Predict (France)

*Correspondence: Sijung Yun, syun16@jhu.edu; Youngmi Ji, charmgene@gmail.comThe integration of multi-omics biological data, such as genomics, transcriptomics, proteomics, etc., is reshaping how we conceptualize and pursue biomarker discovery in endocrinology. Moving beyond reductionist paradigms, contemporary research now unites molecular, cellular, physiological, and population-level information to illuminate the complex regulatory architecture underlying endocrine health and disease. Integrated Diagnostics and Biomarker Discovery in Endocrinology and Biomedical Sciences: Volume II brings together nine original contributions that exemplify this transition toward a systems-oriented and data-driven discipline.Spanning the spectrum from ionic ratios and proteomic signaling to transcriptomic networks, genomic variation, and ecological microbiome interactions, these studies demonstrate how diagnostic precision emerges through the convergence of molecular and systemic perspectives. Collectively, they trace a coherent trajectoryfrom basal biochemistry and molecular communication to clinical integration and population-scale modeling -illustrating how multi-scale data synthesis from ionic ratios to networks can refine both mechanistic understanding and translational application.Taken together, this collection reflects the growing maturity of integrative endocrinology, a field where multi-omics analytics, causal inference, and real-world data harmonization converge to enable predictive and personalized approaches to endocrine disorders. By highlighting these multi-scale insights, Volume II underscores the central message of modern biomarker science: meaningful diagnostic innovation arises not from any single data layer, but from their integration into a unified systems framework that connects molecules to medicine. Lou et al. (2025) systematically validated the Glucose-Potassium Ratio (GPR) -a longrecognized broad clinical predictive marker 1 -as a prognostic biomarker for both shortand long-term all-cause mortality. They showed a strong association with mortality in both hospital and ICU settings. Mortality risk escalated sharply when GPR exceeded this threshold. Sensitivity analyses confirmed the robustness of these findings, positioning GPR as a valuable, non-invasive indicator for early identification and risk stratification of high-risk sepsis patients. This study opens the collection by illustrating that integrated diagnostics can arise not only from macromolecular data but from fundamental ionic interactions reflecting systemic metabolic homeostasis. Together, these nine contributions delineate a rapidly expanding frontier in integrated diagnostics, spanning the full continuum of biological organization-from ionic ratios and proteomic signatures to genomic databases and microbiome-derived ecological biomarkers. Collectively, they illustrate how endocrine science is evolving from isolated molecular characterization toward a fully systems-based discipline, in which the integration of multi-omics, clinical, and environmental data enhances both mechanistic insight and translational precision. This convergence reflects the maturation of data-informed endocrinology, where diagnostic and prognostic innovation emerges from the synthesis of diverse data modalities rather than from any single layer of observation. By harmonizing biochemical, genetic, immunologic, and ecological perspectives, these studies redefine biomarker discovery as a process of multi-scale inference and integration -one that connects molecular precision with population-level relevance and real-world applicability. In this new framework, integration is not merely a methodological approach but a scientific imperative -transforming endocrinology into a discipline that systematically bridges molecules to medicine, and data to diagnosis.

Non-adherence to medication represents a significant concern in managing chronic conditions, resulting in poor health outcomes and increased mortality. Artificial Intelligence (AI) and Machine Learning (ML) models offer promising solutions by analyzing large datasets to improve the understanding and prediction of non-adherence. However, the dimensionality of data and associated computational costs, including environmental impact, remain a challenge in the application of AI systems. This study explores the influence of data dimensionality on the evaluation metrics and computational costs of ML models in predicting medication adherence. Utilizing adherence behaviour data collected in the BEAMER project, dimensionality reduction and Logistic Regression (LR) methods are performed. The results provide a comprehensive understanding of the models' performance after the application of Recursive Feature Elimination with Cross Validation (RFECV) and Principal Component Analysis (PCA), while also emphasizing the importance of evaluating the carbon footprint in ML models. The analysis indicates that pipelines including all variables and those incorporating RFECV present better evaluation metrics compared to those utilizing PCA, although this approach requires the least computational resources. These findings highlight the importance of tailoring project designs to their specific use cases and requirements.Clinical Relevance- This study highlights the importance of balancing accuracy, computational efficiency, interpretability and environmental impact in AI-based adherence to medication models, aiding clinicians in selecting optimal approaches for healthcare applications.

In this work, we provide information about mechanically induced crystallization (MIC) kinetics. First, we show that the usual definition of threshold elongations for crystallization and melting does not always make sense. We then show the effect of the loading path on the crystallization levels reached. Finally, we define a quantity representative of the crystallization kinetics and try to describe it using quantities as elongation or stress. At the same time, we illustrate that crystallization is a self-regulating phenomenon: a saturation of crystallization kinetics can be reached.

Abstract Background Environmental factors, which play a critical role in the rising incidence of inflammatory bowel disease (IBD), may hold the key for prevention. Statins, known for their cardiovascular benefits, possess anti-inflammatory properties and have been hypothesized to reduce the risk of developing IBD, though prior data have been limited. Thus, to advance what is known, we prospectively assessed the association between statin use and the risk of IBD among individuals eligible for statin therapy. Methods Using a prospective new user design within the Danish National Registries, we identified a population-based cohort of individuals ≥ 40 years-old from 2008-2022 who were eligible for statin therapy for the primary prevention of cardiovascular disease. Statin users were matched 1:1 to non-statin users on age, sex, year, and underlying cardiovascular risk factors, with this process repeated one hundred times to ensure an accurate risk estimate. Cox proportional hazards models were then used to obtain adjusted hazard ratios (aHRs) for the development of incident IBD. Results In total, 163,563 statin users were matched to 163,563 non-statin users during each iteration. The overall cohort was predominantly male (56.2%), aged between 50-69 years old (63.5%), and primarily met criteria for statin consideration due to the presence of hypertension in the setting of an elevated low-density lipoprotein (51.9%). Overall, statin users did not have a lower risk of developing IBD compared to non-users (aHR 0.88, 95% CI 0.74-1.04), with subgroup analysis demonstrating that this effect held true for both UC (aHR 0.86, 95% CI 0.70-1.05) and CD (aHR 0.95, 95% CI 0.70-1.28; Table 1; Figure 1). Sensitivity analysis where statin users were censored at the time of treatment discontinuation suggested a possible benefit for the development of UC (aHR 0.78, 95%CI 0.61-1.00), though this was not statistically significant. Conclusion In the largest and first nationwide new-user designed cohort study, statin use—prescribed in individuals ≥40 years for the primary prevention of cardiovascular disease—was not associated with a reduced risk of developing UC or CD. This contrasts with findings from smaller, previous case-control studies, however our data suggest that ongoing use may possibly be protective for the later development of UC.

Abstract Background Axial and peripheral spondyloarthritides (SpA) and inflammatory bowel disease (IBD) are immune-mediated diseases with overlap in clinical and immunological features. However, temporal correlation of SpA and IBD in individual patients is not well characterized. Methods We conducted a nationwide population-based study between January 1, 1998 and July 31, 2022 to determine SpA prevalence before, and incidence after IBD diagnosis. Using ICD-10 codes in cross-linked Danish registers, we identified all residents with IBD diagnosis between January 1, 2006 and July 31, 2014 as well as controls matched 1:5 on age, sex, calendar year, and municipality of residency. We determined the odds ratios of SpA in the IBD group in the preceding years. Using adjusted Cox regression analysis, we determined the hazard ratios of new-onset SpA following IBD diagnosis in the proceeding years. Results Of 111,696 individuals included in the study, 18,616 (16.7%) were diagnosed with IBD. The prevalence of SpA was higher in IBD cases relative to matched controls (Figure). The aOR of SpA among those who later developed IBD, UC, or CD, compared to matched controls, was 1.98 (95% CI 1.81, 2.17), 1.67 (95% CI 1.49, 1.87) and 2.76 (95% CI 2.37, 3.21), respectively, in the 8-year period preceding IBD diagnosis. The odds of SpA were highest in the 2 years preceding IBD diagnosis. Among those with IBD, UC, and CD, compared to those without IBD, the aHR of new-onset SpA was 2.47 (95% CI 2.30, 2.65), 2.18 (2.00, 2.37), and 3.20 (95% CI 2.84, 3.60), respectively. When stratifying the IBD effect of SpA on sex and age, both females and the age group 8-45 years had higher risks for SpA following IBD diagnosis, when compared to male and older age groups, respectively. Conclusion In a population-based cohort, we demonstrate that the odds of SpA diagnosis are increased at least up to 8 years preceding IBD diagnosis and are highest 2 years preceding IBD diagnosis. SpA risk is increased among those with IBD, particularly among women and young adults. These clinical associations can improve early diagnosis and treatment of both IBD and SpA.

Abstract Although immunization through vaccination is one of the most successful public health interventions, actual coverage of vaccination program has decreased rapidly over the last years due to increased vaccine hesitancy. Cognitive biases have been shown to play an important role in affecting vaccine hesitancy. In this study, we conducted a randomized-control trial (N = 2000, N = 1000 from Spain and N = 1000 from Bulgaria), where subjects were randomly assigned to one condition. The conditions differed by whether an electronic Product Information (ePI) was presented to the subjects and by the type of information that was focused to the patient. The current study showed that the provision of digital information in the form of an ePI has important consequences for achieving high vaccination rates. The main result suggests that providing vaccination information in the form of an ePI can increase patients’ vaccine hesitancy. This effect remained when positive and/or negative information in the ePI was focused to the patients. Additionally, we observe that vaccine hesitant individuals spend less time reading the ePI. We conclude, by relating the current study to the relevant literature, that information overload could be the main driver of vaccine hesitancy in the context of this study.

The utilization of modified FOLFIRINOX (mFFX) therapy has shown notable advancements in patient outcomes in both localized and metastatic PDAC. Nevertheless, the effectiveness of mFFX treatment comes at the cost of elevated toxicity, leading to its restriction to patients with adequate performance status. Consequently, the administration of mFFX is contingent upon patient performance rather than rational criteria. The ideal scenario would involve the ability to assess the sensitivity of each drug within the mFFX regimen, minimizing unnecessary toxicity without compromising clinical benefits. We developed transcriptomic signatures for each drug of the mFFX regimen (5FU, oxaliplatin and irinotecan) by integrating transcriptomic data from PDC, PDO and PDX with their corresponding chemo-response profiles to capture the biological components responsible for the response to each drug. We further validated the signatures in a cohort of 167 patients with advanced and metastatic PDAC. All three signatures captured high responder patients for OS and PFS in the mFFX arm exclusively. We then studied the response of patients to 0, 1, 2 and 3 drugs and we identified a positive correlation between the number of drugs predicted as sensitive and the OS and PFS, and the with objective response rate. We developed three novel transcriptome-based signatures which define sensitivity for each mFFX components that can be used to rationalize the administration of the mFFX regimen in patients with metastatic pancreatic cancer and could help to avoid unnecessary toxic effects.

Maintenance has become a critical issue across all sectors of industry.It is not only a technical necessity but also a major, if not the primary, cost driver.With the advancement of technology, maintenance processes are increasingly driven by data collection.Vast amounts of data are gathered, stored, and analysed to provide valuable management insights.This data comes in various formats and encompasses multiple relationships, making proper structuring essential.The data structure must thus be as open as possible, meaning it should be extensible and capable of integrating new data types.Drawing from the feedback and experience gained through our extensive work with the major French railway companies, we have developed a generic and innovative data model.This model supports a cutting-edge tool we call the Intelligent Digital Data Twin (IDDT, or Preditic-IDDT in reference to our company's name) for predictive and preventive maintenance.Unlike traditional Digital Twins, Preditic-IDDT focuses on the data flow surrounding the system under study rather than the physical system itself.

The extreme climatic events are increasing worldwide and are becoming more and more recurrent. It is becoming urgent to take action to reduce them. The Mediterranean region is one of the most affected areas by the effects of climate change. The recent events in France, Spain, Greece, Turkey or even Libya reflect the violence of hydro-meteorological phenomena due in part to the rising sea temperatures. To limit the occurrence of such events, it is necessary to have some feedback. PREDICT uses post-event spatial imaging techniques such as rapid mapping to obtains geospatial data in just a few hours after the event. This permits to manage and document the consequences of such events. Furthermore, PREDICT has been developing a program called COSPARIN (COntribution du SPAtial à la gestion du Risque INondation) for few years now, with European Space Agency (ESA) and the National Centre of Spatial studies (CNES) approval. The main goal of this program is to better understand and monitor extreme events before, during and after their occurrence. This relies on an innovative method based on the use of the most recent satellite data available as well as algorithms and artificial intelligence to establish a global estimation of precipitations and an estimation of potential flooding areas. To improve forecasting and deploy early warning systems, Europe has set up the HORIZON program which includes a new project called GOBEYOND (GeO and weather multi-hazard impact Based Early warning and response systems supporting rapid deploYment of first respONders in EU and beyond). The aim of this project is to improve existing tools and methods by designing two platforms of Multi Risk Impact-based Early Warning System (MR-IEWS) for Europe and Mediterranean area (North Africa included). This draws on all available data, including COSPARIN satellite data, to forecast hydro-meteorological risks (floods, storms, heatwaves) and geological risks (earthquakes, volcanic eruptions, tsunamis). Also, this aims to promote rapid and effective operational response by those in charge of safety and to move from simple hazard forecasting to risk forecasting by taking vulnerability into account. Keywords: Climate change, Spatial data, Forecast, Innovation