Multimodal Imaging Analysis Research Articles

Relative impact of amyloid-β, lacunes, and downstream imaging markers on cognitive trajectories.

SEE COHEN DOI101093/AWW183 FOR A SCIENTIFIC COMMENTARY ON THIS ARTICLE: Amyloid-β and cerebral small vessel disease are the two major causes of cognitive impairment in the elderly. However, the underlying mechanisms responsible for precisely how amyloid-β and cerebral small vessel disease affect cognitive impairment remain unclear. We investigated the effects of amyloid-β and lacunes on downstream imaging markers including structural network and cortical thickness, further analysing their relative impact on cognitive trajectories. We prospectively recruited a pool of 117 mild cognitive impairment patients (45 amnestic type and 72 subcortical vascular type), from which 83 patients received annual follow-up with neuropsychological tests and brain magnetic resonance imaging for 3 years, and 87 patients received a second Pittsburgh compound B positron emission tomography analysis. Structural networks based on diffusion tensor imaging and cortical thickness were analysed. We used linear mixed effect regression models to evaluate the effects of imaging markers on cognitive decline. Time-varying Pittsburgh compound B uptake was associated with temporoparietal thinning, which correlated with memory decline (verbal memory test, unstandardized β = -0.79, P < 0.001; visual memory test, unstandardized β = -2.84, P = 0.009). Time-varying lacune number was associated with the degree of frontoparietal network disruption or thinning, which further affected frontal-executive function decline (Digit span backward test, unstandardized β = -0.05, P = 0.002; Stroop colour test, unstandardized β = -0.94, P = 0.008). Of the multiple imaging markers analysed, Pittsburgh compound B uptake and the number of lacunes had the greatest association with memory decline and frontal-executive function decline, respectively: Time-varying Pittsburgh compound B uptake (standardized β = -0.25, P = 0.010) showed the strongest effect on visual memory test, followed by time-varying temporoparietal thickness (standardized β = 0.21, P = 0.010) and time-varying nodal efficiency (standardized β = 0.17, P = 0.024). Time-varying lacune number (standardized β = -0.25, P = 0.014) showed the strongest effect on time-varying digit span backward test followed by time-varying nodal efficiency (standardized β = 0.17, P = 0.021). Finally, time-varying lacune number (β = -0.22, P = 0.034) showed the strongest effect on time-varying Stroop colour test followed by time-varying frontal thickness (standardized β = 0.19, P = 0.026). Our multimodal imaging analyses suggest that cognitive trajectories related to amyloid-β and lacunes have distinct paths, and that amyloid-β or lacunes have greatest impact on cognitive decline. Our results provide rationale for the targeting of amyloid-β and lacunes in therapeutic strategies aimed at ameliorating cognitive decline.

VoxelStats: A MATLAB Package for Multi-Modal Voxel-Wise Brain Image Analysis.

In healthy individuals, behavioral outcomes are highly associated with the variability on brain regional structure or neurochemical phenotypes. Similarly, in the context of neurodegenerative conditions, neuroimaging reveals that cognitive decline is linked to the magnitude of atrophy, neurochemical declines, or concentrations of abnormal protein aggregates across brain regions. However, modeling the effects of multiple regional abnormalities as determinants of cognitive decline at the voxel level remains largely unexplored by multimodal imaging research, given the high computational cost of estimating regression models for every single voxel from various imaging modalities. VoxelStats is a voxel-wise computational framework to overcome these computational limitations and to perform statistical operations on multiple scalar variables and imaging modalities at the voxel level. VoxelStats package has been developed in Matlab® and supports imaging formats such as Nifti-1, ANALYZE, and MINC v2. Prebuilt functions in VoxelStats enable the user to perform voxel-wise general and generalized linear models and mixed effect models with multiple volumetric covariates. Importantly, VoxelStats can recognize scalar values or image volumes as response variables and can accommodate volumetric statistical covariates as well as their interaction effects with other variables. Furthermore, this package includes built-in functionality to perform voxel-wise receiver operating characteristic analysis and paired and unpaired group contrast analysis. Validation of VoxelStats was conducted by comparing the linear regression functionality with existing toolboxes such as glim_image and RMINC. The validation results were identical to existing methods and the additional functionality was demonstrated by generating feature case assessments (t-statistics, odds ratio, and true positive rate maps). In summary, VoxelStats expands the current methods for multimodal imaging analysis by allowing the estimation of advanced regional association metrics at the voxel level.

Automated analysis of multimodal fluorescence lifetime imaging and optical coherence tomography data for the diagnosis of oral cancer in the hamster cheek pouch model.

It is known that the progression of oral cancer is accompanied by changes in both tissue biochemistry and morphology. A multimodal imaging approach combining functional and structural imaging modalities could therefore provide a more comprehensive prognosis of oral cancer. This idea forms the central theme of the current study, wherein this premise is examined in the context of a multimodal imaging system that combines fluorescence lifetime imaging (FLIM) and optical coherence tomography (OCT). Towards this end, in the first part of the present study, the diagnostic advantage obtained by using both fluorescence intensity and lifetime information is assessed. In the second part of the study, the diagnostic potential of FLIM-derived biochemical features is compared with that of OCT-derived morphological features. For an objective assessment, several quantitative biochemical and morphological features from FLIM and OCT data, respectively, were obtained using signal and image processing techniques. These features were subsequently used in a statistical classification framework to quantify the diagnostic potential of different features. The classification accuracy for combined FLIM and OCT features was estimated to be 87.4%, which was statistically higher than accuracy based on only FLIM (83.2%) or OCT (81.0%) features. Moreover, the complimentary information provided by FLIM and OCT features, resulted in highest sensitivity and specificity for the combined FLIM and OCT features for discriminating benign (88.2% sens., 92.0% spec.), pre-cancerous (81.5% sens., 96.0% spec.), and cancerous (90.1% sens., 92.0% spec.) classes.

Multimodal Image Analysis in Alzheimer's Disease via Statistical Modelling of Non-local Intensity Correlations.

The joint analysis of brain atrophy measured with magnetic resonance imaging (MRI) and hypometabolism measured with positron emission tomography with fluorodeoxyglucose (FDG-PET) is of primary importance in developing models of pathological changes in Alzheimer’s disease (AD). Most of the current multimodal analyses in AD assume a local (spatially overlapping) relationship between MR and FDG-PET intensities. However, it is well known that atrophy and hypometabolism are prominent in different anatomical areas. The aim of this work is to describe the relationship between atrophy and hypometabolism by means of a data-driven statistical model of non-overlapping intensity correlations. For this purpose, FDG-PET and MRI signals are jointly analyzed through a computationally tractable formulation of partial least squares regression (PLSR). The PLSR model is estimated and validated on a large clinical cohort of 1049 individuals from the ADNI dataset. Results show that the proposed non-local analysis outperforms classical local approaches in terms of predictive accuracy while providing a plausible description of disease dynamics: early AD is characterised by non-overlapping temporal atrophy and temporo-parietal hypometabolism, while the later disease stages show overlapping brain atrophy and hypometabolism spread in temporal, parietal and cortical areas.

Development and aging of superficial white matter myelin from young adulthood to old age: Mapping by vertex-based surface statistics (VBSS).

Superficial white matter (SWM) lies immediately beneath cortical gray matter and consists primarily of short association fibers. The characteristics of SWM and its development and aging were seldom examined in the literature and warrant further investigation. Magnetization transfer imaging is sensitive to myelin changes in the white matter. Using an innovative multimodal imaging analysis approach, vertex-based surface statistics (VBSS), the current study vertexwise mapped age-related changes of magnetization transfer ratio (MTR) in SWM from young adulthood to old age (30-85 years, N = 66). Results demonstrated regionally selective and temporally heterochronologic changes of SWM MTR with age, including (1) inverted U-shaped trajectories of SWM MTR in the rostral middle frontal, medial temporal, and temporoparietal regions, suggesting continuing myelination and protracted maturation till age 40-50 years and accelerating demyelination at age 60 and beyond, (2) linear decline of SWM MTR in the middle and superior temporal, and pericalcarine areas, indicating early maturation and less acceleration in age-related degeneration, and (3) no significant changes of SWM MTR in the primary motor, somatosensory and auditory regions, suggesting resistance to age-related deterioration. We did not observe similar patterns of changes in cortical thickness in our sample, suggesting the observed SWM MTR changes are not due to cortical atrophy. Hum Brain Mapp 37:1759-1769, 2016. © 2016 Wiley Periodicals, Inc.

Multimodal Image Analysis in Acquired Vitelliform Lesions and Adult-Onset Foveomacular Vitelliform Dystrophy

Purpose. To characterize vitelliform lesions (VLs) in adult-onset foveomacular vitelliform dystrophy (AOFVD) and acquired vitelliform (AVL) patients using multimodal image analysis. Methods. Retrospective study of twenty-eight eyes from nineteen patients diagnosed with AVL or AOFVD. They were evaluated by color fundus photographs, fundus autofluorescence (FAF), fluorescein angiography (FA), and spectral-domain optical coherence tomography (SD-OCT). Results. Bilateral VLs were associated with AOFVD (p = 0.013). Regular and centered VLs were associated with AOFVD (p = 0.004 and p = 0.016), whereas irregular and noncentered lesions were more frequent in AVL patients. Visual acuity, greatest linear dimension (GLD), lesion height (LH), and pseudohypopyon were similar between groups. Whereas median LH and GLD in AVL group diminished significantly during follow-up (p = 0.009 and p = 0.001), AOFVD lesions tended to become larger and thicker. Conclusions. When consulting a patient presenting a VL with unknown age of onset, familial history, or previous retinal diseases, some aspects of multimodal imaging assessment may lead the ophthalmologist to a correct diagnosis.

Descriptor evaluation and feature regression for multimodal image analysis

In this paper, we present feature descriptor evaluation and feature regression for multimodal image analysis. First, we compare the performances of several popular interest point detectors and feature descriptors from multimodal images with focus on visual and infrared images. The performances of detectors are evaluated mainly by the score of repeatability and accuracy and the descriptors are assessed by using the rate of precision and recall. Secondly, we analyze the relationship between the corresponding descriptors computed from multimodal images. The descriptors are regressed by means of linear regression as well as Gaussian process. Then the features on infrared images are predicted by mapping the descriptors from visual images to the infrared modality through the regression results. Predictions are assessed in two ways: the statistics of absolute error between true values and actual values, and the precision score of matching the predicted descriptors to the original infrared descriptors. We believe that this evaluating information will be useful when selecting an appropriate detector and descriptor for multimodal image analysis. Also the experimental results show that regression methods achieve a well-assessed relationship between corresponding descriptors from multiple modalities.

Tensor Analysis and Fusion of Multimodal Brain Images

Current high-throughput data acquisition technologies probe dynamical systems with different imaging modalities, generating massive data sets at different spatial and temporal resolutions posing challenging problems in multimodal data fusion. A case in point is the attempt to parse out the brain structures and networks that underpin human cognitive processes by analysis of different neuroimaging modalities (functional MRI, EEG, NIRS etc.). We emphasize that the multimodal, multi-scale nature of neuroimaging data is well reflected by a multi-way (tensor) structure where the underlying processes can be summarized by a relatively small number of components or "atoms". We introduce Markov-Penrose diagrams - an integration of Bayesian DAG and tensor network notation in order to analyze these models. These diagrams not only clarify matrix and tensor EEG and fMRI time/frequency analysis and inverse problems, but also help understand multimodal fusion via Multiway Partial Least Squares and Coupled Matrix-Tensor Factorization. We show here, for the first time, that Granger causal analysis of brain networks is a tensor regression problem, thus allowing the atomic decomposition of brain networks. Analysis of EEG and fMRI recordings shows the potential of the methods and suggests their use in other scientific domains.

Multimodal imaging analysis of an orthotopic head and neck cancer mouse model and application of anti-CD137 tumor immune therapy.

Recent technical progress makes sophisticated noninvasive imaging methods available for murine models. For the first time, in this study, we applied fluorodeoxyglucose (FDG)-positron emission tomography (PET)-CT and FDG-PET-MRI to a murine orthotopic model of head and neck cancer immunotherapy. Tumor growth of floor of the mouth tumors was evaluated by multimodal small-animal imaging using FDG-PET-CT and FDG-PET-MRI. The immunotherapeutic effects of anti-CD137 antibody therapy were examined on body weight, tumor growth, and tumor-infiltrating immune cells in longitudinal imaging studies and immunohistochemical analyses. Imaging revealed aggressive, fast-growing tumors without evidence of local or distant metastases. CD137 immunotherapy decreased tumor take and growth and stabilized body weight over time. A clear case of tumor regression was demonstrated by longitudinal PET-CT. The murine model mimics the characteristics of head and neck cancer in humans and offers excellent opportunities to investigate immunomodulatory anticancer drugs. The CD137 antibody showed antitumor effects in some therapy-responsive mice.

Automatic visible and infrared face registration based on silhouette matching and robust transformation estimation

Registration of multi-sensor data (particularly visible color sensors and infrared sensors) is a prerequisite for multimodal image analysis such as image fusion. In this paper, we proposed an automatic registration technique for visible and infrared face images based on silhouette matching and robust transformation estimation. The key idea is to represent a (visible or infrared) face image by its silhouette which is extracted from the image’s edge map and consists of a set of discrete points, and then align the two silhouette point sets by using their feature similarity and spatial geometrical information. More precisely, our algorithm first matches the silhouette point sets by their local shape features such as shape context, which creates a set of putative correspondences that may contaminated by outliers. Next, we estimate the accurate transformation from the putative correspondence set under a robust maximum likelihood framework combining with the EM algorithm, where the transformation between the image pair is modeled by a parametric model such as the rigid or affine transformation. The qualitative and quantitative comparisons on a publicly available database demonstrate that our method significantly outperforms other state-of-the-art visible/infrared face registration methods. As a result, our method will be beneficial for fusion-based face recognition.

Standardised unfold map of the left atrium: regional definition for multimodal image analysis

Background Atrial fibrillation (AF) can be treated by catheter ablation to electrically isolate the pulmonary veins (PVs) from the left atrial (LA) body. To guide the procedure, an anatomical representation of the LA is usually visualised by an electroanatomical mapping system. As a follow-up, late gadolinium enhancement MR sequences can be obtained to assess the extent of ablation delivery. However, visual analysis and quantification is timeconsuming with poor reproducibility. Unfold maps offer advantages over 3D surfaces for visualisation of imaging information while providing a common reference space for inter-subject comparison. We have developed a method to automatically map multimodal information onto a standardised unfold map (SUM) of the LA. We use the SUMs to compare force and scar information from the same patient. Methods Ten patients underwent ablation procedure for AF. The intra-procedure anatomical surfaces containing force measurements from the CARTO system were exported. Six months after the procedure, the patients underwent a cardiac MRI scan in order to quantify the extent of postablation injury. The scan included 3D whole heart (3DWH) and 3D late enhancement (3DLE) sequences. The LA was segmented from the 3DWH datasets using a multi-atlas fusion approach. The segmentations were manually corrected when necessary to generate a high fidelity result. The LA segmentation initialised a 3DLE scar extraction algorithm based on maximum intensity projection. We extracted a surface mesh from the LA segmentation and mapped the 3DLE intensity values to corresponding surface mesh locations. To standardise the resulting surfaces, we automatically clipped the mitral plane and the PVs 10mm distal to the ostia. To bring each mesh to the average mesh space, we computed a landmark based affine transformation using the PV end points and the mitral centroid. We applied this transformation to each patient’ sm esh and performed surface matching via currents. After surface registration, we projected the scar and force values of each patient onto the average mesh. The average mesh was unfolded using a fast surface parameterisation technique developed for texture mapping. The mapping method enables the constraint of the parameterisation to fit a predefined template.

Multimodality imaging and radiological-pathological analysis of ethiodized oil: Imaging biomarker of tumor necrosis after TACE?

TPS503 Background: In transarterial chemoembolization (TACE) for primary and metastatic liver malignancies ethiodized oil (Lipiodol) serves 3 roles: vehicle to deliver high concentration of the chemotherapeutic agent to the tumor, embolic agent to the tumor feeding artery and a contrast agent. The latter allows for intraprocedural visualization during drug delivery. Because of increased retention in liver tumors, Lipiodol can be visualized post-TACE for several weeks to months. Controversial preclinical and clinical data have been published about the role of Lipiodol as an imaging biomarker for treatment success after TACE. Here we investigate whether increased tumor Lipiodol deposition correlates with tumor response and survival using multimodality imaging and radiological/pathological correlation. Methods: This is a prospective phase 2, single center, IRB-approved study. Inclusion criteria include adult patients (Child-Pugh class of A or B and ECOG performance status of 0-2) with unresectable HCC with liver predominant disease or hepatic metastases from any solid tumor. Dynamic contrast-enhanced (DCE) MRI and CT of the liver and PET-CT were obtained before TACE. 24 hours after TACE, a noncontrast CT of the liver was obtained to quantify Lipiodol retention in the targeted tumor. Follow-up imaging consisting of DCE-MRI, CT and PET–CT were also performed approximately 1, 3 and 6 months after the last TACE. Follow up imaging will continue every 2-3 months for up to 4 years. Percent deposition of Lipiodol in the targeted tumor was assessed using 3D quantitative analysis on the 24hr post-TACE CT. Tumor response using RECIST, EASL, modified RECIST, tumor volume (volumetric RECIST), and volumetric tumor enhancement (quantitative EASL) were performed on the DCE-MRI, CT, and PET-CT, and correlated with Lipiodol deposition. Spearman rank correlation coefficient and Pearson’s product-moment correlation were used. The Kaplan-Meier method with the log-rank test was used to assess whether Lipiodol is a biomarker of survival. 12/30 patients with HCC and 12/30 patients with hepatic metastases have been enrolled (#NCT01877187). Clinical trial information: NCT01877187.

Multimodal Imaging Assisting the Early Diagnosis of Cat-Scratch Neuroretinitis

To describe how a multifocal fundus imaging system assisted the early diagnosis of cat scratch neuroretinitis in a case of a 27-year-old male with unilateral visual loss, neuroretinitis, and a peripapillary angiomatous lesion. Multimodal fundus imaging analysis was an essential contributor to the clinical diagnosis of cat scratch neuroretinitis during the early stage of the disease.

Eigenanatomy: Sparse dimensionality reduction for multi-modal medical image analysis

Rigorous statistical analysis of multimodal imaging datasets is challenging. Mass-univariate methods for extracting correlations between image voxels and outcome measurements are not ideal for multimodal datasets, as they do not account for interactions between the different modalities. The extremely high dimensionality of medical images necessitates dimensionality reduction, such as principal component analysis (PCA) or independent component analysis (ICA). These dimensionality reduction techniques, however, consist of contributions from every region in the brain and are therefore difficult to interpret. Recent advances in sparse dimensionality reduction have enabled construction of a set of image regions that explain the variance of the images while still maintaining anatomical interpretability. The projections of the original data on the sparse eigenvectors, however, are highly collinear and therefore difficult to incorporate into multi-modal image analysis pipelines. We propose here a method for clustering sparse eigenvectors and selecting a subset of the eigenvectors to make interpretable predictions from a multi-modal dataset. Evaluation on a publicly available dataset shows that the proposed method outperforms PCA and ICA-based regressions while still maintaining anatomical meaning. To facilitate reproducibility, the complete dataset used and all source code is publicly available.

Non-rigid visible and infrared face registration via regularized Gaussian fields criterion

Registration of multi-sensor data (particularly visible color sensors and infrared sensors) is a prerequisite for multimodal image analysis such as image fusion. Typically, the relationships between image pairs are modeled by rigid or affine transformations. However, this cannot produce accurate alignments when the scenes are not planar, for example, face images. In this paper, we propose a regularized Gaussian fields criterion for non-rigid registration of visible and infrared face images. The key idea is to represent an image by its edge map and align the edge maps by a robust criterion with a non-rigid model. We model the transformation between images in a reproducing kernel Hilbert space and a sparse approximation is applied to the transformation to avoid high computational complexity. Moreover, a coarse-to-fine strategy by applying deterministic annealing is used to overcome local convergence problems. The qualitative and quantitative comparisons on two publicly available databases demonstrate that our method significantly outperforms the state-of-the-art method with an affine model. As a result, our method will be beneficial for fusion-based face recognition.

Changes in the spatial distribution of sclerostin in the osteocytic lacuno-canalicular system in alveolar bone due to orthodontic forces, as detected on multimodal confocal fluorescence imaging analyses

ObjectiveMechanical loading on the bone is sensed by osteocytes. Sclerostin is a molecule secreted by osteocytes that is downregulated by mechanical loading; therefore, its expression level is a potent sensor that indicates the spatial transduction of biomechanical properties in bone. This study applied macroconfocal microscopy to observe the spatial response of alveolar bone to orthodontic forces after immunofluorescence using anti-sclerostin antibodies. DesignOrthodontic tooth movement with the Ni–Ti closed-coil spring was applied between the upper bilateral incisors and the left first molar of mice. Four days after this application, the animals were subjected to multimodal confocal fluorescence imaging analyses. ResultsObvious downregulation of sclerotin in the osteocytic lacuna-canalicular system (LCS) was observed specifically in tensile sites of alveolar bone. Confocal-based three-dimensional fluorescence morphometry further quantitatively demonstrated that the distribution and expression of sclerostin in the tensile sites was significantly reduced compared to that observed in the corresponding control sites. Interestingly, the levels of sclerotin signals in the compression sites were significantly higher than those observed in the control sites, although the distribution of sclerotin was not significantly different. ConclusionsOur observations suggest that spatial changes in the level and distribution of sclerostin in the alveolar LCS trigger successive bone remodelling due to orthodontic tooth movement. The multimodal confocal imaging analyses applied in this work will enhance comprehensive understanding regarding the spatial regulation of molecules of interest from the tissue to the cellular level.

TCT-625 Long term clinical data of the BIOSOLVE-I study with the paclitaxel-eluting absorbable magnesium scaffold (DREAMS) and multi-modality imaging analysis

TCT-625 Long term clinical data of the BIOSOLVE-I study with the paclitaxel-eluting absorbable magnesium scaffold (DREAMS) and multi-modality imaging analysis

Detection of ischemic penumbra using combined perfusion and T2* oxygen challenge imaging.

BackgroundAcute ischemic stroke is common and disabling, but there remains a paucity of acute treatment options and available treatment (thrombolysis) is underutilized. Advanced brain imaging, designed to identify viable hypoperfused tissue (penumbra), could target treatment to a wider population. Existing magnetic resonance imaging and computed tomography-based technologies are not widely used pending validation in ongoing clinical trials. T2* oxygen challenge magnetic resonance imaging, by providing a more direct readout of tissue viability, has the potential to identify more patients likely to benefit from thrombolysis – irrespective of time from stroke onset – and patients within and beyond the 4·5 h thrombolysis treatment window who are unlikely to benefit and are at an increased risk of hemorrhage.AimsThis study employs serial multimodal imaging and voxel-based analysis to develop optimal data processing for T2* oxygen challenge penumbra assessment. Tissue in the ischemic hemisphere is compartmentalized into penumbra, ischemic core, or normal using T2* oxygen challenge (single threshold) or T2* oxygen challenge plus cerebral blood flow (dual threshold) data. Penumbra defined by perfusion imaging/apparent diffusion coefficient mismatch (dual threshold) is included for comparison.MethodsPermanent middle cerebral artery occlusion was induced in male Sprague-Dawley rats (n = 6) prior to serial multimodal imaging: T2* oxygen challenge, diffusion-weighted and perfusion imaging (cerebral blood flow using arterial spin labeling).ResultsAcross the different methods evaluated, T2* oxygen challenge combined with perfusion imaging most closely predicted 24 h infarct volume. Penumbra volume declined from one to four-hours post-stroke: mean ± SD, 77 ± 44 to 49 ± 37 mm3 (single T2* oxygen challenge-based threshold); 55 ± 41 to 37 ± 12 mm3 (dual T2* oxygen challenge/cerebral blood flow); 84 ± 64 to 42 ± 18 mm3 (dual cerebral blood flow/apparent diffusion coefficient), as ischemic core grew: 155 ± 37 to 211 ± 36 mm3 (single apparent diffusion coefficient threshold); 178 ± 56 to 205 ± 33 mm3 (dual T2* oxygen challenge/cerebral blood flow); 139 ± 30 to 168 ± 38 mm3 (dual cerebral blood flow/apparent diffusion coefficient). There was evidence of further lesion growth beyond four-hours (T2-defined edema-corrected infarct, 231 ± 19 mm3).ConclusionsIn conclusion, T2* oxygen challenge combined with perfusion imaging has advantages over alternative magnetic resonance imaging techniques for penumbra detection by providing serial assessment of available penumbra based on tissue viability.

Multimodal and multimedia image analysis and collaborative networking for digestive endoscopy

ObjectiveThe ultimate goal of the Syseo project is to create a chain of collaborative processes to allow the hepato-gastroenterology endoscopy specialist to manage images easily. MethodsA field study has been done to better understand and formalize practices and contexts of use. Based on these results, we have designed tools for gastroenterology, tackling several domains of computer science and reusing well-known format or concepts especially DICOM files, semantic retrieval and infocus-breakpoint. ResultsSyseo consists in four main components: (1) a data management system relying on the well-known standard DICOM format; (2) a polyp ontology and description logics to manage gastroenterological images; (3) software to estimate the size of a neoplasia from colonoscopic images and (4) pearly user interfaces to enhance collaboration. DiscussionPreliminary results of Syseo are quite promising since the proposed solutions enable to efficiently store, annotate, retrieve medical data, while providing relatively accurate measuring tools for physicians and medical staff.

Pseudodrusen Subtypes as Delineated by Multimodal Imaging of the Fundus

To subclassify pseudodrusen based on their appearance in multimodal imaging. Retrospective, observational series. The color fundus photographs and infrared scanning laser ophthalmoscope (IR-SLO) images of patients with pseudodrusen were evaluated along with spectral-domain optical coherence tomography (SD OCT) by masked readers. Distinct types of pseudodrusen could be differentiated. There were 140 eyes of 93 patients with a mean age of 82.4 years. Multimodal imaging analysis showed 3 subtypes of pseudodrusen. One principal type was an orderly array of whitish discrete accumulations principally located in the perifovea, termed dot pseudodrusen. They appeared as hyporeflective spots, often with a target configuration, in IR-SLO images. The second type was interconnected bands of yellowish-white material forming a reticular pattern, called ribbon pseudodrusen, which were located in the perifovea. This subtype was faintly hyporeflective in IR-SLO imaging. Dot pseudodrusen were detected more commonly with IR-SLO imaging than in color photography (P= .014) and ribbon pseudodrusen were seen more frequently in color than in IR-SLO images (P < .001). An uncommon third type of pseudodrusen, yellow-white globules primarily located peripheral to the perifoveal region, appeared hyper-reflective in IR-SLO and were called peripheral pseudodrusen. All 3 types were seen as subretinal drusenoid deposits by SD OCT. Pseudodrusen may be classified into at least 3 categories, each with optimal methods of detection and only 1 that formed a reticular pattern. These findings suggest pseudodrusen could contain differing constituents and therefore may vary in conferred risk for progression to advanced age-related macular disease.