MRI —
Training activity: 17

Details

Review the impact of a commercially available AI-based technique in MR image reconstruction for a common clinical application, including its potential impact on patient care, its limitations and any associated risks

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.

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 are the intended outcomes of the training activity?

• How can the learning outcomes related to modelling signals and image reconstruction and appraising emerging techniques contextualise your preparation and focus your attention on the nuances of AI algorithms?
• What foundational knowledge regarding k-space manipulation and traditional reconstruction limits do you need to know before embarking on an assessment of AI-based alternatives?

What do you anticipate you will learn from the experience?

• What specific insights do you hope to gain from engaging with this activity, particularly concerning the balance between clinical throughput and the risk of image artefacts?
• What is your existing knowledge regarding this specific AI reconstruction task, and how do you expect this experience to build upon your current understanding of MR physics?
• How do you anticipate this activity will shape your perspective on the limitations and risks of emerging technologies in a clinical environment?

What actions will you take in preparation for the experience?

• How will you structure your discussion with your Training Officer to ensure you have a clear understanding of the departmental requirements for AI validation?
• What possible challenges might you face during the activity—such as a lack of vendor transparency regarding training datasets—and how do you plan to handle them?
• How do you feel about embarking on this activity, and how might this influence your approach to appraising this new technology?

In action

What are you doing?

• How are you currently approaching the review of this AI-based technique, and why have you chosen this specific clinical application to focus on?
• What specific decisions are you making as you compare AI-reconstructed images against traditional k-space modelling?
• Which parts of appraising this emerging technology feel intuitive to you, and which aspects—such as identifying subtle AI-generated artefacts—require more of your conscious effort?

How are you progressing with the activity?

• How effective are your current methods in identifying the potential risks and limitations of this ‘black-box’ commercial algorithm?
• What unexpected challenges are you facing right now as you attempt to quantify the impact of AI reconstruction on patient care and diagnostic quality?
• In what ways is the data you are seeing now connecting to your existing knowledge of signal modelling and traditional image reconstruction?
• What insights are you gaining ‘in the moment’ about the role of the MRI physicist in validating AI-based software?

How are you adapting to the situation?

• Are there alternative ways you should be evaluating these reconstruction artefacts that you hadn’t considered before starting the task?
• As you uncover potential risks or limitations, what additional support or technical guidance from your Training Officer might you need right now?
• How are you ensuring that your appraisal of this new technology remains strictly within your professional scope of practice?

On action

What did you notice?

• How would you summarise the key technical findings regarding the AI technique’s impact on image reconstruction compared to traditional methods?
• What were the most significant clinical impacts—positive or negative—that you noticed when reviewing this technique for your chosen clinical application?

What did you learn from the activity?

• In what ways has your knowledge of signal modelling and k-space manipulation improved by evaluating how the AI algorithm “fills” or reconstructs data?
• What did you learn about the limitations and risks (e.g., ‘black-box’ transparency or artefacts) of this emerging technique through your appraisal?
• Were there any unexpected challenges or successes—such as finding a specific risk to patient care—and what did those moments teach you about AI validation?
• How did your reflect-in-action (your thoughts and adjustments during the review) influence the final conclusions of your appraisal?
• How does this experience of evaluating an AI-based technique relate to the standards expected of an MRI physicist in post-programme practice?

What will you take from the experience moving forward?

• What specific areas for continued development in AI or emerging MR technologies have you identified as a result of this activity?
• How can you apply the critical appraisal skills gained here to your routine practice when you are asked to evaluate other new hardware or software developments?
• What are the clear ‘next steps’ you will take—such as further reading on deep-learning artefacts or discussing findings with your Radiologist colleagues—to support your learning?
• What additional support or resources (e.g., vendor-specific technical training or research papers) do you now realise you need to further develop your expertise in MR image reconstruction?

Beyond action

Have you revisited the experiences?

• How has your perspective on the limitations of AI-based reconstruction evolved now that you have compared this training activity with other emerging techniques or hybrid modalities encountered during your training?
• How does your critical appraisal of this commercial AI tool compare with the observed behaviours of senior physicists you shadowed during clinical experiences, such as observing advanced MRI reporting or translational research meetings?
• Having shared your findings with peers or colleagues through professional storytelling, how has their clinical feedback on AI-generated artefacts changed your own technical evaluation of image quality issues?

How have these experiences impacted upon your current practice?

• How has your investigation into signal modelling and reconstruction for AI improved your ability to troubleshoot routine image quality issues or optimise clinical protocols across other areas of your practice?
• In what ways have you applied the communication and report-writing skills developed during this appraisal to explain complex technical risks of “black-box” algorithms to non-physics staff?
• How will the insights gained from this activity support your preparation for observed assessments, such as a Case-Based Discussion (CBD) or a DOPS involving the setup of a volunteer for an advanced quantitative technique?

How might these experiences contribute towards your future practice?

• Which transferable skills—such as the ability to critically evaluate vendor technical claims—did you develop during this activity that will be most valuable in your future post-programme practice?
• How will your experience in appraising the clinical utility and risks of emerging techniques inform your future professional role in service development, equipment specification, or commissioning?
• What clear next steps have you identified for your continued development to ensure you remain at the forefront of AI advancements and their safe application in the clinical environment?

Relevant learning outcomes

#	Outcome
# 2	Outcome Utilise theoretical understanding of pulse sequences, MR signal evolution and image reconstruction in practical problem solving and image optimisation.
# 10	Outcome Appraise the key issues of an emerging technique/technology and a hybrid modality.