Introduction to the 2023 ICMR Guidelines for AI in Biomedical Research and Healthcare

ICMR, the Indian council for medical research- a New Delhi based Apex body released ethical guidelines for application of artificial intelligence (AI) in the second half of March 2023. This being a technical document, is different from other such regulatory technical documents (such as ICH and GVP documents for example) in that it is comparatively much more readable and written with clarity. I am going to write a series on its contents considering how important AI is going to become in the upcoming decades, even lesser time may be and this document tells an Indian government view/take of AI in a very crucial sector such as healthcare (HC) and Biomedical Research (BMR). For the sake of brevity, I have used HC and BMR for the sectors of healthcare and Biomed research henceforth.

This is just an introductory article to give the readers a start into this document, making it much more readable and linear (common regulatory documents problem) and to make one see a bigger picture. Instead of paragraphs, I prefer bullet points and removing all the fluff yet keeping the content adequately covered with improved readability along with retention. In the upcoming second article, I'll write about ICMR thinking in healthcare applications of AI and will cover the ethical considerations of AI in healthcare in more articles.

You may access the ICMR site and download the source document from the link below, as you go through my articles, you can refer to any sections in the PDF you want to delve deeper in.

Without further delay, let's jump right in to the article, shall we??

Background and key facts to know about ICMR:

?ICMR stands for Indian Council of Medical Research

?ICMR is the apex body for conduct of biomedical research funded by Govt of India, responsible for the formulation, coordination and promotion of biomedical research..

?ICMR is at the forefront of setting the standards on ethics in biomedical and health research.

?ICMR head quarter is in New Delhi, India and it has 27 institutes/regional medical research centers.

? It was formed in 1911 when it was known as Indian Research Fund Association (IRFA). In 1949, IRFA was renamed ICMR.

?ICMR encourages registration of clinical trials before the enrollment of the first participant.

?It has been publishing the Indian Journal of Medical Research (IJMR) since 1913.

?These current guidelines discussed in this article were prepared by DHR-ICMR Artificial Intelligence Cell.

?Considering continuous advancements and applications of AI in healthcare, this document is to be periodically reviewed and updated by ICMR.

ICMR Thinking about AI as an advancing technology:

• According to the ICMR, Artificial intelligence(AI) is defined as “a system’s ability to correctly interpret external data and to use those learnings to achieve specific goal and tasks through flexible adaption”.
• AI uses complex computer algorithms to emulate human cognition albeit with far reaching capabilities of analyzing large datasets.

Note: The following write-up is important as it indicates Indian Govt view on emerging AI advances and where they stand and what's the potential they hold:

?According to the ICMR, AI is the most promising emerging technology in the coming decade

?Health/medical sector is one of the sectors most likely to benefit from AI and its advancement

?Artificial Intelligence (AI) and Machine Learning (ML) are certainly the game-changers in the space of healthcare.

? The central goal for such AI?healthcare systems should be to make the AI-assisted platforms available for the benefit of largest section of common people with safety and highest precision possible.

? Human beings are the target and that makes the goal complex and demanding.

But how? Which challenges in healthcare are amenable to be resolved by advances in AI?

1.Lack of medical professionals and infrastructure

2.Rising healthcare costs

3.Difficulties in the implementation of new technology due to the complex healthcare system

The ICMR Guideline document and its structure:

Sections included in ICMR guidance (to be covered in the later articles of this series):

1.Ethical principles for AI in health

2.Guiding principles for stakeholders

3.Ethics review process

4.Governance of AI use for healthcare and research

5.Informed consent process involving human participants and their data.

Scope of ICMR guidelines on AI in healthcare:

1.These guidelines apply to AI based tools created for all biomedical and health research and applications involving human participants and/or their biological data.

2.Considering the far-reaching implications of AI-based technologies in healthcare, these guidelines are applicable to stakeholders who want to utilize health data for biomedical research and healthcare delivery using AI technology and techniques.

3.Both healthcare and AI technologies are rapidly advancing and so shall the associated ethical dimensions.

4.The document therefore shall remain a living document and undergo refinement periodically.

The purpose of these Guidelines:

To provide an ethics framework which can assist in the development, deployment, and adoption of to guide the stakeholders in ethical conduct of research that provides AI solutions in healthcare and will help in identifying and negotiating emerging ethical challenges and concerns.AI-based solutions for biomedical research and healthcare delivery.

Why ethical AI in healthcare is required?

1.To guide the AI technologies development and its application in healthcare.

2.As AI technologies get further developed and applied in clinical decision making, it is important to have processes that discuss accountability in case of errors for safeguarding and protection.

3.AI-based solutions themselves cannot be held accountable for its decisions and judgments.

4.Assignment of accountability and responsibility at all stages of development and deployment of AI for health.

Who are the stakeholders for whom these guidelines are meant?

? Innovators

? Developers

? Patients

? Technologists

? Researchers

? Healthcare professionals

? Ethics committees (ECs)

? Sponsors

? Funding agencies

Potential ethical challenges of AI application in healthcare:

1.Algorithmic transparency

2.Explainability

3.Clarity on liability

4.Accountability

5.Oversight

6.Bias

7.Discrimination

Some of the emerging uses of AI/ML in healthcare/biomedical research:

? Identification of disease

? Screening

? Making diagnosis

? Medical imaging

? Intelligent health operations management

? Personalized medicine

? Digital surveillance for public health

? Outbreak prediction

? Drug discovery

Challenges that need to be tackled As medical profession increasingly applies AI based solutions:

? Data quality

? Data ownership

? Usability

? Trust of the patient-doctor duo on AI software’s effectiveness

? Risk acceptance by the doctor and the patient

? Ethics

Paramount concerns regarding AI in BMR/HC:

? ICMR identifies three paramount concerns regarding AI in healthcare:

1.Patient privacy

2.Confidentiality

3.Ethics

And now, a teaser of the second article in this series on ICMR articles I am writing here on Linkedin and my medium. com page:

Applications of AI in healthcare:

1.Diagnosis

2.Screening

3.Therapeutics

4.Preventive treatments

5.Clinical decision making

6.Public health surveillance

7.Complex data analysis

8.Predicting disease outcomes.

Why it matters and deserves an article of its own?

These applications are still evolving, however, to know about these uses and application in approriate details would help to know where we stand in the practical side of AI at the time of writing, that is April 2023. Additionally, reading about actual applications gets more buy-in and more curiosity about real life AI and not badi badi baatein regarding AI which is going on its own hype cycle and keeps going!

Thanks so much for reading the article, I hope you found it enjoyable and worthy of your time. If you have not already, I urge you to download the ICMR AI guidelines document from the link below and dig deeper in the parts that interest you more. I will definitely come back with more articles based on this document, as understanding Indian point of view (Govt) regarding AI in healthcare and research makes sense from national point of view (as an Indian) as well as from scientific point of view. What's commendable is, compared to other technologies, this guideline document is very advanced in timing.. Usually such guidelines take years to be public, but this also shows the importance and rather urgency govt of India is showing in the developement, deployment and real-life application of AI in a sector so important as healthcare!

Again, thank you for reading. I'd love to know your thoughts about this article, ICMR, this document and AI in healthcare/research and as a whole. You can communicate your thoughts here in comments or send a DM to me on Linkedin. I'd love to connect and discuss!

Have a lovely day :)