From Clinical Medicine to Machine Learning

Hello everyone, my name is Ivonne, and I am a physician from Colombia. I recently decided to study Machine Learning because I increasingly recognize the potential of AI to transform clinical practice.

In my daily work, I encounter many repetitive tasks that take up a significant portion of the consultation, such as reviewing extensive medical records, integrating information from multiple sources, and documenting clinical reasoning—from the physical examination and interpretation of diagnostic tests to the clinical assessment, diagnosis, and treatment plan. Although these tasks are essential, they often leave less time for in-depth clinical reasoning and meaningful patient interaction.

My goal is to understand how to design AI solutions that automate part of these processes and serve as intelligent assistants for healthcare professionals—not to replace clinical judgment, but to support it in an ethical, safe, and patient-centered way. I believe AI reaches its greatest potential when it complements physicians, allowing us to spend more time on what matters most: our patients.

Has anyone else identified similar opportunities in their own profession since they started studying AI?

That’s a great goal, Ivonne. If you can learn enough to build AI-powered interfaces that genuinely support clinicians, I think they could have a real impact on healthcare.

I already use LLMs extensively as productivity tools in my own work, even without building any custom interfaces. For example, I often use voice input instead of typing because long typing sessions have recently started bothering my shoulder. I let the LLM transcribe and rephrase what I say into clear, well-structured text. I also use it to refine grading feedback for my students and to help me think through algorithmic programming problems.

Even using AI “out of the box,” it has made me much more efficient. You might find it useful to start there as well—experiment with existing tools to streamline your daily workflow. As your machine learning skills grow, you can always move on to designing specialized interfaces tailored to clinical practice. Who knows? You may end up building exactly the kind of assistant you’re envisioning.