Image Processing and Machine Learning for Bone Disease Prediction
Location: Research platform, Regional Hospital of Orléans, France
Laboratory: I3MTO - EA 4708, University of Orléans, France
Duration: 12 months – position to be filled by the end of 2019
Salary: 45K€ (~2150€/month after tax)
# I3MTO - EA 4708 (University of Orléans)
The I3MTO laboratory is an academic team at Orléans University that focuses on osteo-articular pathologies, particularly osteoarthritis and osteoporosis. This team is organized in two axes, one “imaging and modeling” (engineering and signal processing, CNU 61) and the other “cellular and molecular biology” (biology, CNU 64).
# PRIMMO (Hospital of Orléans)
The Orleans Mutualized Medical Innovation Research Platform, PRIMMO, brings together in the same place (CHR Orléans), fundamental research laboratories of the Grand Campus, the CNRS and the CHR with a strong industrial partnership. PRIMMO aims to promote and federate fundamental and clinical research skills, and will enable all researchers, doctors and industrialists to continue fundamental research and support pre-clinical research.
Osteoarthritis (OA) is the most common disorder of the musculoskeletal system and the major cause of reduced mobility among seniors. It is now considered as a disease of the whole joint organ involving the articular cartilage, subchondral bone and synovial membrane but also the menisci and ligaments. However, the underlying mechanisms through which this debilitating disease occurs and progresses have not been fully elucidated yet. Moreover, there is still no medical treatment for this pathology, and the lack of predictive biomarkers is a major obstacle to their development.
The past years have shown many studies based on the computer-aided diagnosis for knee osteoarthritis, and more recently the prediction of their evolution. Several research teams around the world have proposed their own methods for imaging markers extraction, but yet no clinical tools have emerged for knee osteoarthritis routine evaluation. Also, the development of high-end imaging modalities (high resolution MRI, QCT, ...) highlighted deep features that were not systematically transposed to the highly available modalities.
The candidate will join the engineering and signal processing axis of the I3MTO which is interested in the combination of texture analysis and artificial intelligence for osteoarthritis prediction. The project has already been initiated and has resulted in several publications.
The objective of this recruitment is to fully automate the process and validate these models on a different database (MOST).A secondary objective is to investigate relations between these models 2D features and the actual changes in the 3D trabecular network.
The candidate will be associated to the ongoing STUDIUM consortium Knee Osteoarthritis Predictive Imaging gathering international experts and will have to interact with teams around the globe.
PhD in computer sciences, image or signal processing, applied mathematics or related and published paper(s) as first author. The candidate must have knowledge/experience in computer programming and technical skills in several of the following areas of expertise: machine learning, texture analysis, data sciences, statistical modeling. The candidate must be able to take initiatives and conduct his/her own scientific research in a multidisciplinary research environment.
Interested candidates should send a CV, a cover letter as well as the names and contact details of references to:
Dr. Eric Lespessailles <email@example.com>.
 Janvier, T. et al. Subchondral tibial bone texture analysis predicts knee osteoarthritis progression: data from the Osteoarthritis Initiative. Osteoarthr. Cartil. 25, 259–266 (2017).
 Riad, R. et al. Texture analysis using complex wavelet decomposition for knee osteoarthritis detection: Data from the osteoarthritis initiative. Comput. Electr. Eng. 68, 181–191 (2018).
 Brahim, A. et al. A Decision Support Tool For Early Detection of Knee OsteoArthritis using X-ray Imaging and Machine Learning: Data from the OsteoArthritis Initiative. Comput. Med. Imaging Graph. (2019).
 Tiulpin, A. et al. Multimodal Machine Learning-based Knee Osteoarthritis Progression Prediction from Plain Radiographs and Clinical Data. (2019).
 Tiulpin, A., Thevenot, J., Rahtu, E., Lehenkari, P. & Saarakkala, S. Automatic Knee Osteoarthritis Diagnosis from Plain Radiographs: A Deep Learning-Based Approach. Sci. Rep. (2017).
 Joseph, G. B. et al. Tool for osteoarthritis risk prediction (TOARP) over 8 years using baseline clinical data, X-ray, and MRI: Data from the osteoarthritis initiative. J. Magn. Reson. Imaging 47, 1517–1526 (2018).
 Podsiadlo, P. et al. Baseline trabecular bone and its relation to incident radiographic knee osteoarthritis and increase in joint space narrowing score: directional fractal signature analysis in the MOST study. Osteoarthr. Cartil. (2016). doi:10.1016/j.joca.2016.05.003
 Woloszynski, T., Podsiadlo, P., Stachowiak, G. & Kurzynski, M. A dissimilarity-based multiple classifier system for trabecular bone texture in detection and prediction of progression of knee osteoarthritis. Proc Inst Mech Eng H 226, 887–894 (2012).
(c) GdR 720 ISIS - CNRS - 2011-2019.