Training activity information
Details
Select, train and optimise an AI model on a healthcare dataset
Type
Developmental training activity (DTA)
Evidence requirements
Evidence the activity has been undertaken by the trainee.
Reflection on the activity at one or more time points after the event including learning from the activity and/or areas of the trainees practice for development.
An action plan to implement learning and/or to address skills or knowledge gaps identified.
Considerations
- Feature engineering
- Feature selection methods
- Classification vs regression
- Supervised vs unsupervised machine learning techniques
- Use of appropriate algorithms
- Processing appropriate to data
- Feed-forward and backpropagation
- Optimisation
- Bias-variance tradeoff
- Best practice, sharing knowledge and output reproducibility
Reflective practice guidance
The guidance below is provided to support reflection at different time points, providing you with questions to aid you to reflect for this training activity. They are provided for guidance and should not be considered as a mandatory checklist. Trainees should not be expected to provide answers to each of the guidance questions listed.
Before action
- What do you need to know before starting this process? This includes understanding various AI model architectures, training methodologies, hyperparameter tuning techniques, and evaluation metrics.
- What do you anticipate you will learn from this experience? Consider gaining practical experience in the end-to-end process of building and refining an AI model. Reflect on your theoretical understanding of machine learning algorithms and model development.
- What actions will you take in preparation for this experience? Will you review different AI model types and their applications? Will you research model training and optimisation strategies? Will you discuss the dataset and desired outcomes with your supervisor? Consider potential challenges in model selection, overfitting, or achieving satisfactory performance and how you might address them. Identify how you feel about embarking on this training activity.
In action
- When selecting the AI model, what types of algorithms are you considering and why? How are you configuring the model architecture and hyperparameters for training?
- During training, how are you monitoring the model’s performance? What decisions are you making about adjusting training parameters or strategies?
- Which aspects of model selection, training, and optimisation feel more familiar, and where do you need to apply more focused effort to ensure good performance and avoid issues like overfitting?
- How effectively do you believe the model is learning from the data? What challenges are you facing in achieving satisfactory performance or optimising the model? What insights are you gaining about the suitability of different models for this type of healthcare data?
- What are you learning about the practical aspects of training and optimising AI models? How does this connect to your theoretical understanding of machine learning principles?
- If you encounter poor model performance or training difficulties, what alternative models or optimisation techniques could you explore? Would reviewing relevant literature or seeking advice from an AI expert be beneficial now? Are you following best practices for training and evaluating AI models?
On action
- Describe the AI model you selected, the healthcare dataset you used for training, the training process, and the optimisation techniques you applied.
- What did you learn about the process of selecting, training, and optimising an AI model? What challenges did you encounter during the training or optimisation process? How did you evaluate the performance of the model? Did you need to adjust your training or optimisation strategies based on the model’s performance?
- What specific AI modelling techniques do you want to explore further? How will you improve your skills in training and optimising AI models for healthcare applications? What are your next steps in learning more about model evaluation metrics and optimisation strategies? Do you require any further resources on specific AI algorithms or training frameworks?
Beyond action
- Have you revisited the AI model you worked with, the training process, and the optimisation techniques you applied? What challenges did you encounter, and how has your understanding of AI model development deepened? Have you trained other models since?
- How has this hands-on experience influenced your ability to critically evaluate AI applications in healthcare in your current practice? Has it highlighted the importance of model performance and generalisation?
- What transferable skills, such as problem-solving and technical AI skills, did you develop? What further learning in AI model selection, training, and optimisation would be beneficial?
Relevant learning outcomes
| # | Outcome |
|---|---|
| # 10 |
Outcome
Design, develop, train and validate AI models using alphanumeric and imaging datasets. |