Good Clinical has teamed up with Te Titoki Mataora, the MedTech Research Translator, to bring you the Te Titoki Mataora Podcast Series.
In this episode, meet Dr Gavin Harris, Anatomical Pathologist from Canterbury Health Laboratories, who discusses the development of a diagnostic assessment tool using machine-learning algorithms to assess breast cancer aggressiveness and recurrence risk.
We chat about several topics such as the process of training the algorithm on clinical data, the challenges of validation, and the regulatory steps required for commercialisation. He emphasises the importance of multidisciplinary collaboration, networks, and funding support from entities like Te Titoki Mataora, and shares his experience navigating the complexities of both the healthcare and commercial sectors. The ultimate goal is to provide a personalised tool that can improve clinical decision-making and patient outcomes by accurately assessing cancer recurrence risk and other components with over 95% accuracy.
Podcast Takeaways:
- Gavin is developing an AI tool to improve breast cancer risk assessment by identifying tumours with aggressive biological signatures that might not be evident through traditional diagnostics, as well as the likelihood of tumour recurrence.
- His tool has the potential to identify genomic features of the tumour that would otherwise only be identified through expensive molecular testing that not everyone can currently access. Having such a novel tool available means increased access, lower cost, and reduced time required for diagnostics and assessment.
- The current accuracy of the tool is around 80%, but the team is aiming for a higher threshold preferably above 95%, to be confident in its diagnostic value.
- Gavin highlights the importance of multidisciplinary collaboration such as with specialists from oncology, pathology, and AI, as well as the support of entities like Te Titoki Mataora and Health New Zealand.
- The project has to meet several parameters related to FDA and TGA regulations, with a pathway to commercialisation taking 3-5 years, involving rigorous testing and validation on independent and international datasets.
- Gavin underscores how crucial research offices, innovation hubs, and funding bodies are in providing the necessary infrastructure and resources for such complex projects.
- Gavin reflects on how much he has learned in transitioning from clinical work to understanding the complexities of commercialisation, regulation, and scalability, emphasising the value of building strong networks and continuously learning.
Timestamps:
- 00:00 Introduction and Background
- 03:07 Gavin Harris's Journey in Pathology
- 05:53 Traditional Approaches to Breast Cancer Diagnosis
- 11:00 Incorporating Algorithmic Approaches
- 19:06 Funding and Project Support
- 25:40 Research Collaborations
- 30:00 Commercialisation and Regulatory Challenges
- 41:15 The Support of Te Titoki Mataora
- 47:35 Advice for Innovators in Healthcare _
