Part 1: Prompt Engineering
Before diving into the core concepts of Generative AI, it's important to build a
foundational understanding of AI. Here are some resources we will use in class to
get you started:
Part 2: Healthcare Data Analytics using Python
Python was born as a language for data analytics, and its core functionality still
revolves around manipulating data. For researchers, Python can be an invaluable
tool in their arsenal. In this module, we will showcase its potential to work on
clinical data. We will introduce you to some of the powerful applications Python
has for handling and manipulating data, conducting statistical analyses, and
generating graphs and other visuals. We will be leveraging sample clinical data to
power real examples throughout the session.
Data Analytics
- NumPy arrays: Utilize NumPy arrays for efficient data manipulation and mathematical operations.
- Pandas DataFrames: Leverage Pandas DataFrames for data manipulation, cleaning, and analysis
- pywalker and Auto EDA: Employ pywalker and Auto EDA for automated exploratory data analysis (EDA) and reporting.
- Storytelling with Python Packages: Use Python packages for data storytelling, making your insights more engaging and actionable.
Visualizing Data
- Matplotlib: Create static and interactive data visualizations using Matplotlib.
- Seaborn: Enhance your visualizations with Seaborn's statistical data visualization capabilities.
- Plotly: Build interactive and dynamic visualizations with Plotly.
- Streamlit: Develop web applications for data visualization and exploration using Streamlit.
Data Pipeline or RPA (Robotic Process Automation)
- pyautogui: Automate GUI interactions and tasks with pyautogui.
- Selenium: Automate web browser interactions and web testing using Selenium.
- Beautiful Soup: Parse and extract data from HTML and XML documents with Beautiful Soup.
- Web Scraping with LLM (Large Language Models): Use LLMs to enhance web scraping and data extraction capabilities.
Case Studies
- LLMs: Explore Large Language Models and their applications in natural language processing
- ChatGPT 4/3.5: Dive into ChatGPT versions 4 and 3.5, understanding their conversational AI capabilities.
- Palm Med2: Investigate the potential of Palm Med2 in medical data analysis.
- Claud: Explore the features and capabilities of Claud, possibly in the context of data analysis.
- Llama v2: Discover the advancements and applications of Llama v2.
- Falcon: Learn about Falcon and its role in data-related tasks
Data Analysis of a Clinical Trial Dataset
Conduct data analysis on a clinical trial dataset, potentially involving textual and image data.
Part 3: Generative AI for Medicine and Healthcare Professionals
GenAI is transforming the practice of medicine. It’s helping doctors diagnose
patients more accurately, make predictions about patients’ future health, and
recommend better treatments. This module will give you practical experience in
applying GenAI to concrete problems in medicine. You will start by
understanding the fundamental concepts of Generative AI and its various
applications in healthcare. We will work with different Language Model variants
like ChatGPT, Med-Palm2, and more. We will harness langchain strategies to
enhance the performance of Language Models in various healthcare tasks.
