Intelligent systems are becoming increasingly prevalent in domains such as medical imaging, pathological analysis, neuropsychology, and rehabilitation. Despite their growing adoption, systematic frameworks for the integration and evaluation of explainability and responsibility within such systems remain insufficiently developed. This research aims to investigate and advance methods, protocols, and tools for the systematic incorporation and assessment of explainability and responsibility in intelligent medical systems. Building upon prior research conducted at Curtin and beyond, this project seeks to support both developers and end users in the confident adoption and deployment of smart technologies.
Aim
This project aims to:
- Develop methods, protocols, and tools for embedding explainability and responsibility in intelligent medical systems.
- Establish methods of evaluating levels of embedded explainability and responsibility using Australian and ISO standards of designing intelligent medical systems.
Objectives
- A comprehensive review of prior research on explainable artificial intelligence, responsible artificial intelligence, trustworthy medical AI, and regulatory governance in healthcare.
- Analyse Australian regulatory guidance and international standards applicable to intelligent medical systems. This includes alignment with the Therapeutic Goods Administration expectations for AI-enabled medical device relevant ISO and IEC standards related to lifecycle processes, risk management, and explainability, e.g., ISO/IEC 42001, ISO/IEC 23894, ISO/IEC 25059, and emerging guidance on explainability and interpretability.
- Identify technical, ethical, clinical, and regulatory requirements through consultation with relevant stakeholders and analysis of use-case needs.
- Design practical methods, implementation protocols, and supporting tools that enable developers to incorporate explainability and responsibility into intelligent medical systems.
- Design of evaluation criteria and metrics: Develop a set of qualitative and quantitative criteria for assessing the degree to which explainability and responsibility are embedded within a system.
- Prototype implementation and case study application: Apply the proposed framework, methods, and evaluation approach to selected case studies or prototype intelligent medical systems. This stage will demonstrate how the proposed methodology performs in realistic healthcare contexts and will provide evidence of feasibility and adaptability across different application domains.
- Validate the proposed methods and evaluation framework through expert feedback from researchers, clinicians, developers, and regulatory or ethics-informed stakeholders where appropriate.
- Refine the methodology, tools, and evaluation model based on findings from implementation and validation.
Significance
This research addresses a critical gap in the safe, effective, and trustworthy adoption of intelligent medical systems by developing systematic methods for embedding and evaluating explainability and responsibility. As intelligent systems are increasingly applied in high-impact healthcare domains such as medical imaging, pathology, neuropsychology, and rehabilitation, the absence of robust frameworks for transparency, accountability, and standards-aligned evaluation presents substantial technical, ethical, and regulatory challenges. The proposed research responds directly to these challenges by integrating explainability and responsibility into the design and assessment of intelligent medical systems in a manner that is aligned with contemporary Australian regulatory expectations for AI-enabled medical device software and internationally recognised AI governance standards. In doing so, the project has the potential to improve confidence among developers, clinicians, patients, and regulators; strengthen evidence-based and human-centred deployment of intelligent systems in healthcare; and contribute practical tools, protocols, and evaluation criteria that support safer innovation, regulatory compliance, and broader societal trust in medical AI.
Ideal Candidate
Strong background in any one or more of the following areas: Biomedical engineering, Mechatronics, Mechanical Engineering, Computational Sciences and Electrical Engineering.
Previous experience in reviewing scientific literature and writing research/ technical reports.
Proficiency in programming in one or more of: Python, C#, C++, Matlab.
Ability to code mathematical concepts in the above-mentioned programming languages.
Additionally, the applicants should meet the eligibility criteria for entry into a PhD program at Curtin University.
This project is open to International and Domestic applicants.
Scholarship
If you are identified as the preferred candidate for this project, you may be considered for an RTP scholarship.
Enquires and How to Apply
For enquires about this opportunity contact Dr Masood Khan at Masood.Khan@curtin.edu.au
To formally apply submit an Expression of Interest to Dr Masood Khan during the Central Scholarship round (July 1st – July 31st 2026)