Instruction:1c2c55e8-b414-4f52-84b9-48f1aef091fb

Modules

AI & Malaria Research

Instructions for:TraineeTrainer

Related Initiative

iRECS

Goal

The aim of this module is to facilitate reflection upon the cross-cutting ethics issues associated with a research proposal to use AI-driven analytics to understand malaria transmission patterns in rural Sub-Saharan Africa.

Learning outcomes

At the end of this module, learners will be able to:

1. Identify and analyse the ethics issues and dilemmas associated with a hypothetical research proposal.
2. Make suggestions for how the ethics issues, including ethics dumping, might be addressed.
3. Identify ethics guidelines and policies that are relevant to the proposed research.

Duration (hours)

1

For whom is this important?

All stakeholders in research

Part of

iRECS

Steps

1Introduction

2Relevant Policies and Guidelines

3Summing Up

4End of Case Study Reflection and Poll

5Module Evaluation

6The Research Proposal and Ethics Approval

7Activity- Live Poll

8A Podcast and Reflection

9Seeking Advice

10Finding Solutions – Activity

11AI Ethics Issues

12Addressing AI Ethics Issues

13Moving Forward

1

Introduction

This project proposes to use AI and machine learning (ML) algorithms to analyse large datasets of malaria transmission in rural regions of Sub-Saharan Africa. Data will be collected on environmental variables, human behaviour, and mosquito population dynamics to build predictive models. The goal is to use AI to identify hotspots of malaria transmission, optimise intervention strategies, and contribute to global malaria eradication efforts.

This is a hypothetical case but draws inspiration from current discussions regarding ethics dumping and the fair and equitable use of AI technologies in research.

As you work through the module, we invite you to consider the ethics issues that are associated with this type of study from a variety of perspectives as well as how they might be addressed by a research ethics committee.

2

The Research Proposal and Ethics Approval

Malaria remains one of the leading causes of morbidity and mortality in Sub-Saharan Africa, particularly in rural regions. Despite significant efforts to curb transmission, the disease persists, with complex transmission patterns that vary based on environmental, social, and biological factors. With the recent rise of artificial intelligence (AI) and big data analytics, new tools are now available to understand and predict disease outbreaks more precisely.

Principal Investigators

Professor Smith (based in the US) and Dr Jones (based in the EU).

Aims

This project proposes to use AI and machine learning (ML) algorithms to analyse large datasets of malaria transmission in rural regions of Sub-Saharan Africa. Data will be collected on environmental variables, human behaviour, and mosquito population dynamics to build predictive models. The goal is to use AI to identify hotspots of malaria transmission, optimise intervention strategies, and contribute to global malaria eradication efforts.

Research Objectives

1. To develop AI-driven predictive models of malaria transmission in rural regions of Sub-Saharan Africa based on environmental, behavioural, and biological factors.
2. To collect and analyse large datasets on malaria incidence, mosquito populations, and environmental variables (e.g., temperature, humidity, rainfall).
3. To provide recommendations to international health organisations on optimal malaria intervention strategies based on AI-predicted transmission patterns.

Research Questions

1. How can AI and machine learning be used to predict malaria transmission patterns in rural Sub-Saharan Africa?
2. What environmental, social, and biological factors are most strongly correlated with malaria transmission in these regions?
3. How can AI models optimise intervention strategies for malaria control in low-resource settings?

Methodology

This study will employ AI and machine learning techniques to analyse large-scale data on malaria transmission. Data will be collected from remote rural areas in Sub-Saharan Africa, specifically focusing on regions in Tanzania, Uganda, and Mozambique, where malaria prevalence is high.

Ethics

Ethics approval is being sought from both Professor Smith’s and Dr Jones’ home institutions in the US and in the EU.

Phase 1: Data Collection

• Environmental Data: Remote sensing and satellite data will be used to gather information on environmental variables such as temperature, rainfall, and vegetation patterns.
• Human Behaviour Data: Survey data on human behaviour related to malaria prevention (e.g., use of bed nets, travel patterns) will be collected through short field visits conducted by external researchers.
• Biological Data: Data on mosquito populations and malaria incidence will be collected through partnerships with local healthcare facilities and mosquito trapping activities.

Phase 2: AI Model Development

• Machine Learning Algorithms: AI algorithms, including neural networks and decision trees, will be used to analyse the datasets and identify the key factors driving malaria transmission.
• Predictive Modelling: AI models will be trained to predict malaria outbreaks based on the environmental, social, and biological data collected.

Phase 3: Reporting and Recommendations

• The AI models will be used to provide recommendations for optimising malaria interventions (e.g., bed net distribution, insecticide spraying) in the study regions.
• Results will be published in high-impact international journals and shared with international health organisations.

Expected Outcomes

1. Publication of Results: The research team expects to publish multiple papers in high-impact journals on the application of AI for malaria control, potentially advancing careers in academia and international research.
2. Recommendations to International Organisations: AI-driven recommendations on malaria interventions will be shared with global health organisations like the WHO and major NGOs working in malaria control, without direct engagement with local policymakers or healthcare systems.

Timeline

• Months 1-3: Initial literature review, AI model development, and planning for data collection trips.
• Months 4-9: Data collection in Tanzania, Uganda, and Mozambique.
• Months 10-15: AI model training, data analysis, and predictive model development.
• Months 16-18: Reporting, writing, and submission of research papers.

Budget

The total estimated budget for this project is $1.2 million, covering:

• Travel expenses for international researchers conducting field visits.
• Equipment for data collection, including satellite imagery access, mosquito traps, and environmental sensors.
• AI development and data analysis software.
• Compensation for local healthcare workers and logistical support in the field.

3

Activity- Live Poll

Question: As a member of a research ethics committee, what is your first impression of this proposal? Do you think that you would be in favour of approval for this project?

4

A Podcast and Reflection

Professor Smith is very excited about this proposal, and looking forward to getting started on the research once the study has ethics approval. She feels that by going for dual approval from both her own and Dr Jones’ institution, the team have covered all necessary requirements for ethics approval. However, she hears that two podcasters have got hold of some information about the project and are raising concerns about ethics dumping issues. Please listen to Brad and Janet’s podcast and see if you agree with them.

What do you think about the points made by Brad and Janet? Do you think that the issues they raised are valid? Do you have any concerns about the proposal that they didn’t mention? Take a few moments to think this through before moving on to the next screen.

5

Seeking Advice

Professor Smith asks Dr Jones if he knows anything about ethics dumping and helicopter research. Dr Jones contacts an old colleague, Dr Langa, to ask for his advice.

Final Thoughts

By aligning the malaria research proposal with the TRUST Code, the project can transition from a potentially extractive model to an inclusive and equitable approach. This will not only safeguard the rights and welfare of the local communities but will strengthen the overall impact and sustainability of the research. Following the TRUST Code will also foster genuine partnerships between African and international researchers, helping to build a foundation for ethical, impactful research that truly addresses the health challenges of local populations.

6

Finding Solutions – Activity

Dr Langa’s advice helps Professor Smith and Dr Jones realise that at the very least they need to address issues around Helicopter research, Benefit Sharing, Knowledge transfer, Informed consent, Data Ownership and Local ethical oversight. Flip the cards to find out ways they might address these issues.

7

AI Ethics Issues

Thus far we have mainly focused on concerns with the ethics dumping issues identified in this proposal. However, it is also important to consider the proposed use of AI technologies in the study and examine the ethics issues that arise from this aspect of the research.

Some AI-related risks have already been touched on in Brad and Janet’s podcast (e.g. explainability, informed consent, accountability), and in Dr Langa’s advice on ethics dumping (e.g. data ownership and access, and capacity building), demonstrating that ethics issues may overlap different domains.

Professor Smith and Dr Jones request a report from their colleague Dr Corry to summarise the AI ethics issues and potential impacts relevant to the proposed study. Watch the video to access Dr Corry’s report.

8

Addressing AI Ethics Issues

Dr Corry’s report helps Professor Smith and Dr Jones identify the primary AI ethics issues that need to be addressed in their study. Flip the cards to find out ways they might address these issues.

9

Moving Forward

Before Professor Smith and Dr Jones proceed any further with this proposal, they realise that they need seek out local partners in each of the three countries and find out:

Is this research wanted or needed by the communities in Mozambique, Uganda and Tanzania?

What research on the prediction of Malaria outbreaks has already been carried out in these communities?

What interventions are already in place?

How successful are the interventions?

Would the proposed research risk undermining local interventions to mitigate the risk of Malaria outbreaks?

Let’s suppose that they get a positive response from a university in one of the countries who would like to collaborate with them and suggest that it would be preferable to start with a pilot of the project.

Of course, as many of the issues cut across both ethics dumping and AI ethics issues, it will be necessary for the research teams to consult ethics guidelines and codes from both angles as they revise their proposal in collaboration with the local partner institution.

10

Relevant Policies and Guidelines

In terms of ethics dumping, the previously mentioned TRUST global code of conduct for equitable research partnerships offers a simple, jargon-free ethics code comprised of 23 articles based around the moral values of Fairness, Respect, Care and Honesty, to help researchers ensure that international research is equitable and carried out without ‘ethics dumping’ or ‘helicopter research’.

In terms of AI ethics, we recommend consulting the Ethics of AI in Healthcare: A checklist for Research Ethics Committees which was developed by irecs colleagues, Alexei Grinbaum and Etienne Aucouturier at CEA (French Alternative Energies and Atomic Energy Commission), as well as the materials in the irecs AI and ethics module.

Chapter 5 of the World Health Organization’s Ethics and Governance of Artificial Intelligence for Health outlines six key ethical principles for AI research in healthcare. These include protecting patient autonomy, promoting human wellbeing, ensuring transparency and explainability, fostering accountability, promoting inclusiveness and equity, and supporting AI that is both responsive and sustainable. These principles serve as essential reminders for researchers and policymakers to prioritise ethical considerations in the development and deployment of AI technologies in healthcare settings.

Another significant issue in the development of AI technologies across all fields is the potential for bias and inaccuracies in algorithms, which in the healthcare domain can result in incorrect diagnoses and treatment recommendations. These risks disproportionately affect vulnerable populations, raising concerns about inclusivity and equity. The EU’s Ethics Guidelines for Trustworthy AI emphasise that AI systems must be lawful, ethical, and robust throughout their Life cycle. This includes compliance with applicable laws, adherence to ethical principles, and ensuring technical and social robustness. Importantly, these guidelines call for mechanisms to prevent algorithmic bias and protect privacy. Unethical applications involving AI are defined as those that risk violating physical or mental integrity, create addiction, risk damaging social processes and public institutions (e.g. by social scoring or contributing to misinformation).

Projects must adhere to essential requirements, which encompass (but are not restricted to):

• People must be made aware that they are interacting with an AI system, its abilities and Limitations, risks and benefits.
• Mechanisms for human oversight, transparency and auditability must be built into the AI system.
• AI-systems must be designed to avoid bias in input data and algorithmic design.
• Compliance with data protection and privacy principles must be demonstrated.

Our hypothetical proposal is not seeking funding from Horizon Europe, however, the EU ethics appraisal scheme (pp74-80), provides relevant guidance for several concerns in this case study. It highlights the importance of transparency, requiring that individuals interacting with AI systems be fully informed about the system’s capabilities, Limitations, risks, and benefits. It also underscores the necessity of building human oversight, transparency, and auditability into AI systems, ensuring that AI development remains accountable and aligned with societal values.

Regulatory oversight has often lagged behind technological advancements, creating additional legal and ethical challenges. The WHO and EU guidelines, among others, stress the need for AI systems to comply with data protection and privacy principles, such as data minimisation, ensuring that only the necessary data is collected and used. This is crucial in building trust and safeguarding against the misuse of sensitive healthcare information.

It is important to remember that different guidelines and regulations will apply to research projects in order to comply with the requirements of different institutions, organisations and geographical locations. Listed in the further resources section are sources to explore on ethics dumping, some of the ethics committees in Africa and the current most relevant EU or international guidelines or standards related to AI in health and healthcare, but you may need to explore further afield to locate those that apply to different situations.

11

Summing Up

This case study explored the ethical complexities of a malaria research proposal using AI to predict disease transmission hotspots in Sub-Saharan Africa. The project, led by researchers from high-income countries, illustrates critical ethical concerns around ethics dumping and AI ethics, particularly when research is conducted in low- and middle-income countries. Key ethics dumping issues included a lack of meaningful involvement of local researchers, limited benefit-sharing with participating communities, and insufficient local control over data ownership.

The AI ethics concerns centered on informed consent complexities, data privacy, and transparency in AI processes which is especially challenging when participants may not fully understand how their data is used by advanced algorithms.

The TRUST Global Code of Conduct principles—fairness, respect, care, and honesty—highlight the need for equitable and respectful research practices. Addressing these values in the proposal would promote ethical research partnerships, prevent exploitative practices, and ensure that the benefits extend to the local communities and health systems involved. This case study underscores the importance of balancing innovative research with robust ethical standards to foster trust, inclusivity, and long-term impact in global health initiatives.

12

End of Case Study Reflection and Poll

Having worked your way through this case study module, how would you feel if a similar proposal was presented to a REC that you were serving on? Do you feel that you would be able to identify where further information or expertise is required, and to locate and apply relevant policies and guidelines? There are many factors to consider, we have touched on several key issues in this case, you might have identified other issues that you would want an ethics committee to take into account in similar situations.

13

Module Evaluation

Thank you for taking this irecs module!

Your feedback is very valuable to us and will help us to improve future training materials.

We would like to ask for your opinions:

1. To improve the irecs e-learning modules

2. For research purposes to evaluate the outcomes of the irecs project

To this end we have developed a short questionnaire, which will take from 5 to 10 minutes to answer.

Your anonymity is guaranteed; you won’t be asked to share identifying information or any sensitive information. Data will be handled and stored securely and will only be used for the purposes detailed above. You can find the questionnaire by clicking on the link below.

This link will take you to a new page: https://forms.office.com/e/UsKC9j09Tx

Thank you!

14

Further Resources on Ethics Dumping

Chatfield, K., Schroeder, D., Guantai, A., Bhatt, K., Bukusi, E., Adhiambo Odhiambo, J., ... & Kimani, J. (2021). Preventing ethics dumping: the challenges for Kenyan research ethics committees. Research Ethics, 17(1), 23-44. Available at: https://journals.sagepub.com/doi/full/10.1177/1747016120925064 (Free to download)

López A, Martins M, Chissico C, et al (2019) OA-87 Research ethics committees in Mozambique: operational and functional characteristics evaluated from a self-assessment tool in 2019 BMJ Global Health 2023;8:A3. https://gh.bmj.com/content/8/Suppl_10/A3.2

Schroeder, D. (2007). Benefit sharing: it’s time for a definition. Journal of medical ethics, 33(4), 205-209.

Schroeder, D., & Pisupati, B. (2010). Ethics, justice and the convention on biological diversity. Available at: https://clok.uclan.ac.uk/9695/1/Ethics,%20Justice%20and%20the%20convention.pdf (Free to download)

Schroeder, D., Cook, J., Hirsch, F., Fenet, S., & Muthuswamy, V. (2018). Ethics dumping: case studies from north-south research collaborations. Springer Nature. Available at: https://link.springer.com/book/10.1007/978-3-319-64731-9 (Free to download)

Schroeder, D., Chatfield, K., Singh, M., Chennells, R., & Herissone-Kelly, P. (2019). Equitable research partnerships: a global code of conduct to counter ethics dumping (p. 122). Springer Nature. Available at: https://link.springer.com/book/10.1007/978-3-030-15745-6 (Free to download)

Schroeder, D., Chatfield, K., Muthuswamy, V., & Kumar, N. K. (2021). Ethics Dumping–How not to do research in resource-poor settings. Journal of Academics Stand Against Poverty, 1(1), 32-55. Available at: https://journalasap.org/index.php/asap/article/view/4 (Free to download)

Wynberg, R., Schroeder, D., & Chennells, R. (2009). Indigenous peoples, consent and benefit sharing: lessons from the San-Hoodia case (Vol. 15). Berlin: Springer.

Research ethics codes

The San Code of Research Ethics, available from: https://www.globalcodeofconduct.org/affiliated-codes/

The TRUST Global Code of Conduct for Equitable Research Partnerships, available from:  https://www.globalcodeofconduct.org/the-code/

Videos

AI and Ethics Dumping: A 25-minute module delivered by Professor Doris Schroeder https://youtu.be/F381tg_upUc?si=FzqE2qOYqpT7TPJa

More videos can be found here: https://www.youtube.com/@trustandprepared1000

Useful websites

The African Society of Human Genetics https://www.afshg.org/

National Commission for Science, Technology and Innovation – (NACOSTI) – Kenya nacosti.go.ke

NACOSTI page on Tanzania -https://nsec.nacosti.go.ke/tanzania/

Tanzania Commission for Science and Technology - https://www.costech.or.tz/

Uganda National Council for Science and Technology (UNCST) Research Committee Accreditation https://www.uncst.go.ug/details.php?option=smenu&id=9&Research%20Ethics%20Committee%20Acreditation.html

The Science for Africa Foundation https://scienceforafrica.foundation/

TRUST (2018) The TRUST Code – A Global Code of Conduct for Equitable Research Partnerships, DOI: https://doi.org/10.48508/GCC/2018.05

Further resources on AI technologies

Alford A, Rathod N (2022) AI could worsen health inequities for UK’s minority ethnic groups - new report 22 February 2022 Imperial News. Imperial College London. Available at: https://www.imperial.ac.uk/news/230413/ai-could-worsen-health-inequitie…;

Colón-Rodríguez CJ (2023) Shedding Light on Healthcare Algorithmic and Artificial Intelligence Bias US. Department of Office of Minority Health. Available at: https://minorityhealth.hhs.gov/news/shedding-Light-healthcare-algorithm…

Morley J, Machado CCV, Burr C, Cowls J, Joshi I, Taddeo M, Floridi L. (2020) The ethics of AI in health care: A mapping review Soc Sci Med. 2020 Sep; 260:113172. doi: 10.1016/j.socscimed.2020.113172.  Epub 2020 Jul 15. PMID: 32702587. Available at: https://pubmed.ncbi.nlm.nih.gov/32702587/ 

Resseguier, Anaïs & Ufert, Fabienne (2024). AI research ethics is in its infancy: the EU’s AI Act can make it a grown-up. Research Ethics 20 (2):143-155. https://philpapers.org/rec/RESARE

Relevant guidelines and policies

It is important to remember that different guidelines and regulations will apply to research projects in order to comply with the requirements of different institutions, organisations and geographical locations. Listed here are the current most relevant EU or international guidelines or standards related to AI in health and healthcare, but you may need to explore further afield to locate those that apply to different situations.

Medical Diagnosis and Artificial Intelligence: Ethical Issues. Joint opinion of the CCNE and CNPEN,

CCNE Opinion 141, CNPEN Opinion 4. November 2022

https://www.ccne-ethique.fr/sites/default/files/2023-05/Opinion%20No.141.pdf

Council of Europe:

Guidelines on artificial intelligence and data protection (2019)

https://ec.europa.eu/info/funding-tenders/opportunities/docs/2021- 2027/horizon/guidance/ethics-by-design-and-ethics-of-use-approaches- for-artificial-intelLigence_he_en.pdf

Deutscher Ethikrat (German Ethics Council):

Opinion: Humans and Machines – Challenges of Artificial Intelligence (2023) https://www.ethikrat.org/en/publications/opinions/humans-and-machines/ (currently only available in German, an English translation will be available in due course)

European Commission:

Ethics By Design and Ethics of Use Approaches for Artificial Intelligence (2021)

https://ec.europa.eu/info/funding-tenders/opportunities/docs/2021-2027/horizon/guidance/ethics-by-design-and-ethics-of-use-approaches-for-artificial-intelligence_he_en.pdf

Ethics guidelines for trustworthy AI (2019)

https://digital-strategy.ec.europa.eu/en/Library/ethics-guideLines- trustworthy-ai

OECD Legal Instruments

Recommendation of the Council on Artificial Intelligence (2019)

https://legaLinstruments.oecd.org/en/instruments/oecd-legal-0449

UNESCO:

Recommendations on the Ethics of Artificial Intelligence (2022)

https://unesdoc.unesco.org/ark:/48223/pf0000381137

World Health Organisation:

Ethics and governance of artificial intelligence for health (2021)

Steps

Other information