Artificial intelligence in Healthcare Course

Lesson 1 • Healthcare Data Landscape

Surveys the types, sources, and formats of clinical data that fuel AI systems. Connects data availability to model feasibility in real-world settings.

Lesson 2 • History of AI in Medicine

Traces AI milestones from early expert systems to modern deep learning applications. Contextualises current capabilities within decades of research progress.

Lesson 3 • Key Stakeholders and Roles

Identifies clinicians, data scientists, administrators, and patients as AI stakeholders. Clarifies how each role influences AI adoption and governance.

Lesson 1 • Feature Engineering for Clinical Data

Transforms raw clinical variables into informative features that improve model performance. Bridges domain knowledge and statistical representation of patient information.

Lesson 2 • Data Collection and Sourcing

Covers primary and secondary data sources including registries, claims, and sensors. Establishes sourcing strategies that ensure representativeness and completeness.

Lesson 3 • Interoperability and Data Standards

Explains health data exchange standards that enable system integration and data reuse. Prepares students to work across heterogeneous clinical information systems.

Lesson 4 • Data Cleaning and Preprocessing

Addresses missing values, outliers, duplicates, and inconsistent coding in clinical datasets. Directly enables reliable model training by improving input data quality.

Lesson 1 • Regulatory Pathways for AI Devices

Surveys medical device regulatory frameworks applicable to AI-based clinical tools. Enables students to navigate approval processes and post-market obligations.

Lesson 2 • Ethical Frameworks for Healthcare AI

Applies bioethical principles—autonomy, beneficence, non-maleficence, justice—to AI decisions. Provides a structured lens for evaluating algorithmic choices.

Lesson 3 • Patient Privacy Principles

Covers de-identification, consent, and data minimisation as core privacy protections. Grounds ethical AI practice in patient rights and trust.

Lesson 4 • Algorithmic Bias and Fairness

Identifies sources of bias in training data and model outputs that affect clinical equity. Teaches mitigation strategies to reduce disparate impact across patient groups.

Lesson 1 • Model Evaluation and Validation

Teaches performance metrics, cross-validation, and calibration specific to clinical AI. Ensures students can critically assess whether a model is ready for clinical use.

Lesson 2 • Supervised Learning in Clinical Settings

Covers classification and regression models for diagnosis, prognosis, and risk scoring. Connects labelled clinical outcomes to model training objectives.

Lesson 3 • Handling Class Imbalance in Healthcare

Addresses rare-event prediction challenges common in clinical datasets such as sepsis or readmission. Provides resampling and cost-sensitive learning solutions.

Lesson 4 • Reinforcement Learning in Treatment Optimisation

Introduces reward-based learning for dynamic treatment regimen optimisation. Illustrates how sequential clinical decisions can be modelled as policy problems.

Lesson 1 • Radiology AI Applications

Applies CNNs to chest X-ray, CT, and MRI interpretation for detection and triage tasks. Demonstrates how imaging AI integrates into radiologist workflows.

Lesson 2 • Convolutional Neural Network Fundamentals

Explains convolution, pooling, and feature map hierarchies that underpin image recognition. Provides the architectural foundation for all medical imaging AI models.

Lesson 3 • Pathology and Dermatology Imaging AI

Covers whole-slide image analysis and skin lesion classification using deep learning. Extends imaging AI beyond radiology to histopathology and dermatology.

Lesson 4 • Image Segmentation and Object Detection

Teaches U-Net, Mask R-CNN, and related architectures for pixel-level clinical annotation. Enables precise organ and lesion delineation for surgical planning and monitoring.

Lesson 1 • Clinical Language Models and BERT Variants

Covers pre-trained biomedical language models and fine-tuning strategies for clinical tasks. Enables students to leverage large-scale pre-training for downstream NLP applications.

Lesson 2 • Large Language Models in Clinical Practice

Evaluates GPT-class models for clinical question answering, coding, and documentation support. Addresses hallucination risks and safe deployment practices in clinical environments.

Lesson 3 • Information Extraction and Summarisation

Applies relation extraction and summarisation to discharge summaries and clinical reports. Reduces documentation burden and surfaces actionable insights from text.

Lesson 4 • Clinical Text Characteristics

Describes the unique linguistic features of clinical notes including abbreviations and negation. Establishes why general NLP tools require adaptation for medical text.

Lesson 1 • Change Management and Clinician Adoption

Applies change management frameworks to overcome resistance and build clinician trust in AI. Directly determines whether technically sound models achieve real-world clinical impact.

Lesson 2 • Clinical Workflow Analysis

Maps existing clinical processes to identify integration points and friction for AI tools. Ensures AI solutions address genuine workflow needs rather than creating new burdens.

Lesson 3 • Post-Deployment Monitoring and Maintenance

Establishes processes for detecting model drift, performance degradation, and safety signals. Ensures sustained clinical value after initial deployment through continuous oversight.

Lesson 4 • Model Deployment Architectures

Covers cloud, on-premise, and edge deployment options for clinical AI systems. Connects infrastructure choices to latency, security, and scalability requirements.

Lesson 1 • Building an AI Readiness Assessment

Evaluates organisational data maturity, talent, infrastructure, and culture for AI adoption. Produces a structured readiness scorecard applicable to any healthcare organisation.

Lesson 2 • Building a Culture of AI Innovation

Establishes leadership practices that sustain continuous AI learning and experimentation. Closes the course by connecting technical mastery to organisational transformation.

Lesson 3 • Emerging AI Technologies in Healthcare

Surveys federated learning, generative AI, digital twins, and multimodal models on the horizon. Prepares students to evaluate and adopt next-generation tools responsibly.

Lesson 4 • AI Governance and Portfolio Management

Designs governance structures to prioritise, oversee, and retire AI initiatives systematically. Prevents ungoverned AI proliferation and ensures strategic alignment.