Robot-assisted surgery, allowing surgeons to perform complex surgeries through tiny incisions, has been significantly increasing in popularity worldwide. However, surgical safety is still a major concern in the high-risk operating environment and remains a threat to the quality of surgical outcome. As global statistics, millions of surgeries per year would encounter safety-critical intraoperative adverse events, most of which were otherwise avoidable if the surgeon can be timely aware of the potential risks in operation. In this project, we aim to introduce smart context-awareness into robot- assisted surgery, by developing novel artificial intelligence techniques to provide automatic cognitive assistance for surgeons during critical moments of the procedure, in order to improve surgical safety and quality. The use case of this project will be robot-assisted hysterectomy, which is the most common gynecological procedure performed on women diagnosed with uterine fibroids or cervical cancer. Both Hong Kong and French teams will explore together innovative multimodal machine learning methods, based on available synchronized clinical video and kinematic data, which will be more advanced and clinically relevant than all existing methods that only used visual perception. Based on our pilot studies, we have identified a set of critical intraoperative scenarios to address avoidable adverse events in hysterectomy, such as injury of the pelvic ureter during both the coagulation of the uterus pedicle and adnexectomy. To achieve our goal, we will solve the following key challenges: 1) How to yield precise and real-time recognition of the surgical context, i.e., surgical workflow, operation actions, surgical instruments, anatomical tissues and the reconstructed 3D surgical environments. 2) How to conduct automatic assessment of the identified critical-context-of-safety (CCS), and further provide informed decision-making support to surgeons for their best practice to avoid safety risks. By a research collaboration between world-class teams with complementary expertise and already-available clinical and annotated data, the i-SaferS project will generate outputs that provide fundamentally new and generic solutions and impactful references to the field. The project outcomes will significantly contribute to the emerging field of intelligent robotic surgery, and further strengthen the leading competitiveness of both partners in this field nationally and internationally.

Each year in Europe, more than 300,000 newborns are born prematurely. Among them, very premature babies have a 10-20% risk of contracting late neonatal infections, with a high risk of mortality or neurobehavioural sequelae. However, the ability to quickly and accurately diagnose the risk of infection in premature newborns based on clinical assessment and laboratory blood tests remains a challenge. In a personalized and precision medicine approach, the proposed project aims to validate and deploy a new non-invasive medical decision support system for early diagnosis of infection in premature newborns. It aims to prepare a proposal for the call H2020 SC1-BHC-06-2020: Digital diagnostics - developing tools for supporting clinical decisions by integrating various diagnostic data from WP2020. Digi-Sepsis uses artificial intelligence methods to detect early infections. It builds on the experience and major progress made in the previous H2020 Digi-NewB project (www.digi-newb.eu) which ends in February 2020. This project resulted in the development of a proof of concept based on a multi-source database of more than 400 patients, an interface validated by ergonomic experts with the nursing staff, a validated and operational architecture for a randomized study, by carrying out a first phase of tests in hospital/neonatology with monitoring of the risk of sepsis. This approach aims to reduce morbidity and mortality associated with late neonatal infections. It requires continuous, real-time measurement of a set of clinical (medical record) and physiological (signal processing) variables and their association with recent and developing biological tools ("-omic", immunological biomarkers and PCR). The Digi-Sepsis project aims to meet the following specific objectives: The issues that will be addressed in the present study are: (i) To propose a definition of the early phase of sepsis which is necessary to evaluate the accuracy and predictive value of the proposed approach. (ii) To evaluate the accuracy of the proposed multi-dimensional approach in terms of sensitivity, specificity, predictive values and significance for the clinicians (iii) To demonstrate a benefice for the patients in the use of AI for the early diagnosis of infection before the occurrence of established sepsis, (iv) To evaluate the perception of the proposed approach by patients, parents and health care givers. In the long term, it is expected that new therapeutic strategies will be developed that combine early treatment with a reduction in the inappropriate use of antibiotics.

Nearly 35% of epileptic patients do not respond to antiepileptic drugs, one fourth of whom can benefit from resective surgery. This represents in Europe around 250 000 patients. The identification of the cortical areas to be removed (the Epileptogenic Zone, EZ), and the evaluation of their functional integrity (the Functional Zone, FZ), is a difficult process which requires intracranial EEG (iEEG) recordings in 25-50% of the cases. However, even when using such iEEG information, epilepsy surgery still fails in a substantial ratio of patients, and post-operative cognitive/psychiatric deficits still exist in a significant number of the cases. This means that iEEG criteria used for identifying epileptogenic and functionally eloquent brain tissues are not clearly determined nor understood. These recent years, emphasis has been put on High Frequency Oscillations (HFOs, 40-500Hz) as putative biomarkers of epileptogenicity or cognition in patients submitted to iEEG evaluation. Although very relevant, identification of HFOs require the use of signal analysis techniques and cognitive tasks that are mainly available as home-made softwares designed by researchers for their own line of research, without immediate clinical applications. Providing reliable HFOs-quantification tools in the clinical environment would undoubtedly have a large impact on the existing presurgical diagnostic procedures, and therefore on epileptic patient’s management. The general objective of the FORCE project is precisely to develop methods and recording technologies into a professional-looking system for optimal recording, detection and characterization of epileptic and cognitive HFOs in clinical routine, in order to improve accuracy and safety of epilepsy surgery. To this aim, FORCE is established on a profoundly interdisciplinary consortium composed of 7 partners, including three industrial companies, all highly considered and renowned in their fields of interest and that together provide the full range of competencies in medicine, cognitive neuroscience, information processing and technology necessary to complete the project. In practice, FORCE will start from the promising methods recently developed by our consortium, that allow to quantify fast activities at seizure onset (ictal HFOs) and between the seizures (interictal HFOs), as well as fast activities elicited by various cognitive tasks (Gamma Band Responses, GBRs). IEEG signals, visually analysed and validated by experts in a large cohort of epileptic patients (around 200), will be used to optimize the detection, quantification and visualization of HFOs and GBRs. Time-frequency methods will occupy a central place, and we will revisit the processing of iEEG signals for HFOs/GBRs visualization in a MRI-based environment. A new hybrid macro-microelectrode will be developed in order to assess the contribution of micro-electrodes for the recording of HFOs/GBRs. The validated tools will be integrated in a user-friendly software environment that will be incorporated in commercial EEG reviewing stations as readily-usable, CE-marked tools for the routine analysis of HFOs and GBRs. The clinical relevance of HFOs/GBRs quantification to localize the EZ and FZ will be evaluated by comparing these methods with the traditional ones (visual inspection, cortical electrical stimulation), with respect to surgical outcome. Only a handful of clinical centers worldwide have the technology and human expertise to study in routine human epilepsy and human cognition based on intracranially-recorded HFOs/GBRs. FORCE represents a major advance in the field, that also will reinforce the leadership of industrial partners based on the newly developed products which extend the capabilities of their already-existing systems and devices.

In the context of the growing interest in data-sharing in medicine, this project aims 1/ to perform systematic re-analyses of Randomised Controlled Trials (RCTs) in order to assess their reproducibility and 2/ to develop a tool to identify transparent and reproducible scientific practices. We will critically appraise all initiatives for data-sharing in medicine and identify all platforms that enable RCT data to be shared. A random sample of 62 RCTs will be selected and re-analysed. The results of these re-analyses will be compared with the results of the original analyses. The same method will be applied for all new drugs submitted to the European Medicine Agency in order to replicate results and to provide practical information for drug regulation. The impact of certain contextual factors on reproducibility will be explored. A scale for scoring the sharing and reproducibility of useful data will be developed. France is involved in the G7 initiative calling for open practices in Science. This project will provide empirical data on these practices and their usefulness.

Dormancy is an adaptive trait that is established during seed maturation and prevents seed germination on the parent plant or out of proper season after seed dispersal. It is also an important agronomic trait as germination before harvest (vivipary) is a major cause of crop yield losses. Abscisic acid (ABA) is the key phytohormone promoting dormancy whereas nitrate (NO3-) stimulates germination by triggering ABA catabolism. Partners 1 & 2 have previously identified a new whole MAPK (Mitogen-activated protein kinase) module which is activated by both ABA and NO3- in Arabidopsis plantlets. They have also recently shown that mutants impaired in this module produce seeds which are more dormant. Strikingly and coherently, mutations in homologous MAPK genes in wheat and barley were reported to reduce vivipary. Taken together, these preliminary results suggest that this MAPK module is a new player controlling seed dormancy conserved throughout angiosperms. The fact that this module is activated by both ABA and NO3- also suggests that it may have a pivotal role as integrator of signaling pathways controlling dormancy. This project aims to better characterize this module in the frame of seed germination using Arabidopsis as a model plant and to exploit the results to develop new strategies to manipulate crop germination in the field. To achieve these goals, the first WP will aim to functionally validate the module by unveiling where and when it is required to modulate seed dormancy and which are the kinases involved in this function, a MAPK module being composed of at least 3 kinases. Importantly, we will test how the MAPK signaling module is modulated by and/or modulates ABA and NO3- signaling by using a combination of biochemical and genetic approaches to study mutants impaired in these signaling cascades. Furthermore, the MAPK module presents unique features when compared to other plant and animal MAPK modules described so far. The second WP will thus be devoted to the characterization of these specificities and will particularly study the translational and post-translational regulations of the module as well as decipher the unknown function of protein domains in the central MAP2K. The third WP will focus on the downstream events that are regulated by the module. Firstly, we will identify substrates which are phosphorylated by MAPKs and are important to control seed dormancy. Secondly, we will unveil the cellular processes which are regulated by the module by performing transcriptomic and metabolomic studies of mutants impaired in the module. Finally, a fourth WP will aim to identify molecules targeting this MAPK module and use them as chemical probes to investigate to which extent, across the plant kingdom, this module is important for seed physiology and to modulate seed dormancy in crops. This project relies on the collaboration of 4 groups recognized as leaders in their respective fields, who will bring their expertise and skills to challenge the novel hypothesis that the recently discovered MAPK module integrates distinct environmental signaling pathways to trigger the downstream processes that determine whether seeds germinate or not. Their joint work will lead to a better understanding of the factors that control seed physiological traits and new essential knowledge to enhance resilience through advanced breeding programs and to provide guidelines for optimal seed production, treatment and storage. It will also use an original strategy based on chemical genetics in yeast aiming at the identification of small molecules that modify the activity of the MAPK module and modulate dormancy in model species and crops. Thus, the MAPKSEED project brings together multidisciplinary expertises to tackle an important issue for optimizing sustainable agriculture in a changing environment by novel and original basic research.