Wuraola Oyewusi

Machine learning skills you'll gain:

• ML fundamentals: supervised vs. unsupervised learning, features, labels, and model evaluation
• Classification: heart failure outcome prediction with real clinical data
• Regression: predicting heart ejection fraction from patient features
• Feature importance analysis and feature scaling techniques
• Unsupervised learning: clustering (k-means) and dimensionality reduction
• Deep learning overview, transfer learning, and pretrained models for healthcare
• Data privacy, ethics, and career pathways in healthcare ML

Tools covered: Google Colab, scikit-learn, TensorFlow, Python for healthcare ML, model evaluation techniques

Perfect for: Healthcare professionals and data scientists entering the healthcare AI field.

Real-world healthcare projects you'll complete:

• Outpatient clinic analysis: wait times, demographics, cost patterns, and correlations
• Medication demand forecasting with time-series analysis using Prophet
• Patient experience analytics with sentiment analysis (TextBlob)
• Geospatial health facility mapping and buffer analysis with GeoPandas and folium
• Interactive healthcare dashboards and visualizations with Plotly
• Advanced data wrangling and cleaning for healthcare datasets

Tools covered: pandas, GeoPandas, Plotly, TextBlob, Prophet, folium, statistical analysis, forecasting

Perfect for: Healthcare analysts and data professionals working with clinical datasets.

Productivity tools you'll master:

• Google Gemini: writing, data exploration, image description, and text extraction from images
• Microsoft Copilot: image generation, design creation, and brand kit development with Designer
• Prompt engineering fundamentals and best practices for GenAI tools

Academic research tools you'll master:

• Elicit AI: scientific literature search and research automation
• Litmaps: seed map generation, literature discovery, and visual citation networks
• SciSpace: literature review, PDF parsing with Copilot, citation generation, and scholarly paraphrasing

Perfect for: Academic researchers, graduate students, professionals, healthcare workers, educators, and knowledge workers across all sectors.

AI tools for healthcare administration:

• ChatGPT: automating patient communication, drafting policies, and streamlining workflows
• Google Gemini: creating patient education materials and analyzing healthcare data
• Microsoft Copilot: enhancing documentation, email management, and scheduling efficiency
• Claude: generating professional correspondence and summarizing medical information

Administrative workflows you'll optimize:

• Grant writing and funding proposals for healthcare programs
• Patient education materials: creating clear, accessible health information
• Scheduling optimization and appointment management automation
• Email drafting, responses, and professional communication
• Policy drafting, procedure documentation, and compliance materials
• Meeting summaries and administrative report generation

Prompt engineering for healthcare tasks:

• Crafting effective prompts for healthcare-specific AI outputs
• Ensuring HIPAA compliance and patient privacy in AI workflows
• Refining AI-generated content for medical accuracy and professionalism
• Customizing AI tools for clinic, hospital, and healthcare organization needs

Perfect for: Healthcare administrators, clinic managers, medical office staff, hospital administrators, healthcare operations professionals.

Healthcare Excel projects you'll complete:

• Patient data management: biodata forms, care coordination, demographic analysis with PivotTables
• Healthcare operations: nursing staff scheduling, lab sample tracking, Gantt chart project timelines
• Clinical applications: medication dosage calculators, symptom trackers, discharge planning checklists
• Analytics dashboards: wait time analysis, patient satisfaction surveys, disease outbreak monitoring
• Supply chain management: inventory tracking, stock forecasting for vaccination centers, dynamic pricing

Excel skills covered: PivotTables, data validation, conditional formatting, formulas (AVERAGE, MEDIAN), charts, dashboards

Perfect for: Nurses, administrators, clinical staff, and healthcare managers.

Python fundamentals you'll master:

• Core programming concepts: data types, structures (lists, dictionaries, tuples, sets)
• Python operations: arithmetic, comparisons, logical operators, and string manipulation
• Functions, conditional statements, loops, and control flow for healthcare tasks
• File handling: reading, writing, and processing healthcare data files
• Built-in functions and Python libraries for data analysis workflows
• Code documentation, error handling, and debugging techniques

Healthcare examples: Grouping medications by pharmacology class, managing dosage strengths, patient data organization

Perfect for: Nurses, physicians, pharmacists, and healthcare administrators with zero prior coding experience.

Deep learning applications you'll build:

• X-ray image classification: normal vs. pneumonia detection with CNNs
• Medical image object detection for diagnostic support
• Zero-shot learning approaches for healthcare applications
• Transfer learning with pre-trained architectures (ResNet)

Neural network architectures covered: Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Transformer models

Tools: TensorFlow, PyTorch, Keras, computer vision for medical imaging

Portfolio projects you'll complete:

• Disease prediction: Diabetes ML model with SHAP explainability
• Sentiment analysis: Online medication reviews using transformer models
• Medical image classification: Shoulder implant X-ray detection with CNNs and transfer learning
• Text visualization: Word clouds and Scattertext for disease and medication data

Tools: Machine learning models, SHAP, transformer models, CNNs, transfer learning, word clouds, Scattertext

Perfect for: Creating demonstrable skills for healthcare AI roles.

Annotation tools and techniques you'll master:

• Computer vision: Image classification, object detection, semantic segmentation with Pigeon, CVAT, Roboflow
• NLP annotation: Named entity recognition, text classification, sentiment analysis with Universal Data Tool and Prodigy
• Advanced techniques: Polygon masking, Segment Anything Model (SAM), video annotation
• Cloud platforms: AWS SageMaker Ground Truth, Azure ML Studio, Google Vertex AI

Tools: CVAT, Roboflow Annotate, Prodigy, Universal Data Tool, AWS/Azure/GCP annotation platforms

Perfect for: Building ML skills or starting a remote annotation side hustle.

Essential Colab skills you'll master:

• AI-powered coding: Autocomplete, code generation, and Gemini AI support in notebooks
• File management: Upload/download, Google Drive mounting, URL imports, Google Cloud Storage integration
• System interactivity: GPU/TPU access, magic commands, bash commands, Python os/sys modules
• Development workflow: Create Python scripts, markdown documentation, GitHub cloning and commits

Tools: Google Colab, Gemini AI, Google Drive, Google Cloud Storage, GitHub, GPU/TPU computing

Perfect for: Data scientists, ML practitioners, and anyone needing a free cloud-based Python environment.

Text processing techniques you'll master:

• Text representation: bag-of-words (BoW), TF-IDF, and n-grams
• Word embeddings: Word2Vec, GloVe, and fastText for semantic understanding
• Text preprocessing: tokenization, stemming, lemmatization, and stopword removal
• Part-of-speech (POS) tagging and named entity recognition with spaCy
• Dependency parsing and syntactic analysis

Speech processing techniques you'll learn:

• Speech feature extraction: MFCCs, spectrograms, and acoustic features with librosa
• Audio preprocessing and signal processing fundamentals
• Speech representation techniques for machine learning models

Advanced NLP algorithms covered:

• Text classification and sentiment analysis
• Topic modeling and document clustering
• Transformer models and contextual embeddings for language understanding

Tools: NLTK, spaCy, librosa, Gensim, scikit-learn, transformer models, Python

Project 1: Named Entity Recognition with spaCy

• Extract entities (people, organizations, locations) from text using spaCy
• Customize NER pipelines for domain-specific entities
• Visualize entity relationships and build entity extraction workflows

Project 2: Topic Modeling with Gensim

• Discover hidden topics in document collections using Latent Dirichlet Allocation (LDA)
• Preprocess text corpora and build topic models with Gensim
• Visualize topics interactively with pyLDAvis
• Optimize topic model parameters for better interpretability

Project 3: Sentiment Analysis with VADER & Transformers

• Perform rule-based sentiment analysis using VADER
• Implement deep learning sentiment analysis with pretrained transformer models
• Compare traditional vs. transformer-based approaches for sentiment classification
• Build sentiment analysis pipelines for real-world text data

Tools: spaCy, Gensim, pyLDAvis, VADER, Hugging Face transformers, Python

Advanced NLP techniques you'll master:

• Clinical Named Entity Recognition (CNER) with pretrained medical models
• Clinical Entity Resolution: abbreviation expansion and biomedical knowledge base linkage
• Medical text representation using word embeddings (fastText) and sentence encoding (Universal Sentence Encoder)
• Transformer models for clinical diagnosis prediction and masked word prediction
• Named entity recognition using transformer architectures for medical text

Tools covered: scispaCy, fastText, Universal Sentence Encoder (USE), transformer models for healthcare NLP

Hugging Face ecosystem:

• Explore and use pretrained models from the Hugging Face Model Hub
• Integrate transformers for NLP, computer vision, and audio tasks
• Deploy models using Hugging Face Inference API and Spaces
• Fine-tune models with Hugging Face Transformers and Datasets libraries

OpenAI API integration:

• Build applications with GPT models (chat completions, text generation, embeddings)
• Implement function calling and structured outputs for production workflows
• Use DALL-E API for image generation and manipulation
• Manage API keys, rate limits, and cost optimization strategies

Google AI Studio & Gemini API:

• Integrate Google's Gemini models for multimodal AI applications
• Use AI Studio for rapid prototyping and prompt engineering
• Build applications with Google's generative AI SDKs

GitHub Models & AI tools:

• Access and integrate models through GitHub Models marketplace
• Use GitHub Copilot API for code generation workflows
• Build AI-powered developer tools and automation

Tools: Hugging Face, OpenAI API, Google AI Studio, Gemini API, GitHub Models, Python SDKs

Diffusion model fundamentals:

• Understanding diffusion models: forward diffusion process and reverse denoising
• Diffusion vs. autoregressive models: architectural differences, strengths, and trade-offs
• Score-based generative modeling and noise scheduling strategies
• Text embedding with BERT and representation learning for diffusion models

Implementation skills:

• Building diffusion models for text generation from scratch
• Implementing the denoising process using neural networks
• Training diffusion models on text datasets
• Sampling and generating text with trained diffusion models
• Evaluating text quality and model performance

Applications covered:

• Controlled text generation with diffusion guidance
• Text editing and infilling with diffusion models
• Comparing diffusion-based vs. transformer-based text generation

Tools: PyTorch, Hugging Face Transformers (BERT embeddings), diffusion model architectures, Python

Jupyter AI setup and configuration:

• Installing and configuring Jupyter AI extension in JupyterLab
• Connecting to AI model providers (OpenAI, Anthropic, Cohere, and more)
• Managing API keys and model selection for different tasks
• Customizing AI behavior and preferences in JupyterLab

Jupyternaut chat interface:

• Interactive AI chat directly in JupyterLab for coding assistance
• Asking questions about code, data, and analysis workflows
• Getting explanations of complex functions and libraries
• Troubleshooting errors and debugging with AI guidance

Magic commands for AI-powered workflows:

• %%ai magic commands for in-cell AI generation and execution
• Generating code snippets, functions, and data transformations
• Creating visualizations and plots with AI assistance
• Automating data cleaning and preprocessing tasks

Practical AI applications:

• Code generation and completion for Python data science tasks
• Automated documentation and code commenting
• Data exploration and analysis suggestions
• Learning new libraries and frameworks with AI tutoring

Tools: JupyterLab, Jupyter AI extension, Jupyternaut, AI magic commands, LLM providers (OpenAI, Anthropic, Cohere)