member

Previous page Daniel RACOCEANU university professor Team “Algorithms, Models and Methods for Images and Signals of the Human Brain ” http://daniraco.free.fr/short_bio.htm https://twitter.com/daniraco https://www.linkedin.com/in/daniraco/

Biography

Professor at Sorbonne University, and PI of the INRIA - ARAMIS team of the Paris Brain Institute (CNRS UMR 7225 - Inserm U 1127), I teach in the field of medical image analysis, pattern recognition and machine learning (including deep learning), I focus my research in the field of integrative digital anatomopathology, and in particular on the creation of new paradigms, protocols and technologies for grading and prognosis in histopathology, at the frontier between imaging and omics. Professor at Pierre et Marie Curie University between 2011 and 2018, I was 6 years Professor at the National University of Singapore from 2009 to 2015, as well as Chargé (2005-2011) and Directeur de Recherche CNRS (2011-2014), being also the Director (2008-2014) of the UMI CNRS 2955 IPAL (Image & Pervasive Access Lab) based in Singapore. Between 2014 and 2016, I was co-Director of the theranostics and cancer team of the Laboratoire l'Imagerie Bio-médicale (CNRS UMR 7371 - Inserm U1146), being also involved in the creation and the first steering committee of the Institut Universitaire d'Ingénierie en Santé (IUIS) of Sorbonne University. In Besançon, as Senior Lecturer at the University of Franche-Comté, and researcher at the FEMTO-ST Institute (1999-2005), I was involved in the European Project (FP5-ITEA - Proteus) and the Industrial Project NEMOSYS - DCNS. During this period, I worked on intelligent dynamic monitoring methods, developing new tools such as recurrent neural networks and predictive diagnosis methods using neuro-fuzzy systems. Business manager at General Electric France and logistics manager at Gaussin between 1997 and 1999, I defended my PhD thesis in 1997 at the UTBM and my Habilitation to Supervise Research in 2006 at the University of Franche-Comté. In terms of funded scientific collaborations, I led the ANR TecSan MICO project (Cognitive Microscopy for Breast Cancer grading - 2011-2014), being involved - between 2013 and 2016 - in the FUI (Fonds Uniques Interministériels) project FlexMIm, on collaborative telepathology based on semantic imaging, as well as in the A*STAR / JCO (Agency for Science, Technology And Research / Joint Council Office - Singapore) IMS project, dedicated to a stand-alone integrated 3D cellular microscopy system. In 2016, I obtained and participated in a European project (funded by EIT Health) entitled PAPHOS, focusing on the use of massive databases in Digital Anatomopathology. Member of the Advisory Board of the European Society for Digital Integrative Pathology (ES-DIP), I have been actively involved in its creation (2016) by being successively its Vice-President (2016-2018) and its President (2018-2020). For a period of 2 years (2016-2018), I was Professor at the Pontifical Catholic University of Peru, which allowed me to organize (General Chair) the international conference MICCAI 2020. Since 2018, I am a member of the Board of Directors of the international learned society MICCAI (Medical Image Computing & Computer Assisted Intervention).

Research work

Semantic Exploration and Understanding of Large High Content Microscopic Image Databases. Computational Integrative Histopathology. My work focuses on understanding, analysing, modelling and simulating the tumour microenvironment, using virtual slide images from histopathology (digital pathology) as a reference. By understanding and modelling the elements indexed by pathologists when analysing intra-tumour heterogeneity, we are currently able to build mathematical models capable of providing local and spatial quantification of processes, allowing a holistic approach to the pathological landscape. In cancers, the objective is to correlate and consolidate the phenotypic signature with an omics signature (integrative computational anatomopathology), within the tumour genesis process. The tools studied, developed and used in my research are mathematical morphology on sparse sets (simplicial complexes), scalable stochastic models (Markovian point processes), biomedical semantics and deep learning. My methodological concern is the combination of these tools (generative - discriminative), in a global traceable and explainable approach, capable of creating a breakthrough and accelerating the adoption of DL tools in the biomedical field. This work has led to a project funded by AVIESAN - ITMO Cancer, entitled MALMO (Mathematical Approaches to Modelling Metabolic Plasticity and Heterogeneity in Melanoma) dedicated to mathematical approaches to modelling metabolic plasticity and heterogeneity in melanoma. Furthermore, around my competences in the field of computational histopathology, a recent part of my work at the Brain Institute concerns the correlation between Tau proteins (neurites, tangles and plaques - known to be part of the symptoms of Alzheimer's disease) observed at the microscopic level in grey matter and the stratification of Alzheimer's disease patients. A BBT (Big Brain Theory) project entitled STRATIFIAD (Refining Alzheimer Disease Patients' stratification using interpretable AI) has been competitively funded around these topics for a period of 2 years (2021-2023).

Publications

1. Jiménez, G., Racoceanu, D. (2019). Deep Learning for Semantic Segmentation versus Classification in Computational Pathology: Application to mitosis analysis in Breast Cancer grading, Front. Bioeng. Biotechnol., 21 June 2019, doi: 10.3389/fbioe.2019.00145.
2. Zemouri, R., Zerhouni, N., Racoceanu, D. (2019). Deep Learning in the Biomedical Applications: Recent and Future Status, Appl. Sci., 9(8), 1526; doi:10.3390/app9081526.
3. Saha, M., Chakraborty, C., Racoceanu, D. (2018) Efficient Deep Learning Model for Mitosis Detection using Breast Histopathology Images, Computerized Medical Imaging and Graphics, 64:29-40. doi: 10.1016/j.compmedimag.2017.12.001, Epub 2017 Dec 16.
4. Ehteshami Bejnordi, B. et al. (2017). Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women with Breast Cancer, JAMA, 318(22): 2199-2210. doi:10.1001/jama.2017.14585.
5. Sirinukunwattana, K., Pluim, J., Chen, H., Qi, X., Heng, P-A., Bo, Y., Wang, L.Y., Matuszewski, B., Bruni, E., Sanchez, U., Böhm, A. Ronneberger, O., Ben Cheikh, B., Racoceanu, D., Philipp Kainz, P., Pfeiffer, M., Urschler, M., Snead, D., & Rajpoot, N. (2017). Gland Segmentation in Colon Histology Images: The GlaS Challenge Contest. Medical Image Analysis, 35: 489-502.