Short- or mid-term rainfall forecasting is a major task with several environmental applications such as agricultural management or flood risk monitoring. Existing data-driven approaches, especially deep learning models, have shown significant skill at this task, using only rainfall radar images as inputs. In order to determine whether using other meteorological parameters such as wind would improve forecasts, we trained a deep learning model on a fusion of rainfall radar images and wind velocity produced by a weather forecast model. The network was compared to a similar architecture trained only on radar data, to a basic persistence model and to an approach based on optical flow. Our network outperforms by 8% the F1-score calculated for the optical flow on moderate and higher rain events for forecasts at a horizon time of 30 min. Furthermore, it outperforms by 7% the same architecture trained using only rainfall radar images. Merging rain and wind data has also proven to stabilize the training process and enabled significant improvement especially on the difficult-to-predict high precipitation rainfalls.

Around 30% of patients with RA have an inadequate response to MTX. We aimed to use routine clinical and biological data to build machine learning models predicting EULAR inadequate response to MTX and to identify simple predictive biomarkers. Models were trained on RA patients fulfilling the 2010 ACR/EULAR criteria from the ESPOIR and Leiden EAC cohorts to predict the EULAR response at 9 months (± 6 months). Several models were compared on the training set using the AUROC. The best model was evaluated on an external validation cohort (tREACH). The model's predictions were explained using Shapley values to extract a biomarker of inadequate response. We included 493 therapeutic sequences from ESPOIR, 239 from EAC and 138 from tREACH. The model selected DAS28, Lymphocytes, Creatininemia, Leucocytes, AST, ALT, swollen joint count and corticosteroid co-treatment as predictors. The model reached an AUROC of 0.72 [95% CI (0.63, 0.80)] on the external validation set, where 70% of patients were responders to MTX. Patients predicted as inadequate responders had only 38% [95% CI (20%, 58%)] chance to respond and using the algorithm to decide to initiate MTX would decrease inadequate-response rate from 30% to 23% [95% CI: (17%, 29%)]. A biomarker was identified in patients with moderate or high activity (DAS28 > 3.2): patients with a lymphocyte count superior to 2000 cells/mm³ are significantly less likely to respond. Our study highlights the usefulness of machine learning in unveiling subgroups of inadequate responders to MTX to guide new therapeutic strategies. Further work is needed to validate this approach.

Objectives Around 30% of patients with rheumatoid arthritis (RA) do not respond to tumour necrosis factor inhibitors (TNFi). We aimed to predict patient response to TNFi using machine learning on simple clinical and biological data. Methods We used data from the RA ESPOIR cohort to train our models. The endpoints were the EULAR response and the change in Disease Activity Score (DAS28). We compared the performances of multiple models (linear regression, random forest, XGBoost and CatBoost) on the training set and cross-validated them using the area under the receiver operating characteristic curve (AUROC) or the mean squared error. The best model was then evaluated on a replication cohort (ABIRISK). Results We included 161 patients from ESPOIR and 118 patients from ABIRISK. The key selected features were DAS28, lymphocytes, ALT (aspartate aminotransferase), neutrophils, age, weight, and smoking status. When predicting EULAR response, CatBoost achieved the best performances of the four tested models. It reached an AUROC of 0.72 (0.68–0.73) on the train set (ESPOIR). Better results were obtained on the train set when etanercept and monoclonal antibodies were analysed separately. On the test set (ABIRISK), these models respectively achieved on AUROC of 0.70 (0.57–0.82) and 0.71 (0.55–0.86). Two decision thresholds were tested. The first prioritised a high confidence in identifying responders and yielded a confidence up to 90% for predicting response. The second prioritised a high confidence in identifying inadequate responders and yielded a confidence up to 70% for predicting non-response. The change in DAS28 was predicted with an average error of 1.1 DAS28 points. Conclusion The machine learning models developed allowed predicting patient response to TNFi exclusively using data available in clinical routine.

Short- or mid-term rainfall forecasting is a major task with several environmental applications such as agricultural management or flood risk monitoring. Existing data-driven approaches, especially deep learning models, have shown significant skill at this task, using only rainfall radar images as inputs. In order to determine whether using other meteorological parameters such as wind would improve forecasts, we trained a deep learning model on a fusion of rainfall radar images and wind velocity produced by a weather forecast model. The network was compared to a similar architecture trained only on radar data, to a basic persistence model and to an approach based on optical flow. Our network outperforms by 8% the F1-score calculated for the optical flow on moderate and higher rain events for forecasts at a horizon time of 30 min. Furthermore, it outperforms by 7% the same architecture trained using only rainfall radar images. Merging rain and wind data has also proven to stabilize the training process and enabled significant improvement especially on the difficult-to-predict high precipitation rainfalls.

Depuis les années 1920, les cheminots bénéficient d’une réputation de combativité dans l’imaginaire collectif de la société française. En novembre 1974, les postiers ont engagé une grève illimitée et générale depuis octobre. La conjoncture laisse penser qu’ils ont tout intérêt à élargir leur mouvement, en premier lieu en direction des cheminots. Un mot d’ordre de grève circule effectivement à la SNCF début novembre 1974, mais le mouvement ne « prend » pas. Il s’agit donc pour nous de comprendre l’image des cheminots qui s’est formée dans la société pendant et à l’issue de la grève des postiers. Cette grève et le mouvement raté qui s’en est suivi à la SNCF ont-ils eu des répercussions sur les représentations de la combativité cheminote ? Comment caractériser les attentes qui se fixent sur les cheminots pendant la grève des postiers ? Et quelles sont les réactions à l’échec d’un mouvement cheminot en novembre 1974 ? Il doit être replacé dans le début du reflux du mouvement ouvrier et de ses organisations politiques et syndicales dans les années 1970 et 1980. Mais si novembre 1974 est une parenthèse, on ne peut pas dire avec certitude que cet épisode ait ébranlé la réputation, la perception de la combativité des cheminots dans la société.

Drug development is a lengthy and high-risk process, with most investigational drug candidates failing in phase II randomized clinical trials (RCT) due to insufficient efficacy. It makes early prediction of trial outcomes crucial for reducing attrition and guiding strategic decisions, especially in immunology and inflammation (I&I) diseases. Herein, we present EVA, the first pre-trained foundation model in complex inflammatory diseases tailored to support drug development. EVA learns generalizable patterns from large-scale data of cell biology and immunology, enabling superior predictive performance and generalization compared to traditional approaches. EVA is pre-trained on tens of millions of single-cell RNA-seq samples and tens of thousands of bulk RNA-seq samples from I&I diseases patients, enabling it to learn disease-relevant transcriptomic patterns in this therapeutic area. By fine-tuning EVA in few-shot settings on both preclinical (mouse) and clinical (human) data and harnessing its wide pre-training knowledge, EVA predicts drug responses in I&I with high precision at both cohort and patient levels, as illustrated by accurate forecasting of anti-TNF therapeutic activity in ulcerative colitis. By deciphering its decision process, we further highlight that EVA’s ability to stratify patients based on predicted drug response can also be leveraged to discover drug response biomarkers as early as preclinical stages. EVA’s applications in precision immunology encompass therapeutic target validation prior to clinical entry, identification of patient subpopulations most likely to benefit from treatment, and comparative efficacy analysis against competitor compounds. EVA’s versatility makes it an invaluable tool for strategic decision-making throughout the drug development pipeline: by leveraging it to prioritize the most promising drug candidates and optimize RCT designs, it can contribute to reduce late-stage failures and accelerate the delivery of effective therapies. Overall, this work represents a significant advancement in utilizing a pre-trained foundation model for precision drug development in complex inflammatory diseases. Graphical EVA is a pre-trained foundation model specific to immune-mediated inflammatory diseases. It enables the prediction of therapeutic efficacy in patients leveraging data from preclinical disease models.