2 Internships - #1 Ultrasound Simulation | #2 Computer Vision

To push the boundaries of transcranial Contrast-Enhanced Ultrasound (CEUS), we need a reliable digital twin of our imaging chain. A simulator is only useful if it accurately captures both our specific hardware and the complex physics of transcranial wave propagation and microbubble scattering.

The goal of this internship is to build a plug-and-play, hardware-accurate ultrasound simulation workflow calibrated against experimental data. Starting from controlled in vitro benchtop measurements, you will develop a simulator and align it with our hardware, extend it to transcranial brain imaging, and validate it against in vivo clinical trial datasets.

The internship will include:

• State-of-the-Art Review: Benchmark existing spatio-temporal ultrasound simulator approaches on physical accuracy, hardware parameterization, and compute speed.
• Hardware & Physics Modeling: Model Resolve Stroke’s specific hardware (transducer characteristics, acoustic field, sequences) alongside the key physical effects of transcranial CEUS (skull aberration/attenuation, microbubble dynamics).
• Experimental Calibration (In Vitro): Calibrate and validate the simulator against lab measurements (hydrophone, wire and flow phantoms) so synthetic raw data matches real acquisitions.
• Pipeline Benchmark & Clinical Comparison (In Vivo): Implement and benchmark the end-to-end simulation pipeline, comparing synthetic outputs with real clinical trial datasets.
• Production Integration: Package the simulator into a plug-and-play module directly compatible with Resolve Stroke’s production software.

All software development activities will be done in Python.

Position #2 - Computer vision and feature recognition: AI-Assisted transcranial ultrasound imaging

Obtaining high-quality ultrasound images of the brain depends heavily on the correct positioning of the probe on the patient’s temple. Today, the device only displays a live image, with no guidance to help the physician find the optimal position.

The goal of the internship is to assist the physician during the positioning stage by providing real-time feedback on what constitutes a “good position”. You will develop quality scores, anatomical feature recognition, and indicators derived from the live imaging stream to help optimize probe placement.

The internship will include:

• State-of-the-art review: benchmark existing approaches to feature recognition and image quality assessment in ultrasound, and understand what distinguishes a good from a bad probe position.
• Solution design: identify and select the most relevant, robust and efficient approaches (classical image processing, machine learning, deep learning).
• Real-time implementation: develop a real-time pipeline running on both simulated and real hardware.
• Evaluation: validate the solution on existing datasets and in live conditions.

All software development activities will be done in Python.

MSc, Master, or Engineering Degree student ideally specialized in signal/image processing, biomedical engineering or medical imaging.

• Experience with Python, especially scientific computing (NumPy, SciPy).
• Experience with data analysis techniques, statistics, and imaging processing.
• Experience in medical imaging.
• Position #2 (Computer Vision):
• MSc, Master, or Engineering Degree student ideally specialized in machine learning, computer vision, signal/image processing or medical imaging.
• Experience with Python, especially scientific computing (NumPy, SciPy) and machine learning frameworks (PyTorch, OpenCV).
• Experience with computer vision is a strong plus.
• Experience in medical imaging or ultrasound signal processing is a plus.

Who you are:

• Fluent in spoken and written English. Knowledge of French is a plus.
• Ability to work effectively in a fast-paced and dynamic start-up environment.
• Strong autonomy, analytical, organizational, and problem-solving skills.
• A committed, enthusiastic person with a “can-do”, creative mindset.

• A respectful, fun, and collaborative work environment.
• Monthly compensation: €1,400.