In the Fundamentals of Health Informatics course, Bachelor in Bioinformatics (BBI) students explored the modern applications of AI tools in healthcare through an exciting hands-on activity.

💡 Through role-play simulations of doctor–patient consultations, students generated clinical notes and applied AI to unlock healthcare insights:
✅ ICD-10 Coding – Automated coding and verification of patient records using AI tools.
✅ Gene Discovery – Identifying disease-related genes.
✅ Protein Insights – Exploring 3D structures with AlphaFold.
✅ AI Diagnostics – Developing a simple case study model on chest X-rays (normal, COVID-19, pneumonia) and building a basic diagnostic website enhanced with ChatGPT for vibe coding.

🌍 This activity bridges clinical data, genomics, and AI-driven diagnostics, providing students with a holistic understanding of modern AI applications in healthcare while sharpening practical skills for real-world challenges.

This class activity provides a comprehensive, hands-on experience in clinical medicine and data analysis. Students begin by role-playing a doctor-patient consultation to gather real-time clinical notes for a cardiology case. They then transition to the digital realm, using these notes to create a simulated EHR document. The activity bridges traditional and modern methods by having students practice ICD-10 coding using both an AI tool and manual verification with the WHO database. The exercise then delves into the intersection of medicine and genetics, teaching students to convert clinical findings into potential genetic factors using a gene analysis tool. Finally, they use a database like NCBI Gene to research and analyze the top genes, examining their direct association with the patient's disease, thereby connecting a patient's symptoms to their genetic underpinnings and foundational scientific research. 

This advanced activity takes students beyond the basic functions of an Electronic Health Record (EHR) to explore the powerful integration of clinical informatics and bioinformatics. After role-playing a patient case using a simulated EHR, students embarked on a journey to convert clinical data into genomic insights, showcasing the future of precision medicine.

Activity Description: EHR Data to Genomic Analysis

The activity begins with the clinical notes from our simulated patient case, which includes the patient's diagnosis and symptoms. The objective is to analyze the patient's condition at a molecular level by identifying associated genes and potential biomarkers. This process demonstrates a key workflow in modern medical research and diagnostics.

Step 1: Converting Clinical Data

The first step involves transforming the unstructured text from the EHR's clinical notes into structured, machine-readable data. This is a crucial step that bridges the gap between clinical documentation and scientific analysis.

• Medical Coding: Students use an ICD-10 coding tool to convert the patient's diagnosis (e.g., stable angina) into a standardized code. This code serves as a precise query for searching medical and genetic databases.

• Data Extraction: The text is analyzed using methods similar to Natural Language Processing (NLP) to identify key terms, diseases, and genes mentioned in the notes.

Step 2: Bioinformatics and Gene Analysis

With the coded and structured data, students then use bioinformatics tools to analyze the genomic landscape related to the patient's condition.

• Gene Identification: Using the ICD-10 code, students query a genomic database to find a list of genes associated with the diagnosis. This provides a starting point for deeper investigation.

• Gene Function Analysis: The identified genes are then searched in specialized databases (e.g., Gene Ontology, OMIM) to understand their specific biological functions and their known roles in the disease process. This helps to pinpoint which genes are most influential in the patient's condition.

Step 3: Identifying Biomarkers

The final and most advanced part of the activity focuses on identifying biomarkers. A biomarker is a measurable indicator of a biological state or condition. For example, a specific gene mutation or protein level could serve as a biomarker for disease risk, progression, or response to a particular drug.

• Biomarker Discovery: By cross-referencing the patient's diagnosis, associated genes, and genomic data, students can identify potential biomarkers. These could be genes with known mutations linked to the disease or genes that are expressed differently in affected individuals.

• Precision Medicine: This process highlights how integrating EHR data with bioinformatics can lead to personalized treatment plans. By identifying specific biomarkers, doctors could one day select medications that are most likely to be effective for an individual patient, minimizing side effects and improving outcomes.

This activity provides students with a tangible understanding of how medical informatics is evolving, connecting traditional patient care with the cutting-edge fields of genomics and personalized medicine.