Computer Vision and Machine learning for Objective clinical psychiatry in severe depression, the CALYPSO project (in French: Clinique et AnaLYses PSychiatriques Objectives)

23 Avril 2023

---

Catégorie : Post-doctorant

---

Summary of the project

Depression is a common psychiatric disorder. The WHO estimates that more than 300 million people worldwide suffer from this disorder. In France, between 7 and 10% of the population is affected. Depression is the cause of significant costs (direct and indirect) for society and generates a high level of disability. It is also associated with a high risk of suicide. It is therefore a frequent and serious disorder. However, little is known about depression regarding its clinical manifestations, prognosis, pathophysiology and therapeutics. For example, there is no biomarker used in routine psychiatric clinical practice. Indeed, the diagnosis and assessment of depression relies on the physician's expertise in detecting and identifying symptoms. Thus, it becomes crucial to find a way to perform an easy and objective clinical assessment of patients with severe depression in order to improve the accuracy of diagnosis, as well as prognosis and potential therapeutic actions.

Interestingly, in depression, many clinical manifestations are observable and measurable. For example, there is a slowing down of movements, but also a reduction in physical activity, tone of voice, facial expression and head movements. Recent technical advances in computer science (computer vision) and artificial intelligence make it possible to objectify, measure and analyse these symptoms.

The objectives of the CALPYPSO project are, on the one hand, to identify and describe objective markers associated with the severity of depression and its clinical features and, on the other hand, to evaluate the predictive value of the objective markers on the remission of the major depressive episode at the different times of the study (3, 6 and 12 months).

 

Key facts on the post-doctoral project:

• Faculty sponsor/ PI: Pr. Mohamed Daoudi (Computer Science and AI) & Pr Ali Amad (Psychiatry)
• Funding source: Fondation de France (https://www.fondationdefrance.org/fr/)
• Salary will be determined based on the experience of the candidate and the University guidelines for postdoctoral fellows.
• Location/ Department: UMR 9189 CRIStAL / Lille University Hospital, psychiatry department.

 

Job description

The postdoctoral researcher will work on a research project focused on detecting objective markers of depression, including facial expressions, face movements, speed of movement, and voice analysis. The researcher will be responsible for analyzing these markers, combining them, detecting subgroups of depressive patients based on these objective characteristics, and predicting the evolution of depression.

 

Tasks and Responsibilities:

• Involvement in the installation and functioning of the device from a technical point of view
• Develop and implement algorithms for analyzing facial expressions, face movements, speed of movement, and voice analysis…
• Combine the analyzed data to form a comprehensive understanding of the objective markers of depression.
• Use machine learning techniques to identify subgroups of depressive patients based on the objective markers.

• Predict the evolution of depression using the analyzed data.
• Conduct experimental evaluations of the developed algorithms and techniques.
• Collaborate with interdisciplinary teams to integrate the findings into clinical practice.
• Present the results at scientific conferences and publish them in peer-reviewed journals.
• Contribute to the overall direction and strategy of the research project.

The postdoctoral researcher will have the opportunity to work on cutting-edge research in the field of computer vision and contribute to the development of novel solutions for depression diagnosis and treatment.

 

Required Education and Skills

• A Ph.D. in Computer Science, Electrical Engineering, or a related field with a focus on computer vision, machine learning, and artificial intelligence.

 

Professional Experience:

• Strong research background in computer vision, machine learning, and artificial intelligence.
• Familiarity with deep learning techniques, such as convolutional neural networks and recurrent neural networks.
• Experience in working with large-scale datasets and conducting experimental evaluations.

 

Technical Skills:

• Strong programming skills in Python and experience with computer vision libraries such as OpenCV, TensorFlow, PyTorch, or similar.
• Knowledge of machine learning techniques and deep learning frameworks.
• Knowledge of computer vision techniques for face and gesture recognition.
• Strong analytical and problem-solving skills.
• Experience in working with interdisciplinary teams and collaborating with experts from other fields.

In addition to these technical skills, the postdoctoral researcher should possess excellent communication and interpersonal skills, be able to work independently and as part of a team, and have a strong motivation for conducting impactful research in the field of computer vision and artificial intelligence for healthcare applications.

 

Application Guidelines

Please submit application materials to mohamed.daoudi@imt-nord-europe.fr & ali.amad@univ-lille.fr, subject: [CALYPSO]. To be considered, your application must include: (i) an updated CV, with full track record and a sample of published work, (ii) a cover letter highlighting relevant research experience to the project, interest in the position and the earliest date you could start, (iii) two letters of endorsement/support from academic mentor/reference (including their contact information), and specifically written for this position.

Deadline submission: 1st June 2023

The position is funded for 18 months or 2 years from the date of joining.