report generation. This efficiency reduces the workload for healthcare providers, allowing them to focus on patient care.¹⁶ Furthermore, by improving low-cost screening tools and reducing reliance on expensive infrastructure, AI makes breast cancer screening more affordable and accessible in resource-limited settings.¹⁷

Ensuring fairness in screening processes

AI-driven solutions can improve healthcare access by facilitating remote screening and diagnosis. Smartphone-based AI applications and telemedicine platforms enable women in rural areas to undergo preliminary screenings and receive expert evaluations without the need to travel long distances.^18,19 AI has the potential to revolutionize breast cancer screening, especially in low-resource settings. It can improve accuracy, increase accessibility, and lower costs. By integrating AI into healthcare systems, we can significantly enhance early detection efforts, which could lead to reduced mortality rates and better outcomes for underserved communities.

AI SCREENING TECHNOLOGIES IN RESOURCE-LIMITED ENVIRONMENTS

Artificial intelligence (AI) is transforming breast cancer screening by introducing innovative technologies that are both accessible and cost-effective, making them suitable for low-resource settings.

AI-based mammography

AI algorithms have shown great promise in improving the interpretation of mammograms. Trained on large datasets, these models enhance diagnostic accuracy by identifying subtle abnormalities that human radiologists might overlook.²⁰ For example, Google Health's AI model has outperformed human radiologists in retrospective studies, demonstrating higher accuracy in breast cancer detection.²¹ Similarly, DeepMind's breast cancer detection algorithm has significantly reduced both false positives and false negatives in large-scale trials.²¹ These advancements help streamline the interpretation of mammograms and address the shortage of skilled radiologists in underserved areas.

AI in ultrasound imaging

Ultrasound is often more accessible and affordable than mammography in many low-resource settings, which makes it a practical alternative for screening. AI-powered solutions further improve the effectiveness of this method. For instance, Butterfly Network's handheld ultrasound devices, which are integrated with AI, offer portable and cost-effective options for breast cancer screening.²² These devices are user-friendly, allowing healthcare providers with minimal training to conduct preliminary screenings effectively.

Thermography and AI

Thermal imaging provides a radiation-free screening alternative that, when combined with artificial intelligence (AI), shows promise for detecting breast cancer.²³ AI algorithms analyze the heat maps produced by thermography to identify abnormal patterns that may indicate malignancy.^5,24 This method is especially beneficial in low-resource settings due to its non-invasive nature and minimal infrastructure requirements.

Smartphone-based applications

AI-driven mobile applications are empowering individuals to take an active role in breast health. These applications guide women through breast self-examinations, provide risk assessments, and recommend seeking medical attention when abnormalities are detected.²⁵ With the widespread use of smartphones, these tools have the potential to bridge gaps in healthcare access by facilitating early detection and encouraging timely interventions.²⁶ By utilizing cost-effective, portable, and efficient solutions such as AI-enhanced mammography, ultrasound, thermography, and mobile applications, these innovations are enhancing access to early detection and potentially lowering breast cancer mortality in underserved areas.

CASE STUDIES AND SUCCESS STORIES

Integrating artificial intelligence (AI) into breast cancer screening has produced promising outcomes in low-resource settings. Various successful initiatives demonstrate how AI technologies are transforming healthcare delivery and closing gaps in early detection services.

RAD-AID international

RAD-AID International is leading the way in implementing AI-powered radiology solutions in resource-limited countries. By incorporating AI tools into existing healthcare systems, RAD-AID has shown that even underserved regions can access advanced diagnostic technologies. Their projects have enhanced access to accurate and timely breast cancer screenings, addressing the challenges posed by shortages of trained radiologists and imaging infrastructure.^27,28

Tata memorial centre, India

In India, the Tata Memorial Centre has investigated the potential of AI-based software for analyzing thermographic images as a tool for breast cancer screening. Thermography, which is a non-invasive and radiation-free method, becomes even more effective in low-resource settings when combined with AI algorithms that interpret heat patterns associated with cancer.²⁹ The study emphasized that AI-enhanced thermography could provide an affordable and accessible screening alternative, especially in regions where mammography is either unavailable or too expensive.³⁰

Rwanda’s AI-driven ultrasound project

Rwanda has successfully piloted the use of AI-enabled portable ultrasound devices to enhance breast cancer screening in rural areas. These portable and cost-effective devices have significantly improved access to early detection services for women in underserved communities.³¹ The integration of AI algorithms simplifies image interpretation, enabling healthcare workers with limited training to conduct screenings effectively. This initiative has not only increased screening coverage but also showcased the potential of scalable AI solutions to address disparities in healthcare access.³²

These case studies highlight the significant role of AI technologies in breast cancer screening, especially in low-resource settings. By utilizing innovative approaches, these initiatives offer a model for enhancing early detection and decreasing breast cancer-related mortality globally.

CHALLENGES IN IMPLEMENTING AI-BASED SCREENING

Implementing AI-based technologies for breast cancer screening has great potential, yet significant challenges in low-resource settings must be addressed. These challenges include technical, infrastructural, economic, and ethical factors, that hinder the widespread adoption and effectiveness of AI-driven solutions.

Data availability and quality

AI algorithms need large, high-quality datasets for training and validation to deliver accurate results. However, low-resource settings often struggle to access diverse and representative medical datasets due to poor digital record-keeping and limited diagnostic infrastructure. The lack of locally relevant data can result in biased algorithms that may not perform well in specific populations, ultimately compromising diagnostic accuracy and reliability.³³

Infrastructural barriers

AI-based technologies depend on digital infrastructure, which includes dependable electricity, internet connectivity, and compatible devices. In many low-resource areas, this necessary technological infrastructure is lacking, making it difficult to deploy and operate AI tools. Moreover, in regions with limited technical expertise, maintaining and repairing advanced equipment, such as AI-integrated devices, can pose significant challenges.^9,34

Cost and sustainability

The initial investment in AI-powered screening tools, along with the costs of training personnel, software licensing, and maintaining devices, presents significant barriers. Although AI technologies have the potential to lower long-term expenses, the upfront financial requirements can be prohibitive for healthcare systems in low-resource settings. Additionally, securing ongoing funding for these programs remains a critical challenge.^12,35

Resistance to adoption

Healthcare professionals and policymakers may hesitate to adopt AI-based technologies because of unfamiliarity, concerns about reliability, and fears of job displacement. To build trust in AI tools, it is essential to provide extensive education and training, along with compelling evidence demonstrating their effectiveness in real-world settings.^19,36

Ethical and regulatory concerns

The use of AI in healthcare raises ethical issues related to data privacy, informed consent, and the transparency of algorithms. In low-resource settings, the lack of strong legal and regulatory frameworks can jeopardize patient confidentiality and equitable access to AI-assisted care.³⁷

To address these concerns, collaborative efforts among governments, NGOs, and technology developers are essential. This collaboration will help ensure that AI-based breast cancer screening tools are accessible, affordable, and tailored to meet the unique needs of low-resource environments.

STRATEGIES FOR EFFECTIVE AI INTEGRATION

Effectively integrating AI-based breast cancer screening technologies in low-resource settings necessitates a comprehensive approach that addresses technological, infrastructural, and socioeconomic challenges. By implementing these strategies, we can enhance accessibility, affordability, and effectiveness, ultimately leading to improved early detection outcomes.

Creating strong and localized datasets

Creating AI algorithms that are specifically designed for local populations is crucial for obtaining accurate and reliable results. This process requires the development of large, high-quality datasets that accurately reflect the demographic, genetic, and clinical characteristics of the region.³⁸ Collaborating with local healthcare institutions and research organizations can help in gathering this data while upholding ethical standards, including patient privacy and consent.

Strengthening digital infrastructure

Investing in essential infrastructure is crucial for the adoption of AI. Governments and non-governmental organizations can work together to enhance internet connectivity, electricity supply, and access to digital tools.³⁹ Additionally, deploying portable, AI-powered devices that require minimal infrastructure can help address logistical challenges in remote areas.

Cost reduction and financial assistance

To ensure the long-term sustainability of artificial intelligence technologies, it is essential to minimize initial costs. Public-private partnerships can play a key role by subsidizing AI tools, providing affordable software licenses, and funding training programs for healthcare workers. Promoting open-source AI platforms further reduces expenses and encourages broader adoption, ultimately making advanced technologies more accessible to various sectors.⁴⁰

Training and capacity building

Healthcare providers in low-resource settings need specialized training to use AI technologies effectively. Training programs should emphasize understanding AI outputs, operating AI-integrated devices, and troubleshooting technical issues.¹⁶ By empowering local healthcare workers, we can ensure that technology is used efficiently and that there is increased confidence in AI-driven solutions.⁴¹

Policy and regulatory frameworks

Establishing clear and comprehensive policies is vital for the ethical use of artificial intelligence. These policies should include strong guidelines for data privacy to protect personal information and promote algorithm transparency, allowing stakeholders to understand decision-making processes and identify biases. Ensuring equitable access to AI technologies, especially for marginalized and low-resource communities, is also crucial.⁴² Governments should collaborate with technology developers to create tailored standards that address these challenges, fostering a responsible AI environment that safeguards individual rights and interests.³³

Community involvement and awareness

Establishing trust in AI technologies is a multifaceted process that involves educating communities about the numerous benefits these innovations can offer while also addressing common concerns related to automation and the reliability of such systems. To effectively foster acceptance, it is essential to launch comprehensive awareness campaigns that engage local leaders, healthcare providers, and community influencers.⁴³ By doing so, these campaigns can highlight the positive impacts of AI on healthcare-such as improved screening accuracy and access to timely medical interventions-ultimately encouraging more individuals to participate in AI-driven screening programs. Together, these efforts can empower communities to embrace the advantages of AI, transforming apprehension into confidence.⁴⁴

By employing these strategies, AI technologies can be effectively integrated, closing gaps in breast cancer screening and enhancing healthcare outcomes in underserved areas.

FUTURE DIRECTIONS

The future of AI-driven breast cancer screening in low-resource settings depends on the development of technologies that are cost-effective, portable, and tailored to the specific needs of local healthcare systems. Innovations such as smartphone-based screening tools, open-source AI platforms, and cloud-based solutions can significantly improve accessibility and scalability. Collaborative efforts among governments, technology developers, and healthcare organizations will ensure ethical integration, sustainable funding, and broad adoption of these technologies.⁴⁵ Moreover, incorporating AI into comprehensive healthcare models that emphasize education, prevention, and treatment will enhance its overall impact. Ongoing research and pilot programs in various settings will help refine AI applications, paving the way for equitable and effective breast cancer screening worldwide.

CONCLUSION

AI technologies hold significant promise for improving breast cancer screening in low-resource settings. By addressing key issues such as accessibility, affordability, and workforce shortages, AI can play a vital role in reducing global disparities in breast cancer outcomes. However, to fully realize this potential, it is essential to focus on overcoming implementation challenges, ensuring ethical use, and customizing solutions to meet the specific needs of these environments. With ongoing innovation and collaboration, AI can contribute to a future where early detection of breast cancer is available to everyone, regardless of their socioeconomic status.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: Not required

REFERENCES

1. 1. Wilkinson L, Gathani T. Understanding breast cancer as a global health concern. Br J Radiol. 2022;95(1130):20211033.
2. 2. Ong SK, Haruyama R, Yip CH, et al. Feasibility of monitoring Global Breast Cancer Initiative Framework key performance indicators in 21 Asian National Cancer Centers Alliance member countries. EClinicalMedicine. 2023;67:102365.
3. 3. Ginsburg O, Yip CH, Brooks A, et al. Breast cancer early detection: A phased approach to implementation. Cancer. 2020;126(10):2379-93.
4. 4. Anderson BO, Jakesz R. Breast cancer issues in developing countries: an overview of the Breast Health Global Initiative. World J Surg. 2008;32(12):2578-85.
5. 5. Ahn JS, Shin S, Yang SA, et al. Artificial Intelligence in Breast Cancer Diagnosis and Personalized Medicine. J Breast Cancer. 2023;26(5):405-35.
6. 6. Uchikov P, Khalid U, Dedaj-Salad GH, et al. Artificial Intelligence in Breast Cancer Diagnosis and Treatment: Advances in Imaging, Pathology, and Personalized Care. Life (Basel). 2024;14(11):1451.
7. 7. Ley P, Hong C, Varughese J, Camp L, Bouy S, Maling E. Challenges in the Management of Breast Cancer in a Low Resource Setting in South East Asia. Asian Pac J Cancer Prev. 2016;17(7):3459-63.
8. 8. Galukande M, Kiguli-Malwadde E. Rethinking breast cancer screening strategies in resource-limited settings. Afr Health Sci. 2010;10(1):89-92.
9. 9. Pandey S, Chandravati. Breast screening in north India: a cost-effective cancer prevention strategy. Asian Pac J Cancer Prev. 2013;14(2):853-7.
10. 10. Mahumud RA, Gow J, Keramat SA. Distribution and predictors associated with the use of breast cancer screening services among women in 14 low-resource countries. BMC Public Health. 2020;20(1):1467.
11. 11. Asadzadeh VF, Broeders MJ, Kiemeney LA, Verbeek AL. Opportunity for breast cancer screening in limited resource countries: a literature review and implications for Iran. Asian Pac J Cancer Prev. 2011;12(10):2467-75.
12. 12. Yin J, Ngiam KY, Teo HH. Role of Artificial Intelligence Applications in Real-Life Clinical Practice: Systematic Review. J Med Internet Res. 2021;23(4):25759.
13. 13. Uwimana A, Gnecco G, Riccaboni M. Artificial intelligence for breast cancer detection and its health technology assessment: A scoping review. Comput Biol Med. 2025;184:109391.
14. 14. Taylor CR, Monga N, Johnson C, Hawley JR, Patel M. Artificial Intelligence Applications in Breast Imaging: Current Status and Future Directions. Diagnostics (Basel). 2023;13(12):2041.
15. 15. Sacca L, Lobaina D, Burgoa S. Promoting Artificial Intelligence for Global Breast Cancer Risk Prediction and Screening in Adult Women: A Scoping Review. J Clin Med. 2024;13(9):2525.
16. 16. Maleki Varnosfaderani S, Forouzanfar M. The Role of AI in Hospitals and Clinics: Transforming Healthcare in the 21st Century. Bioengineering (Basel). 2024;11(4):337.
17. 17. Bhandari A. Revolutionizing Radiology With Artificial Intelligence. Cureus. 2024;16(10):72646.
18. 18. Iqbal J, Cortés Jaimes DC, Makineni P. Reimagining healthcare: unleashing the power of artificial intelligence in medicine. Cureus. 2023;15(9):44658.
19. 19. Ezeamii VC, Okobi OE, Wambai-Sani H. Revolutionizing Healthcare: How Telemedicine Is Improving Patient Outcomes and Expanding Access to Care. Cureus. 2024;16(7):63881.
20. 20. Jiang Z, Gandomkar Z, Trieu PDY, Taba ST, Barron ML, Lewis SJ. AI for interpreting screening mammograms: implications for missed cancer in double reading practices and challenging-to-locate lesions. Sci Rep. 2024;14(1):11893.
21. 21. Reuter E. Google’s AI beats humans at detecting breast cancer, sometimes. MedCity. Avaialble at: https://medcitynews.com. Accessed on 21 September 2024.
22. 22. Thawkar S, Sharma S, Khanna M, Singh LK. Breast cancer prediction using a hybrid method based on Butterfly Optimization Algorithm and Ant Lion Optimizer. Comput Biol Med. 2021;139:104968.
23. 23. Rakhunde MB, Gotarkar S, Choudhari SG. Thermography as a breast cancer screening technique: a review article. Cureus. 2022;14(11):31251.
24. 24. Wang X, Chou K, Zhang G. Breast cancer pre-clinical screening using infrared thermography and artificial intelligence: a prospective, multicentre, diagnostic accuracy cohort study. Int J Surg. 2023;109(10):3021-31.
25. 25. Zhang J, Wu J, Zhou XS, Shi F, Shen D. Recent advancements in artificial intelligence for breast cancer: Image augmentation, segmentation, diagnosis, and prognosis approaches. Semin Cancer Biol. 2023;96:11-25.
26. 26. Indah N, Nilawati Usman A, Sanusi Baso Y, Syarif S, Ahmad M, Agus Mumang A. Early detection of self-breast examination using smartphone breast application. Breast Dis. 2024;43(1):135-44.
27. 27. Lugossy AM, Anton K, Dako F. Building radiology equity: themes from the 2023 RAD-AID conference on international radiology and global health. J Am Coll Radiol. 2024;21(8):1194-200.
28. 28. Mollura DJ, Culp MP, Pollack E. Artificial intelligence in low- and middle-income countries: innovating global health radiology. Radiology. 2020;297(3):513-20.
29. 29. Low-cost AI tools aim to predict Indian population cancer risk and therapeutic benefit, Winship Cancer Institute of Emory University. 2023. Available at: https://winshipcancer.emory.edu. Accessed on 22 December 2024.
30. 30. Mahajan A, Bothra M. Mining artificial intelligence in oncology: Tata memorial hospital journey. Cancer Res, Stat, and Treat. 2020;3(3):622.
31. 31. AI Helpful in Triaging Breast Masses in Low-Resource Areas. 2017. Available at: Accessed https://www.rsna.org. Accessed on 22 December 2024.
32. 32. Akingbola A, Adegbesan A, Ojo O, Otumara JU, Alao UH. Artificial intelligence and cancer care in Africa. J Med Surg and Pub Heal. 2024;3:100132-3.
33. 33. Ahmed MI, Spooner B, Isherwood J, Lane M, Orrock E, Dennison A. A systematic review of the barriers to the implementation of artificial intelligence in healthcare. Cureus. 2023;15(10):46454.
34. 34. Pashkov VM, Harkusha AO, Harkusha YO. Artificial intelligence in medical practice: regulative issues and perspectives. Wiad Lek. 2020;73(2):2722-7.
35. 35. Khanam M, Akther S, Mizan I. The Potential of Artificial Intelligence in Unveiling Healthcare's Future. Cureus. 2024;16(10):71625.
36. 36. Lei F. The application of artificial intelligence in lung cancer research. Cancer Control. 2024;31:10732748241297373.
37. 37. Jeyaraman M, Balaji S, Jeyaraman N, Yadav S. Unraveling the Ethical Enigma: Artificial Intelligence in Healthcare. Cureus. 2023;15(8):43262.
38. 38. Esmaeilzadeh P. Challenges and strategies for wide-scale artificial intelligence (AI) deployment in healthcare practices: A perspective for healthcare organizations. Artif Intell Med. 2024;151:102861.
39. 39. Reddy S. Generative AI in healthcare: an implementation science informed translational path on application, integration and governance. Implement Sci. 2024;19(1):27.
40. 40. Shuaib A. Transforming Healthcare with AI: Promises, Pitfalls, and Pathways Forward. Int J Gen Med. 2024;17:1765-71.
41. 41. Uygun Ilikhan S, Özer M, Tanberkan H, Bozkurt V. How to mitigate the risks of deployment of artificial intelligence in medicine. Turk J Med Sci. 2024;54(3):483-92.
42. 42. Al-Roomi K, Alzayani S, Almarabheh A. Familiarity and applications of artificial intelligence in health professions education: perspectives of students in a community-oriented medical school. Cureus. 2024;16(11):73425.
43. 43. Alowais SA, Alghamdi SS, Alsuhebany N. Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC Med Educ. 2023;23(1):689.
44. 44. Alowais SA, Alghamdi SS, Alsuhebany N. Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC Med Educ. 2023;23(1):689.
45. 45. Zheng D, He X, Jing J. Overview of Artificial Intelligence in Breast Cancer Medical Imaging. J Clin Med. 2023;12(2):419.