Modelling the expertise of medical practitioners in diagnosing spasticity from multi-sensors gesture data

M2 internship in AI and Data Science:

Spasticity is a motor disorder characterized by muscular hyperactivity caused by impaired nerve conduction. Diagnosis of this pathology relies on the assessment of the degree of resistance of the limb following a passive movement performed by the practitioner, and is used to determine the treatment to be followed. However, this assessment remains subjective and requires practical experience, making it difficult to learn for young practitioners.

The overall goal of this research project ( is to develop a simulator capable of reproducing different degrees of spasticity, to enable young practitioners to practice before actually performing their gestures on a patient. A key component of the simulator is the ability to analyze the gestures performed by the practitioners to infer their underlying expertise and knowledge. This, in turn, can be used to provide more personalized spasticity cases in the simulator to improve learning.

In this internship, we will focus on the gesture analysis part of the project, namely inferring the levels of expertise of a practitioner solely based on data captured while performing the spasticity test. To do so, we have already collected a dataset of gestures performed by several practitioners, captured via video recording and motion sensors. The first part of the internship will involve exploring, cleaning, aligning and fusing these data as needed. Next, artificial intelligence (AI), data mining and data analysis methods will be explored to model how experts perform the spasticity test from the data, as well as to build a predictive model of the practitioners’ expertise.

The intern should be in the final year of a masters’/engineering degree (≈M2 students) in computer science, or a closely related field. Strong programming skills and knowledge about methodologies and tools for AI and data analysis are expected.

The internship will take place at the LIP6 laboratory at Sorbonne University in Paris, one of the main computer science research laboratories in France. The intern will be supervised by Vanda Luengo and Sébastien Lallé, two professors specialized in technology-enhanced learning, AI in education, and learning analytics (cf. The intern is expected to work mostly in person on the campus, 4 Place Jussieu 75005 Paris. The start date is flexible, but ideally early 2024, for a 6-months duration, paid about 580€/month.

Importantly, this project comes with a PhD scholarship, meaning that the intern will be offered the opportunity to continue as a PhD student, depending on the outcome of the internship. The PhD research will be on the same project, focusing on how to infer specific skills from the simulators’ data and how to best recommend new spasticity cases to learners, to personalize their learning. The PhD will involve a tight collaboration with a multidisciplinary team involving physiological therapists and biomechanics & mechatronics researchers, as well as participation in the development and evaluation of the training simulator.

To apply, please send a CV and a motivation statement to:

Campus Jussieu, 75005 Paris
Vanda Luengo
Sébastien Lallé
Référent Universitaire: 
2 024

User login