Search

Programmed death-ligand 1 (PD-L1) expressions play a crucial role in guiding therapeutic interventions such as the use of tyrosine kinase inhibitors (TKIs) and immune checkpoint inhibitors (ICIs) in lung cancer. Conventional determination of PD-L1 status includes careful surgical or biopsied tumor specimens. These specimens are gathered through invasive procedures, representing a risk of difficulties and potential challenges in getting reliable and representative tissue samples. Using a single center cohort of 189 patients, our objective was to evaluate various fusion methods that used non-invasive computed tomography (CT) and 18 F-FDG positron emission tomography (PET) images as inputs to various deep learning models to automatically predict PD-L1 in non-small cell lung cancer (NSCLC). We compared three different architectures (ResNet, DenseNet, and EfficientNet) and considered different input data (CT only, PET only, PET/CT early fusion, PET/CT late fusion without as well as with partially and fully shared weights to determine the best model performance. Models were assessed utilizing areas under the receiver operating characteristic curves (AUCs) considering their 95% confidence intervals (CI). The fusion of PET and CT images as input yielded better performance for PD-L1 classification. The different data fusion schemes systematically outperformed their individual counterparts when used as input of the various deep models. Furthermore, early fusion consistently outperformed late fusion, probably as a result of its capacity to capture more complicated patterns by merging PET and CT derived content at a lower level. When we looked more closely at the effects of weight sharing in late fusion architectures, we discovered that while it might boost model stability, it did not always result in better results. This suggests that although weight sharing could be beneficial when modality parameters are similar, the anatomical and metabolic information provided by CT and PET scans are too dissimilar to consistently lead to improved PD-L1 status predictions. [ABSTRACT FROM AUTHOR]

Quantitative Gait Analysis (QGA) is considered as an objective measure of gait performance. In this study, we aim at designing an artificial intelligence that can efficiently predict the progression of gait quality using kinematic data obtained from QGA. For this purpose, a gait database collected from 734 patients with gait disorders is used. As the patient walks, kinematic data is collected during the gait session. This data is processed to generate the Gait Profile Score (GPS) for each gait cycle. Tracking potential GPS variations enables detecting changes in gait quality. In this regard, our work is driven by predicting such future variations. Two approaches were considered: signal-based and image-based. The signal-based one uses raw gait cycles, while the image-based one employs a two-dimensional Fast Fourier Transform (2D FFT) representation of gait cycles. Several architectures were developed, and the obtained Area Under the Curve (AUC) was above 0.72 for both approaches. To the best of our knowledge, our study is the first to apply neural networks for gait prediction tasks. [ABSTRACT FROM AUTHOR]

Independent validation studies of automatic diabetic retinopathy screening systems have recently shown a drop of screening performance on external data. Beyond diabetic retinopathy, this study investigates the generalizability of deep learning (DL) algorithms for screening various ocular anomalies in fundus photographs, across heterogeneous populations and imaging protocols. The following datasets are considered: OPHDIAT (France, diabetic population), OphtaMaine (France, general population), RIADD (India, general population) and ODIR (China, general population). Two multi-disease DL algorithms were developed: a Single-Dataset (SD) network, trained on the largest dataset (OPHDIAT), and a Multiple-Dataset (MD) network, trained on multiple datasets simultaneously. To assess their generalizability, both algorithms were evaluated whenever training and test data originate from overlapping datasets or from disjoint datasets. The SD network achieved a mean per-disease area under the receiver operating characteristic curve (mAUC) of 0.9571 on OPHDIAT. However, it generalized poorly to the other three datasets (mAUC < 0.9). When all four datasets were involved in training, the MD network significantly outperformed the SD network (p = 0.0058), indicating improved generality. However, in leave-one-dataset-out experiments, performance of the MD network was significantly lower on populations unseen during training than on populations involved in training (p < 0.0001), indicating imperfect generalizability. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new collaborative process that aims to detect macrocalcifications from mammographic images while minimizing false negative detections. This process is made up of three main phases: suspicious area detection, candidate object identification, and collaborative classification. The main concept is to operate on the entire image divided into homogenous regions called superpixels which are used to identify both suspicious areas and candidate objects. The collaborative classification phase consists in making the initial results of different microcalcification detectors collaborate in order to produce a new common decision and reduce their initial disagreements. The detectors share the information about their detected objects and associated labels in order to refine their initial decisions based on those of the other collaborators. This refinement consists of iteratively updating the candidate object labels of each detector following local and contextual analyses based on prior knowledge about the links between super pixels and macrocalcifications. This process iteratively reduces the disagreement between different detectors and estimates local reliability terms for each super pixel. The final result is obtained by a conjunctive combination of the new detector decisions reached by the collaborative process. The proposed approach is evaluated on the publicly available INBreast dataset. Experimental results show the benefits gained in terms of improving microcalcification detection performances compared to existing detectors as well as ordinary fusion operators. [ABSTRACT FROM AUTHOR]

Microcalcifications (MCs) are considered as the first indicator of breast cancer development. Their morphology, in terms of shape and size, is considered as the most important criterion that determines their malignity degrees. Therefore, the accurate delineation of MC is a cornerstone step in their automatic diagnosis process. In this paper, we propose a new conditional region growing (CRG) approach with the ability of finding the accurate MC boundaries starting from selected seed points. The starting seed points are determined based on regional maxima detection and superpixel analysis. The region growing step is controlled by a set of criteria that are adapted to MC detection in terms of contrast and shape variation. These criteria are derived from prior knowledge to characterize MCs and can be divided into two categories. The first one concerns the neighbourhood searching size. The second one deals with the analysis of gradient information and shape evolution within the growing process. In order to prove the effectiveness and the reliability in terms of MC detection and delineation, several experiments have been carried out on MCs of various types, with both qualitative and quantitative analysis. The comparison of the proposed approach with state-of-the art proves the importance of the used criteria in the context of MC delineation, towards a better management of breast cancer. Graphical Abstract Flowchart of the proposed approach. [ABSTRACT FROM AUTHOR]

Recent clinical research studies in forensic identification have highlighted the interest in sphenoid sinus anatomical characterization. Their pneumatization, well known as extremely variable in degrees and directions, could contribute to the radiologic identification, especially if dental records, fingerPrints, or DNA samples are not available. In this paper, we present a new approach for automatic person identification based on sphenoid sinus features extracted from computed tomography (CT) images of the skull. First, we present a new approach for fully automatic 3D reconstruction of the sphenoid hemisinuses which combines the fuzzy c-means method and mathematical morphology operations to detect and segment the object of interest. Second, deep shape features are extracted from both hemisinuses using a dilated residual version of a stacked convolutional auto-encoder. The obtained binary segmentation masks are thus hierarchically mapped into a compact and low-dimensional space preserving their semantic similarity. We finally employ the ℓ2 distance to recognize the sphenoid sinus and therefore identify the person. This novel sphenoid sinus recognition method obtained 100% of identification accuracy when applied on a dataset composed of 85 CT scans stemming from 72 individuals. Automatic Forensic Identification using 3D Sphenoid Sinus Segmentation and Deep Characterization from Dilated Residual Auto-Encoders. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of image segmentation by iterative region aggregations starting from an initial superpixel decomposition. Classical approaches for this task compute superpixel similarity using distance measures between superpixel descriptor vectors. This usually poses the well-known problem of the semantic gap and fails to properly aggregate visually non-homogeneous superpixels that belong to the same high-level object. This work proposes to use random forests to learn the merging probability between adjacent superpixels in order to overcome the aforementioned issues. Compared to existing works, this approach learns the fusion rules without explicit similarity measure computation. We also introduce a new superpixel context descriptor to strengthen the learned characteristics towards better similarity prediction. Image segmentation is then achieved by iteratively merging the most similar superpixel pairs selected using a similarity weighting objective function. Experimental results of our approach on four datasets including DAVIS 2017 and ISIC 2018 show its potential compared to state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Purpose : Toward an efficient clinical management of hepatocellular carcinoma (HCC), we propose a classification framework dedicated to tumor necrosis rate estimation from dynamic contrast-enhanced CT scans. Based on machine learning, it requires weak interaction efforts to segment healthy, active and necrotic liver tissues. Methods : Our contributions are two-fold. First, we apply random forest (RF) on supervoxels using multi-phase supervoxel-based features that discriminate tissues based on their dynamic in response to contrast agent injection. Second, we extend this technique in a hierarchical multi-scale fashion to deal with multiple spatial extents and appearance heterogeneity. It translates in an adaptive data sampling scheme combining RF and hierarchical multi-scale tree resulting from recursive supervoxel decomposition. By concatenating multi-phase features across the hierarchical multi-scale tree to describe leaf supervoxels, we enable RF to automatically infer the most informative scales without defining any explicit rules on how to combine them. Results : Assessment on clinical data confirms the benefits of multi-phase information embedded in a multi-scale supervoxel representation for HCC tumor segmentation. Conclusion : Dedicated but not limited only to HCC management, both contributions reach further steps toward more accurate multi-label tissue classification. [ABSTRACT FROM AUTHOR]