An artificial intelligence algorithm for co-clustering to help in pharmacovigilance before and during the COVID-19 pandemic.

AIMS: Monitoring drug safety in real-world settings is the primary aim of pharmacovigilance. Frequent adverse drug reactions (ADRs) are usually identified during drug development. Rare ones are mostly characterized through post-marketing scrutiny, increasingly with the use of data mining and disproportionality approaches, which lead to new drug safety signals. Nonetheless, waves of excessive numbers of reports, often stirred up by social media, may overwhelm and distort this process, as observed recently with levothyroxine or COVID-19 vaccines. As human resources become rarer in the field of pharmacovigilance, we aimed to evaluate the performance of an unsupervised co-clustering method to help the monitoring of drug safety. METHODS: A dynamic latent block model (dLBM), based on a time-dependent co-clustering generative method, was used to summarize all regional ADR reports (n = 45 269) issued between 1 January 2012 and 28 February 2022. After analysis of their intra and extra interrelationships, all reports were grouped into different cluster types (time, drug, ADR). RESULTS: Our model clustered all reports in 10 time, 10 ADR and 9 drug collections. Based on such clustering, three prominent societal problems were detected, subsequent to public health concerns about drug safety, including a prominent media hype about the perceived safety of COVID-19 vaccines. The dLBM also highlighted some specific drug-ADR relationships, such as the association between antiplatelets, anticoagulants and bleeding. CONCLUSIONS: Co-clustering and dLBM appear as promising tools to explore large pharmacovigilance databases. They allow, 'unsupervisedly', the detection, exploration and strengthening of safety signals, facilitating the analysis of massive upsurges of reports.

Unobserved classes and extra variables in high-dimensional discriminant analysis.

In supervised classification problems, the test set may contain data points belonging to classes not observed in the learning phase. Moreover, the same units in the test data may be measured on a set of additional variables recorded at a subsequent stage with respect to when the learning sample was collected. In this situation, the classifier built in the learning phase needs to adapt to handle potential unknown classes and the extra dimensions. We introduce a model-based discriminant approach, Dimension-Adaptive Mixture Discriminant Analysis (D-AMDA), which can detect unobserved classes and adapt to the increasing dimensionality. Model estimation is carried out via a full inductive approach based on an EM algorithm. The method is then embedded in a more general framework for adaptive variable selection and classification suitable for data of large dimensions. A simulation study and an artificial experiment related to classification of adulterated honey samples are used to validate the ability of the proposed framework to deal with complex situations.

Cardiac dyspnea risk zones in the South of France identified by geo-pollution trends study.

The incidence of cardiac dyspnea (CD) and the distribution of pollution in the south of France suggests that environmental pollution may have a role in disease triggering. CD is a hallmark symptom of heart failure leading to reduced ability to function and engage in activities of daily living. To show the impact of short-term pollution exposure on the increment of CD emergency room visits, we collected pollutants and climate measurements on a daily basis and 43,400 events of CD in the Région Sud from 2013 to 2018. We used a distributed lag non-linear model (DLNM) to assess the association between air pollution and CD events. We divided the region in 357 zones to reconciliate environmental and emergency room visits data. We applied the DLNM on the entire region, on zones grouped by pollution trends and on singular zones. Each pollutant has a significant effect on triggering CD. Depending on the pollutant, we identified four shapes of exposure curves to describe the impact of pollution on CD events: early and late effect for NO2; U-shape and rainbow-shape (or inverted U) for O3; all the four shapes for PM10. In the biggest cities, O3 has the most significant association along with the PM10. In the west side, a delayed effect triggered by PM10 was found. Zones along the main highway are mostly affected by NO2 pollution with an increase of the association for a period up to 9 days after the pollution peak. Our results can be used by local authorities to set up specific prevention policies, public alerts that adapt to the different zones and support public health prediction-making. We developed a user-friendly web application called Health, Environment in PACA Region Tool (HEART) to collect our results. HEART will allow citizens, researchers and local authorities to monitor the impact of pollution trends on local public health.

Co-clustering of Time-Dependent Data via the Shape Invariant Model.

Multivariate time-dependent data, where multiple features are observed over time for a set of individuals, are increasingly widespread in many application domains. To model these data, we need to account for relations among both time instants and variables and, at the same time, for subject heterogeneity. We propose a new co-clustering methodology for grouping individuals and variables simultaneously, designed to handle both functional and longitudinal data. Our approach borrows some concepts from the curve registration framework by embedding the shape invariant model in the latent block model, estimated via a suitable modification of the SEM-Gibbs algorithm. The resulting procedure allows for several user-defined specifications of the notion of cluster that can be chosen on substantive grounds and provides parsimonious summaries of complex time-dependent data by partitioning data matrices into homogeneous blocks. Along with the explicit modelling of time evolution, these aspects allow for an easy interpretation of the clusters, from which also low-dimensional settings may benefit.

Accurate assessment of nonalcoholic fatty liver disease lesions in liver allograft biopsies by a smartphone platform: A proof of concept.

Macrovesicular steatosis (MS) is a major risk factor for liver graft failure after transplantation and pathological microscopic examination of a frozen tissue section remains the gold standard for its assessment. However, the latter requires an experienced in-house pathologist for correct and rapid diagnosis as well as specific equipment that is not always available. Smartphones, which are must-have tools for everyone, are very suitable for incorporation into promising technology to generate moveable diagnostic tools as for telepathology. The study aims to compare the microscopic assessment of nonalcoholic fatty liver disease (NAFLD) spectrum in liver allograft biopsies by a smartphone microscopy platform (DIPLE device) to standard light microscopy. Forty-two liver graft biopsies were evaluated in transmitted light, using an iPhone X and the microscopy platform. A significant correlation was reported between the two different approaches for graft MS assessment (Spearman's correlation coefficient: r = .93; p < .001) and for steatohepatitis feature (r = .56; p < .001; r = .45; p < .001). Based on these findings, a smartphone integrated with a cheap microscopy platform can achieve adequate accuracy in the assessment of NAFLD in liver graft and could be used as an alternative to standard light microscopy when the latter is unavailable.