Annonce

13 septembre 2024

Postdoctoral position in artificial intelligence for diabetic retinopathy management

Catégorie : Post-doctorant

The Laboratory of medical information processing (LaTIM – UMR 1101 INSERM) is opening a 1-year postdoc position in artificial intelligence for diabetic retinopathy management, within the framework of the ANR RHU EviRed project (Intelligent Evaluation of Diabetic Retinopathy).

Context

EviRed Project

Diabetic retinopathy (DR), an ocular complication of diabetes, is a leading cause of blindness in developed countries. An important obstacle to fight DR is the use of a classification based on an old imaging technique: color fundus photography (CFP). This classification is insufficient to finely predict the future evolution: in ~50% of cases, ophthalmologists over or under-estimate the advent of complications. New available imaging techniques are more powerful in this context: ultra-wide-field color fundus photography (UWF-CFP) gives useful 2-D information on the periphery of the retina, not seen on standard CFP. Structural optical coherence tomography (OCT), which produces few microns resolution cross-sectional 3-D imaging, is the reference for the diagnosis of diabetic macular edema, one complication of DR. It has been enriched with OCT angiography (OCTA), which can show the vasculature of the retina non-invasively. However, these new imaging modalities produce an expanding amount of data which requires high human expertise. Any clinical score based on them will be complex and challenging for most ophthalmologists. Therefore, the purpose of the EviRed project is to replace the current classification with an AI-based expert system integrating multi-modal data to propose diagnosis and prediction.

The EviRed project, led by Paris Hospitals (AP-HP), groups together the LaTIM laboratory, 14 hospitals in France and two companies (Evolucare Technologies and ZEISS).

LaTIM Laboratory

The postdoc will be hosted by LaTIM, in Brest, France, which leads AI development in the EviRed project. Born from the complementarity between Health and Communication sciences, the LaTIM ("laboratoire de traitement de l’information médicale" for laboratory of medical information processing) develops multidisciplinary research driven by members from University of Western Brittany (UBO), IMT Atlantique, INSERM and Brest University Hospitals (CHRU de Brest). Information is at the heart of the research project of the unit; being by nature multimodal, complex, heterogeneous, shared and distributed, it is integrated by researchers into methodological solutions for improving medical care. Benefiting from a unit within the CHRU, the UMR (joint research unit) has (in addition to access to its own platforms) a privileged access to hospital technical platforms, as well as to all clinical data and patients, in a strong dynamic of translational research.

Research Topic

A key challenge in the EviRed project was integrating visual data from different imaging modalities (such as UWF-CFP, OCT, and WF-OCTA) along with contextual information gathered from ophthalmology departments (e.g., visual acuity) and diabetology departments (e.g., diabetes stability). We have already developed neural architectures capable of processing these multimodal data sources to assess the severity and progression of diabetic retinopathy (DR). Another challenge involved analyzing follow-up examinations to enhance the accuracy of DR severity and progression assessments. The goal was to leverage previous examinations alongside the current one. To achieve this, we developed time series analysis tools that can track and predict changes across consecutive examinations.

The postdoctoral position will take place during the final year of the EviRed project. A key responsibility will be integrating all AI developments made since the project's inception, including patented algorithms and software. This will involve exploring efficient methods to combine these algorithms while ensuring they remain independent of specific devices and populations. Additionally, visualization techniques will need to be explored to clearly highlight changes and early warning signs targeted by the AI in examination records. These two tasks will imply novel developments in the fields of domain generalization and explainability, with the potential to result in methodological publications. Close collaboration with our industrial partners (Evolucare Technologies and ZEISS) will be required for technology transfer, as well as working closely with clinicians to evaluate the technology and publish findings in high-impact journals.

Recent publications from the team and the consortium can be found on the EviRed website: https://evired.org/publications/

Keywords: deep learning, multimodal data analysis, longitudinal data analysis, explainable AI, domain generalization