You are here: indexbrainstorming

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
brainstorming:index [2020/12/08 13:10]
nbuisard
brainstorming:index [2022/01/31 18:33] (current)
nbuisard
Line 19: Line 19:
  
 ===== Coordination ===== ===== Coordination =====
-  * [[:members:julien.berthelemy:​index| ​Julien Berthelemy]]+  * [[:members:john.baxter:​index| ​John Baxter]] 
 +  * [[:​members:​noemie.buisard:​index| Noémie Buisard]]
  
 ====== Agenda ====== ====== Agenda ======
 +====== Vos rendez-vous ======
 +
 +
 +===== Mardi 14 Janvier 2021 13:00-14:00 =====
 +Présentation : Tiphaine Casy  \\  ​
 +**"​Enregistrement des données climatiques : Earth blackbox" ​ \\ ** 
 +"Basé sur le modèle des boîtes noires aériennes permettant de comprendre les causes d'un crash, des chercheurs et une compagnie de communication Australienne ont imaginé une "Earth black box" (ou "​Boîte noire de la Terre"​). Le rôle de cette dernière serait l'​enregistrement quotidiens des mouvements climatiques (fonte des glaces, augmentation du climat,...) des dépenses militaires, de la croissance démographique et des contenus de différents médias. Sa conception permettrait un stockage de 30 à 50 ans dans le but d'​informer les générations futures sur les actions ou non-actions orientant la Terre sur son déclin. Cette idée est venue suite à l'​observation de la progression continue du réchauffement climatique et de ses impacts écologiques,​ économiques et humains. Les chiffres actuels correspondent à ceux prédis en 1972 dans une étude menée par le MIT ("The limits of growth"​) et tendent vers un déclin total de la Terre vers la fin du siècle s'il n'y a aucune modification sociétale." ​ \\  ​
 +
 +
 +===== Mardi 21 Décembre 2021 13:00-14:00 =====
 +Présentation : Nicolas Vallée ​ \\  ​
 +**Présentation du projet CUFFSIM** ​
 +
 +===== Mardi 14 Décembre 2021 13:00-14:00 =====
 +Présentation : Enora Giffard ​ \\  ​
 +**Le stress post traumatique ​ \\ ** 
 +
 +===== Mardi 30 Novembre 2021 13:00-14:00 =====
 +Présentation : Tamás Haidegger ​ \\  ​
 +For the Noon sharing Tamás Haidegger, associate professor at Óbuda University, serving as the director of the University Research and Innovation Center (EKIK), and as the technical lead of medical robotics research at the Antal Bejczy Center for Intelligent Robotics.
 +His main field of research is on medical technologies,​ control/​teleoperation of surgical robots, image-guided therapy and digital health technologies. " ​ \\  ​
 +
 +
 +===== Mardi 23 Novembre 2021 13:00-14:00 =====
 +Présentation : Arnaud Huaulmé ​ \\  ​
 +"​Synapse is a platform for supporting scientific collaborations centered around shared biomedical data sets. Our goal is to make biomedical research more transparent,​ more reproducible,​ and more accessible to a broader audience of scientists. Synapse serves as the host site for a variety of scientific collaborations,​ individual research projects, and DREAM challenges." ​ \\  ​
 +
 +
 +===== Mardi 02 Novembre 2021 13:00-14:00 =====
 +Présentation : Alfonso Estudillo-Romero ​ \\  ​
 +"​Voxel-Based Diktiometry - Merging convolutional neural networks with voxel-based analysis for characterization of diffusion MRI for Parkinson’s disease." ​ \\  ​
 +
 +
 +===== Mardi 19 Octobre 2021 =====
 +Michaël Nguyen - Répétition de soutenance de stage  \\  14h : Delphine Tanguy - Répétition de soutenance de thèse
 +
 +===== Mardi 12 Octobre 2021 13:00-14:00 =====
 +Présentation : Alexandre Tronchot ​ \\  ​
 +**"​MENISIM,​ Simulation VR pour apprendre les bases de l'​arthroscopie : Retour sur les travaux de l'​année écoulée" ​ \\ ** 
 +
 +Voilà 1 an que le projet MENISIM a démarré (projet majeur de ma thèse de sciences). L'​objectif de ce Noon Sharing sera de vous faire un rappel sur l'​intérêt de ce travail dans la formation pratique à l'​arthroscopie des chirurgiens orthopédistes,​ une description de la méthodologie employée, et un retour sur les premiers résultats. Je conclurais cette présentation par les perspectives du projet, et une brève explication du projet de master 2 de Nicolas qu'il présentera de façon plus détaillée en fin d'​année.
 +
 +===== Mardi 05 Octobre 2021 13:00-14:00 =====
 +Présentation : Marie Le Duff  \\  ​
 +**"Les représentations sociales du changement climatique et des actions pro-environnementales du Plateau des Petites roches" ​ \\ ** 
 +
 +
 +===== Mardi 14 septembre 2021 13:00-14:00 =====
 +Présentation : Sonia Guérin ​ \\  ​
 +**Répétition de soutenance de Master : "​Outils d'​analyse d'une chirurgie : Hystérectomie robot-assitée" ​ \\ ** 
 +
 +
 +===== Mardi 07 septembre 2021 13:00-14:00 =====
 +Présentation : John Baxter ​ \\  ​
 +**Présentation de l'​expérience IHM (interaction homme-machine) sur un outil de segmentation du STN par l'​apprentissage automatique. ​ \\ ** 
 +
 +
 +===== Mardi 31 août 2021 13:00-14:00 =====
 +Répétition de soutenance de stage de fin d'​étude ​ \\  ​
 +Présentation : Enora Giffard
 +
 +
 +===== Mardi 15 Juin 2021 13:00-14:00 =====
 +
 +==== "​CARS2021 - Adapting the Listening Time for Micro-Electrode Recordings in Deep Brain Stimulation Interventions"​ ====
 +Présentation : Thibault Martin
 +
 +Abstract :
 +
 +" Deep Brain Stimulation (DBS) is a common treatment for a variety of neurological disorders which involves the precise placement of electrodes at particular subcortical locations such as the subthalamic nucleus. This placement is often guided by auditory analysis of micro-electrode recordings (MERs) which informs the clinical team as to the anatomic region in which the electrode is currently positioned. Recent automation attempts have lacked flexibility in terms of the amount of signal recorded, not allowing them to collect more signal when higher certainty is needed or less when the anatomy is unambiguous. Therefore, we have addressed this problem by evaluating a simple algorithm that allows for MER signal collection to terminate once the underlying model has sufficient confidence. We have parameterized this approach and explored its performance using three underlying models composed of one neural network and two Bayesian extensions of said network. Our results shown that one particular configuration,​ a Bayesian model of the underlying network'​s certainty, outperforms the others and is relatively insensitive to parameterization. Further investigation shows that this model also allows for signals to be classified earlier without increasing the error rate. This work presents a simple algorithm that records the confidence of an underlying neural network, thus allowing for MER data collection to be terminated early when sufficient confidence is reached. This has the potential to improve the efficiency of DBS electrode implantation by reducing the time required to identify anatomical structures using MERs. "
 +
 +===== Mardi 8 Juin 2021 13:00-14:00 =====
 +
 +==== " Robot-assisted laparoscopic hysterectomy:​ experts vs juniors " ====
 +Présentation : Sonia Guérin
 +
 +===== Mardi 1er Juin 2021 13:00-14:00 =====
 +
 +==== " Study of anatomical and clinical correlation in tractography after deep brain stimulation of the anterior nucleus of the thalamus (ANT) for focal, inoperable and drug resistant epilepsy. " ====
 +Présentation : Mathilde Gaudian
 +
 +===== Mardi 25 Mai 2021 13:00-14:00 =====
 +
 +==== " Présentation des résultats du stage " ====
 +Présentation : Tiphaine Casy
 +
 +===== Mardi 18 Mai 2021 13:00-14:00 =====
 +
 +==== " REPEAT - Simulation par Réalité Virtuelle : utilisation et apport dans l'​apprentissage de compétences non techniques dans la formation de l'​interne en neurochirurgie " ====
 +Présentation : Julien Tinois
 +
 +===== Mardi 11 Mai 2021 13:00-14:00 =====
 +
 +==== " Multimodal investigation of epileptic network, insight from electrophysiology and hemodynamic processes " ====
 +Présentation : Christophe Grova (PhD, Associate Professor, Physics Dpt, Concordia University ​ and  Adjunct Professor in Biomedical Engineering,​ and Neurology and Neurosurgery Dpt at McGill University)
 +
 +===== Mardi 20 avril 2021 13:00-14:00 =====
 +
 +==== "Le numérique responsable"​ ====
 +Présentation : Pierre Jannin
 +
 +===== Mardi 06 avril 2021 13:00-14:00 =====
 +
 +==== "​Bénéfice de l'​état de flow sur l'​efficacité des distractions dans la prise en charge de l’anxiété préopératoire en anesthésie pédiatrique"​ ====
 +Présentation : Camille Tordet
 +
 +"Les méthodes de distraction (jouet, dessin animé, jeu vidéo) ont été évaluées comme efficace dans la prise en charge de l’anxiété préopératoire des enfants. Les activités distractrices détournent l’attention des enfants des indices menaçants de la situation vers des activités et des pensées plus agréables. Toutefois, la variabilité d’efficacité de cette stratégie s’avère peu étudiée. En effet, si le bénéfice de la méthode de distraction s’explique par le transfert des ressources attentionnelles de la situation anxiogène vers l’activité distractrice,​ alors on devrait observer une variabilité d’efficacité selon le degré de ressources attentionnelles alloué sur l’activité. Ce travail cherche à analyser comment l’état de flow (état de concentration et d'​absorption intense) peut fournir des explications,​ dans le but d’améliorer les distractions actuellement utilisées. ​
 +Cette présentation abordera le contexte puis l’étude que nous avons menée au CHU de Rennes Sud, afin d’apporter des éléments de réponse à ces questions."​
 +
 +
 +===== Mardi 30 Mars 2021 13:00-14:00 =====
 +
 +==== "Brain correlates of behavioural disinhibition in behavioural variant frontotemporal dementia."​ ====
 +Présentation : Delphine Tanguy
 +
 +"​Disinhibition is a core symptom in neurodegenerative diseases, in particular in fronto-temporal dementia (FTD), and a major cause of stress for caregivers. However, the concept of this symptom is hard to disentangle:​ while a distinction between behavioural and cognitive disinhibition is common, an operational definition of behavioural disinhibition remains unclear. Therefore, available assessments of these behavioural disorders are imprecise and the corresponding neuroanatomical correlates are non-univocal.Thus,​ we aimed to characterise the anatomical networks underlying different subtypes of behavioural disinhibition. For this purpose, we used an original semi-ecological task assessing two specific dimensions of behavioural disinhibition,​ such as compulsivity and social disinhibition,​ in 17 behavioural variant of FTD patients and 18 healthy controls, associated with classic neuropsychological tests and brain imaging (voxel-based morphometry for the grey matter, tractography for the white matter). Compulsivity correlated with right temporal regions atrophy and with alterations of the bilateral cingulum and uncinate fasciculus, the right inferior longitudinal fasciculus and the right arcuate fasciculus. Social disinhibition was associated to huge bilateral frontal atrophy and impairments in the cingulum, the forceps minor and the left uncinate fasciculus. By combining ecological behavioural analyses with neuroanatomy,​ we identified two subtypes of behavioural disinhibition associated with differential neural networks."​
 +
 +===== Mardi 23 Février 2021 13:00-14:00 =====
 +
 +==== " Automatic Cortical Target Point Localisation in MRI for Transcranial Magnetic Stimulation via a Multi-Resolution Convolutional Neural Network " ====
 +Présentation : John S H Baxter
 +
 +" Transcranial Magnetic Stimulation (TMS) is a growing therapy for a variety of psychiatric and neurological disorders that arise from or are modulated by cortical regions of the brain represented by singular 3D target points. These target points are often determined manually with assistance from a pre-operative T1-weighted MRI, although there is growing interest in automatic target point localisation using an atlas. However, both approaches can be time-consuming which has an effect on the clinical workflow and the latter does not take into account patient variability such as the varying number of cortical gyri where these targets are located. This paper proposes a multi-resolution convolutional neural network for point localisation in MR images for a priori defined points in increasingly finely resolved versions of the input image. This approach is both fast and highly memory efficient, allowing it to run in high-throughput centres, and has the capability of distinguishing between patients with high levels of anatomical variability. Preliminary experiments have found the accuracy of this network to be 7.26+5.30 mm, compared to 9.39+4.63 mm for deformable registration and 6.94+5.10 mm for a human expert. The human-level performance of this network indicates that it can improve TMS planning by automatically localising target points in seconds, avoiding more time-consuming registration or manual point localisation processes. " ​
 +
 +===== Mardi 16 Février 2021 13:00-14:00 =====
 +
 +==== " PETRAW: PEg TRAnsfer Workflow recognition by different modalities " ====
 +Présentation : Arnaud Huaulmé
 +
 +" Automatic and online recognition of surgical workflows is mandatory to bring computer-assisted surgery (CAS) applications inside the operating room. According to the type of surgery, different modalities could be used for workflow recognition. In the case of robotic-assisted surgeries and virtual reality training sessions, video and kinematic data are easily available. However, even if these modalities are available, numerous methods on state of art only focus on one of them. Last year we submitted the MISAW sub-challenge as part of EndoVis at MICCAI2020 (challenge description and results available at https://​www.synapse.org/​MISAW,​ paper under writing), offering to participant to combine both modalities for the workflow recognition. To the best of our knowledge, there are very few studies about the added value to combine multiple modalities. Whereas, some study as demonstrate that the addition of external information improve the recognition,​ as the presence of tools
 +
 +PETRAW is a sub-challenge of ENDOVIS 2021 currently on minor revision for MICCAI2021. PETRAW focuses on the comparison of 3 modalities (video, kinematic, and segmentation) and their combination on surgical workflow recognition. On this noon sharing, I will present the challenge design and probably partial results." ​
 +
 +===== Mardi 9 Février 2021 13:00-14:00 =====
 +
 +==== " Preliminary results on the characterization of PD from DT images on the PPMI dataset by means of CNNs" ====
 +Présentation : Alfonso Estudillo Romero
 +
 +" We present our preliminary results on the analysis of diffusion tensor images (DTI) by means of a Convolutional Neural Network (CNN) to classify Parkinsonian patients (PD) and healthy controls (HC) from brain images. ​ An analysis of the whole population registered in a common space is proposed to compute some of the classical anisotropy measures in a voxel-based approach. Moreover, we retrieved the back-propagated gradients of these images through the CNN to understand how sensitive the CNN is to some anisotropy measures at some particular regions. " ​
 +
 +===== Mardi 2 Février 2021 13:00-14:00 =====
 +
 +==== " Extending Deep Neural Networks with Bayesian Inference and Recurrent Layers for Localizing the Subthalamic Nucleus from Micro-Electrode Recordings"​ ====
 +Présentation : Thibault Martin
 +
 +" Deep brain stimulation (DBS) is an interventional treatment for Parkinson'​s disease which involves the precise positioning of stimulation electrodes within particular subcortical anatomy. Although originally identified via imaging, additional inter-operative guidance is necessary to localize the target anatomy. Analysis of Micro-Electrode Recordings (MERs) allows for a trained neurophysiologist to infer the underlying anatomy at a particular electrode position using human audition, although it is subjective and requires a high degree of expertise. This project aims to develop and validate a deep neural network to assist in MER analysis during DBS which can use arbitrary length MERs while providing feedback to the neurophysiologist as to the model'​s certainty. We evaluate six different deep learning based classifiers which can use arbitrary length MERs for Sub-Thalamic Nucleus (STN) segmentation,​ finding a particular Baysian network to perform the best. With a view to providing decision support, the use of Bayesian inference has also made it possible to interpret the level of confidence in the predictions underlying the predictive model. This work represents a step forward in integrating automated analysis of MERs into the DBS surgical workflow by automatically finding and exploiting MER in a data-driven predictive model. " ​
 +
 +===== Mardi 5 Janvier 2021 13:00-14:00 =====
 +
 +==== " Place de l’Apprentissage de la Chirurgie Arthroscopique utilisant la Simulation en Réalité Virtuelle (VR) à propos d’un Geste : la Méniscectomie - MENISIM-VR"​ ====
 +Présentation : Alexandre Tronchot
 +
 +"Les dernières années ont été marquées par d'​importantes avancées dans la formation initiale des jeunes chirurgiens avec du temps alloué supplémentaire et de nouveaux outils pédagogiques. Toutefois, le quotidien reste essentiellement occupé par le temps de service. Paradoxalement,​ les internes vont être mis en responsabilité plus vite avec un temps passé à l'​hôpital plus court. La formation initiale doit donc à la fois assurer sa mutation et rester efficiente en trouvant des compléments et alternatives au traditionnel compagnonnage. La simulation est l'un des outils proposés pour cela, et elle a déjà fait ses preuves dans le domaine comportemental. Son évolution vers la réalité virtuelle (VR) est maintenant proposée pour la formation procédurale car elle offre un outil à la frontière entre réalités théorique et pratique et, surtout, inlassablement reproductible. La chirurgie arthroscopique est particulièrement adaptée à cette simulation VR du fait de son retour visuel per-opératoire sur écran 2D. Toutefois, et bien qu'il y ait de plus en plus d'​articles scientifiques parlant de l'​intérêt de la simulation VR en chirurgie arthroscopique,​ peu rapportent sa validité lors de la transposition des performances du simulateur vers le patient réel. Montrer que l'​étudiant en progressant sur simulateur devient également performant au bloc opératoire serait une avancée considérable dans la formation médicale du troisième cycle, correspondant parfaitement aux demandes des Ministères de la Santé et de l'​Enseignement Supérieur et la Recherche à la fois en termes de sécurité patient ("​Jamais la première fois sur le patient"​),​ mais aussi de sécurité du praticien et d'​enjeux économiques de formation." ​
 +
 +===== Mardi 8 Décembre 2020 13:00-14:00 =====
 +
 +==== "​Hystérectomie : indications,​ modalités et apports de la chirurgie robot assistée."​ ====
 +Présentation : Sonia Guerin
 +
 +"​Hysterectomy is the most commonly performed gynecological procedure in high income countries. Eight-seven percent of indications are benign diseases. Several surgical approaches are possible, each with its own advantages and disadvantages. With the arrival of the surgical robot, it is now possible for us to study precisely the gestures performed by the surgeon thanks to kinematics and video data analysis.
 +Initially, we will look at the indications,​ approaches and complications of hysterectomy before discussing the contribution of kinematic and video data in general surgery and in hysterectomy in particular." ​
 +
 +===== Mardi 1er Décembre 2020 13:00-14:00 =====
 +
 +==== "A semantic database for integrated management of image and dosimetric data in low radiation dose research in medical imaging"​ ====
 +Présentation : Bernard Gibaud
 +
 +"​Medical ionizing radiation procedures and especially medical imaging are a non negligible source of exposure to patients. Whereas the biological effects of high absorbed doses are relatively well known, the effects of low absorbed doses are still debated. This work presents the development of a computer platform called Image and Radiation Dose BioBank (IRDBB) to manage research data produced in the context of the MEDIRAD project, a European project focusing on research on low doses in the context of medical procedures. More precisely, the paper describes a semantic database linking dosimetric data (such as absorbed doses to organs) to the images corresponding to X-rays exposure (such as CT images) or scintigraphic images (such as SPECT or PET images) that allow measuring the distribution of a radiopharmaceutical.
 +
 +The main contributions of this work are: 1) the implementation of the semantic database of the IRDBB system and 2) an ontology called OntoMEDIRAD covering the domain of discourse involved in MEDIRAD research data, especially many concepts from the DICOM standard modelled according to a realist approach."​
 +
 +
 +===== Mardi 10 Novembre 2020 13:00-14:00 =====
 +
 +==== "​Flexible recurrent neural network for Audio-Visual sequence analysis"​ ====
 +Présenté par : Tsiry Mayet
 +
 +"Deep learning models are now brought closer to the user rather than running in a distant cloud, helping to reduce latency, network congestion, and improving data security and privacy. However, smartphones and user devices impose additional constraints such as limited computation or energy. Handling these constraints has become an active research topic. User devices can also host multiple processes running at the same time and starting or stopping abruptly, modifying the constraints affecting the processes. Few works have considered models that can be modified at run time to adapt to an evolving computational limit." ​
  
 ===== Mardi 20 Octobre 2020 13:00-14:00 ===== ===== Mardi 20 Octobre 2020 13:00-14:00 =====
Line 32: Line 211:
 ==== "MISAW - Part of the Endoscopic Vision Challenge"​ ==== ==== "MISAW - Part of the Endoscopic Vision Challenge"​ ====
 Présenté par : Arnaud Huaulmé Présenté par : Arnaud Huaulmé
 +
 +"​Automatic and online recognition of surgical workflows is mandatory to bring computer-assisted surgery (CAS) applications inside the operating room. According to the type of surgery, different modalities could be used for workflow recognition. In the case where the addition of multiple sensors is not possible, the information available for manual surgery is generally restricted to video-only. In the case of robotic-assisted surgery, kinematic information is also available. It is expected that multimodal data would make easier automatic recognition methods.
 +
 +The “MIcro-Surgical Anastomose Workflow recognition” (MISAW) sub-challenge provides a unique dataset for online automatic recognition of surgical workflow by using both kinematic and stereoscopic video information on a micro-anastomosis training task. Participants are challenged to recognize online surgical workflow at different granularity levels (phases, steps, and activities) by taking advantage of both modalities available. Participants can submit results for the recognition of one or several granularity levels. In the case of several granularities,​ participants are encouraged (but not required) to submit the result of a multi-granularity workflow recognition,​ i.e. recognize different granularity levels thanks to a unique model."​
  
 ===== Mardi 22 Septembre 2020 13:00-14:00 ===== ===== Mardi 22 Septembre 2020 13:00-14:00 =====
inserm rennes1 ltsi