HUMBERT , Olivier

Purpose We evaluated the prognostic value of immunotherapy-induced organ inflammation observed on (18)FDG PET in patients with non-small cell lung cancer (NSCLC) treated with immune checkpoint inhibitors (ICPIs). Methods Data from patients with IIIB/IV NSCLC included in two different prospective trials were analyzed. (18)FDG PET/CT exams were...

International audience

International audience

Semi supervised learning (SSL) provides an effective means of leveraging unlabelled data to improve a model's performance. Even though the domain has received a considerable amount of attention in the past years, most methods present the common drawback of being unsafe. By safeness we mean the quality of not degrading a fully supervised model...

Image registration is a key task in medical imaging applications, allowing to represent medical images in a common spatial reference frame. Current literature on image registration is generally based on the assumption that images are usually accessible to the researcher, from which the spatial transformation is subsequently estimated. This...

Abstract Background Presently, there is a wide variety of classification methods and deep neural network approaches in bioinformatics. Deep neural networks have proven their effectiveness for classification tasks, and have outperformed classical methods, but they suffer from a lack of interpretability. Therefore, these innovative methods are...

Abstract Background Presently, there is a wide variety of classification methods and deep neural network approaches in bioinformatics. Deep neural networks have proven their effectiveness for classification tasks, and have outperformed classical methods, but they suffer from a lack of interpretability. Therefore, these innovative methods are...

Image registration is a key task in medical imaging applications, allowing to represent medical images in a common spatial reference frame. Current approaches to image registration are generally based on the assumption that the content of the images is usually accessible in clear form, from which the spatial transformation is subsequently...

International audience

Semi-supervised learning is a powerful techniquefor leveraging unlabeled data to improve machinelearning models, but it can be affected by the pres-ence of "informative" labels, which occur whensome classes are more likely to be labeled thanothers. In the missing data literature, such labelsare called missing not at random. In this paper,we...

Abstract Purpose FDOPA PET shows good performance for the diagnosis of striatal dopaminergic denervation, making it a valuable tool for the differential diagnosis of Parkinsonism. Textural features are image biomarkers that could potentially improve the early diagnosis and monitoring of neurodegenerative parkinsonian syndromes. We explored the...

BACKGROUND:When using 18F-FDG PET, glucose metabolism quantification is affected by various factors. We aimed to investigate the benefit of different SUV normalizations to improve the accuracy of 18F-FDG uptake to predict breast cancer aggressiveness and response to treatment.METHODS:Two hundred fifty-two women with locally advanced breast...

Early changes in tumor glucose metabolism (SUVmax) and in tumor blood flow (BF) have been evaluated separately for monitoring breast cancer response to neoadjuvant chemotherapy (NAC). This study used a single F-18-FDG dual-acquisition PET examination to simultaneously assess these two imaging features and to determine whether they correlate...

Abstract Purpose Iodine-125 ( 125 I) seeds can be used as landmarks to locate non-palpable breast lesions instead of implanting metal wires. This relatively new technique requires a nuclear probe usually used for technetium-99m ( 99m Tc) sentinel node detection. This study aimed to compare the performances of different probes and valid the...

Abstract Background Supervised classification methods have been used for many years for feature selection in metabolomics and other omics studies. We developed a novel primal-dual based classification method (PD-CR) that can perform classification with rejection and feature selection on high dimensional datasets. PD-CR projects data onto a low...

Abstract Background Supervised classification methods have been used for many years for feature selection in metabolomics and other omics studies. We developed a novel primal-dual based classification method (PD-CR) that can perform classification with rejection and feature selection on high dimensional datasets. PD-CR projects data onto a low...

Abstract Background Supervised classification methods have been used for many years for feature selection in metabolomics and other omics studies. We developed a novel primal-dual based classification method (PD-CR) that can perform classification with rejection and feature selection on high dimensional datasets. PD-CR projects data onto a low...

International audience