2023 was a whirlwind in the tech world… the year that AIs like ChatGPT went mainstream!
There has been a groundbreaking shift in awareness about artificial intelligence (AI) that has both tech enthusiasts and healthcare experts on their toes as they adapt to the wide availability of this technology. Current AI models can’t perfectly replicate what we as humans can do. However, they are still capable enough to be incredibly useful, both in medicine and your life outside of work.
Here we take a quick look at two important aspects of AI that provide key insights into how AI works:
Natural language processing in healthcare and everyday life
Natural language processing is a branch of artificial intelligence that involves teaching computers to understand and interpret human language, including text and speech. This enables them to communicate with us in a more human-like way.
Here are a couple of examples of natural language processing in action:
First, think about all the times you’ve called a customer service number, and were greeted by an artificial voice telling you to “Press or say ‘one’ for technical support, ‘two’ for warranty service…”, et cetera. Those systems use natural language processing to analyse your speech and interpret its meaning.
Next, do you or your colleagues ever dictate texts?
Or have you ever dictated notes from a patient appointment into an electronic medical record?
If you’re using software that transforms your speech into text, that’s AI-powered natural language processing in action. Nowadays there are software applications like Mobius, designed to comply with privacy laws, that are a convenient way to incorporate AI and natural language processing into healthcare practice.
AIs like ChatGPT, Bard, Bing Chat, and Claude are large language models that use the power of natural language processing to draft emails for you, help with the writing of a research paper, or even act as your editor. We will be reviewing these in detail over the next few months and they are discussed in the Essentials for Clinicians course by Medmastery.
Machine learning in healthcare
Machine learning occurs when computers learn and improve from experience, without being specifically programmed to do so. It involves training machines to recognize patterns and make predictions or decisions based on data, which enables them to become more accurate and efficient over time.
Machine learning’s practicality becomes clear in everyday situations like identifying spam emails. Consider this: you’ve learned from experience that emails from people claiming to be a prince, offering you wealth in exchange for your bank details, are spam. Machines can learn this too. After analysing numerous spam email examples, they can identify language patterns unique to such messages. And that’s how AI-powered spam filters are able to stop these deceptive emails from reaching your inbox.
Now let’s turn to some practical examples of machine learning in healthcare.
In radiology, AI can be used with CT to suggest ideal positioning of the patient and scan settings. It can also assist with the analysis of diagnostic images.
And think about all the information present in a patient’s chart or electronic medical record. It’s basically a large collection of data points that need analysis when you’re deciding on a diagnosis and treatment plan. Machine learning excels at handling large datasets, enabling AI to potentially assist in analyzing all that data. That can save time and allow clinicians to arrive at the correct diagnosis or treatment plan faster than what may have been possible before.
The future of AI in healthcare
Artificial intelligence has become an integral part of our daily lives, and is becoming an increasingly important part of the way we provide healthcare. From understanding human speech and transforming it into text to identifying patterns in large datasets, AI is revolutionising the way tasks are approached and executed.
It’s important to recognize that while it doesn’t replace human judgement, it can absolutely act as a capable assistant, streamlining processes and enhancing decision-making. As technology continues to evolve, the potential of AI is likely to expand, leading to a future where human-machine collaboration drives further innovation and efficiency.
- Seneviratne M. Deep Learning – Pushing the boundaries of health AI. How do we make it fair and the data safe? LITFL
- Chandru P. Network Five: Cardiology. LITFL
Research and Review
- Stewart J, Lu J, Goudie A, Bennamoun M, Sprivulis P, Sanfillipo F, Dwivedi G. Applications of machine learning to undifferentiated chest pain in the emergency department: A systematic review. PLoS One. 2021 Aug 24;16(8):e0252612
AI in HEALTHCARE
Want to become a pro at prompting, and consistently get usable results? Be sure to check out Medmastery’s AI prompting course. Learn techniques to apply to the plethora of AI resources in constant development.
BA MA (Oxon) MBChB (Edin) FACEM FFSEM. Associate Professor Curtin Medical School, Curtin University. Emergency physician, Sir Charles Gairdner Hospital. Passion for rugby; medical history; medical education; and asynchronous learning #FOAMed evangelist. Co-founder and CTO of Life in the Fast lane | Eponyms | Books | Twitter |