Medical Imaging and Deep Learning Engineer (Regulated Medical Devices)

About Ceretas

Ceretas is a Brisbane-based medical device company developing focused ultrasound (FUS) therapeutic systems for neurological conditions, including Alzheimer's disease. Our first system is heading into clinical trials in mid-2026, and a next-generation platform is in active development.

We're a small, fast-moving engineering team building a regulated medical device from the ground up. You will work alongside firmware, software, and hardware engineers, and directly with the CTO, on problems that have clear clinical impact.

About the Role

We are developing an Image Guided Navigation (IGN) system for planning and delivering transcranial focused ultrasound treatments. Personalised, patient-specific treatment planning depends on medical imaging algorithms that turn patient MRI data into inputs for acoustic simulation and treatment delivery.

We need an engineer who can own the medical imaging and deep learning side of this pipeline. This is primarily a deep learning role for medical imaging, with the hands-on software engineering skills needed to take models all the way into a shipping, regulated application under IEC 62304.

You will work with real patient data, build and deploy models that directly affect clinical outcomes, and collaborate closely with the engineers responsible for acoustic simulation, treatment planning, and delivery. What we need from day one is strong fundamentals in medical imaging and deep learning.

What You'll Do

Medical image analysis — deep learning and beyond

We use the right tool for each problem: deep learning where it gives a real advantage, atlas-based registration or other established neuroimaging methods where they do.

• Develop and deploy brain segmentation on patient MRI to produce anatomical outputs that feed treatment planning. Open to deep learning, atlas-based registration, or other established neuroimaging methods (e.g. statistical, surface-based) — whichever gives the best clinical result.
• Develop and deploy deep learning models for cross-modality image synthesis (e.g. generating synthetic CT from MRI) to support acoustic modelling and treatment personalisation.
• Design training and validation strategies that handle scanner and protocol variability and small-dataset regimes.
• Stay current with the medical imaging literature and bring relevant advances into production.

Medical imaging pipeline work

• Own the end-to-end medical imaging pipeline — from DICOM ingestion through to treatment-ready outputs — and the image pre-processing, registration, and post-processing steps around the ML models.
• Improve and maintain existing classical imaging components.

Integration and productisation

• Integrate algorithms into our C++/Qt planning and treatment applications — not just Python research scripts.
• Meet clinical performance, memory, and reliability constraints.
• Deliver production-quality software, not research prototypes.

Validation and regulatory

• Design validation strategies for algorithm accuracy against appropriate ground-truth datasets.
• Document algorithms to medical device standards (IEC 62304, ISO 13485).
• Support verification and validation activities for regulatory submissions.

What We're Looking For

Must have

• Degree in Biomedical Engineering, Computer Science, Medical Physics, Applied Mathematics, or a closely related field (Master's or PhD preferred).
• Strong hands-on experience with medical imaging — MRI and CT — in a research or product setting.
• Demonstrated experience developing and deploying deep learning models for medical image analysis (segmentation, synthesis, or similar), using PyTorch or equivalent.
• Proficiency in Python for model development and for building the surrounding pipelines.
• Working knowledge of ITK / VTK or an equivalent open-source medical imaging stack.
• Solid understanding of medical image processing fundamentals — filtering, segmentation, registration, interpolation, coordinate systems.
• Experience with DICOM and NIfTI in a clinical or research setting.
• Ability to take an algorithm from research prototype to production-quality code that runs inside a desktop application.
• Ability to work independently, take ownership, and deliver working solutions under real deadlines.
• Excellent communication skills — able to explain complex algorithms to clinicians, regulators, and non-specialist engineers.

Discover our full range of relocation jobs with comprehensive support packages to help you relocate and settle in your new location.

Strongly preferred

• Experience with brain MRI and neuroimaging pipelines — deep learning, atlas-based registration, or other established neuroimaging methods (e.g. statistical, surface-based).
• Experience with cross-modality image synthesis (e.g. MRI-to-CT) or other image-to-image deep learning problems.
• C++ experience for integration with desktop applications (Qt, ITK/VTK in C++).

Nice to have

• Experience in a regulated environment — medical devices, clinical research, or similar.
• Published research in medical imaging, neuroimaging, or related fields.
• Experience with GPU inference optimisation for deployed models.
• Familiarity with MLOps practices — dataset versioning, reproducible training, model registry.

The Team

You'll join a small, senior engineering team based in Brisbane CBD, spanning software, firmware, and hardware. We also collaborate with university research groups.

This is a small team, so your contributions will be visible and impactful from day one.

Tech Stack & Tools

• Languages: Python, C++
• Medical imaging: ITK, VTK, DICOM, NIfTI
• ML frameworks: PyTorch
• Desktop applications: Qt (C++)

Interested in relocating to Australia? Check out our comprehensive Relocation Jobs in Australia page with detailed relocation packages and benefits.

• Infrastructure: GitLab, CI/CD, Linux

Why This Role

• Direct clinical impact — your models will run inside a real medical device treating patients with neurological conditions.
• Hard, meaningful problems — brain segmentation, cross-modality synthesis, and imaging-driven tissue property estimation are genuinely challenging and scientifically interesting problems, especially in a regulated clinical context.
• Ownership — you will own the medical imaging and deep learning pipeline end to end, not a small slice of a large system.
• Small team, big responsibility — no bureaucracy, direct access to the CTO and to clinical context.
• Growth — build deep expertise at the intersection of medical imaging, deep learning, transcranial FUS, and regulated clinical medical devices.

Details

• Location: Brisbane CBD. Hybrid — minimum 50% office attendance required. We believe regular in-person collaboration with the engineering team is critical at this stage of the company. Relocation support available for the right candidate.
• Type: Full-time permanent
• Start: As soon as available
• Salary: Competitive, commensurate with experience

How to Apply

Send your CV and a brief note covering:

1. Your most relevant medical imaging deep learning project and your specific contribution to it
2. Your experience with MRI and CT data
3. Your availability and start date