Bell Eapen MD, PhD.

Bringing Digital health & Gen AI research to life!

About Me

Translational research in health IT is my passion.

Bell Eapen MD, PhD.

Digital Health & Gen AI Consultant
I am Bellraj (Bell) Eapen, a techie MD (dermatologist) with a PhD in Information Systems from McMaster University. As an R&D lead engineer at Mayo clinic, I specialize in cloud architecture for Gen AI and multimodal machine learning using FHIR.

I maintain several software libraries on GitHub. I facilitate the adoption of Generative AI and FHIR in healthcare organizations. As a consultant, I deliver prototypes & models instead of reports and recommendations!

Checkout my GitHub repo and Contact Me for your next project!

Services

Things that I work on.

Deploy healthcare machine learning pipelines on the cloud

Facilitate ML and AI research in healthcare

Chatbot & Conversational AI for clinical workflows

Design AI applications that fits enterprise architecture

Data warehousing and health data analytics for healthcare (FHIR & OHDSI OMOP)

Customize OSCAR, OpenMRS, DHIS2 and RedCap

Skills

Things that I'm good at.

FHIR
90%
GCP
80%
AI
90%
JAVA
80%
Python
70%
OSCAR EMR
100%
DERMATOLOGY
100%
TensorFlow
85%

Blog

My thoughts.

LLM notations and symbols

Architecting LLM solutions for healthcare – Part II

Healthcare data and applications are complex and require careful design choices. In a previous post, I have outlined some examples and tools for architecting LLM solutions. One challenge of developing LLM applications for healthcare is the complexity and diversity of the architectures involved. LLMs can be used for different purposes, such as information retrieval, text […]

Navigating the Complexities of Gen AI in Medicine: 5 Development Blunders to Avoid

Navigating the Complexities of Gen AI in Medicine: 5 Development Blunders to Avoid

Below, I have listed five critical missteps that you should steer clear of to ensure the successful integration of Gen AI in Medicine. This post is primarily for healthcare professionals managing a software team developing a Gen AI application. Image credit: Nicolas Rougier, GPL via Wikimedia Commons #1 Focus on requirements Gen AI is an […]

Medprompt: How to architect LLM solutions for healthcare.

Medprompt: How to architect LLM solutions for healthcare.

Leveraging the power of advanced machine learning, particularly large language models (LLMs), has increasingly become a transformative element in healthcare and medicine. The applications of LLMs in healthcare are multifaceted, showing immense potential to improve patient outcomes, streamline administrative tasks, and foster medical research and innovation. Architecting LLM solutions in the healthcare domain is challenging […]

Named Entity Recognition using LLMs: a cTakes alternative?

Named Entity Recognition using LLMs: a cTakes alternative?

TLDR: The targeted distillation method described may be useful for creating an LLM-based cTakes alternative for Named Entity Recognition. However, the recipe is not available yet.  Image credit: Wikimedia Named Entity Recognition is essential in clinical documents because it enhances patient safety, supports efficient healthcare workflows, aids in research and analytics, and ensures compliance with […]

Distilling LLMs to small task-specific models

Distilling LLMs to small task-specific models

Deploying large language models (LLMs) can be difficult because they require a lot of memory and computing power to run efficiently. Companies want to create smaller task-specific LLMs that are cheap and easy to deploy. Such small models may even be more interpretable, an important consideration in healthcare. Distilling LLMs refers to the process of […]

Kedro for multimodal machine learning in healthcare 

Kedro for multimodal machine learning in healthcare 

Healthcare data is heterogenous with several types of data like reports, tabular data, and images. Combining multiple modalities of data into a single model can be challenging due to several reasons. One challenge is that the diverse types of data may have different structures, formats, and scales which can make it difficult to integrate them […]