Drawing inspiration from the past: AI's potential role in reshaping paediatrics
Keywords:artificial intelligence, paediatrics
In 1852, Dr Charles West established the Great Ormond Street Hospital for Sick Children. Simultaneously, The Times speculated about London drowning in horse manure.
I’m sure 19th-century paediatricians would have counted infectious diseases and poor sanitation, in part caused by horse manure, among their greatest challenges.
London’s prosperity was heavily horse-driven, with over 50,000 horses on its streets daily, delivering bricks to build places like GOSH, transport people and scale agriculture. This was mirrored in every growing city, with each horse producing about 10kg of waste.
Despite the world’s experts coming together in New York for the first international conference in 1898, all that could be agreed was that there wasn’t an alternative and horses would have to stay.
Not long after this, the problem was remarkably resolved. In 1912, the Model T Ford was launched and the motor car took over. Henry Ford said, “if I’d have asked people what they wanted, they’d have said a faster horse.”
The moral of the story is that, often, technological advances provide solutions to our seemingly unsurmountable challenges.
Today, I balance paediatrics with my role as the Chief Clinical Information Officer at Microsoft. From this perspective, I anticipate a revolutionary shift, akin to the automobile’s impact, driven by general purpose AI.
Like the engine, you don’t need to know how it exactly works to benefit from it. This “new” AI, most easily experienced with ChatGPT - a large language model (LLM), is brought to life by using natural language i.e. the way we naturally speak and powered by huge data models (it’s fuel) which run through novel reasoning engines. For those interested in learning more, I recommend this visual explainer from the Financial Times as good place to start.1
As we integrate new technologies into healthcare, anticipating challenges is crucial. I am optimistic that our speciality, entrepreneurial in nature, will lean in. Given that our patients are experiencing this tech revolution first hand, it’s vital that we actively participate in shaping the responsible and safe use of AI.
To illustrate the potential capabilities, here are some examples describing how it can help us and our patients today:
- Engagement: In general paeds, our digitally-native patients, often supported by families, are receptive to innovations. Enhanced chatbots, accessible in multiple languages and backed by verified resources (not the whole internet), can enhance patient understanding about their condition and medications whilst being able to properly safety net. This dynamic interaction surpasses traditional handouts and static websites, moving to personalised communication.
- Time Efficiency: Burnout among clinicians is alarming, a US study found 76% of us feel overwhelmed, mainly due to ever increasing tasks that need to be completed in less time.2 Clinical ambient technology, like Microsoft's DAX Copilot, captures consultations in real-time, auto-generates medical transcripts, and dispatches clinical notes for example as discharge letters. I think this could easily save 30% of our time, which translates to saving hours each shift, and more importantly getting home on time.
These developments are just the beginning. I’d love to hear from you about your experiences, hopes and even fears. Catch up with me on LinkedIn and/or at the RCPCH conference in March.
Financial Times. Generative AI exists because of the transformer. Available from https://ig.ft.com/generative-ai/
NIHCM Foundation. Physician burnout and moral injury. The hidden health care crisis. Available from https://nihcm.org/publications/physician-burnout-suicide-the-hidden-health-care-crisis#:~:text=More%20than%2070%20percent%20of%20physicians%20say%20they,mention%20the%20impact%20it%20has%20on%20their%20patients
How to Cite
Copyright (c) 2023 Umang Patel
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution 4.0 that allows others to share the work with an acknowledgement of the works authorship and initial publication in this journal.