Diagnostic and Prognostic Potential of Multiparametric Renal MRI in Kidney Transplant Patients.

BACKGROUND: Multiparametric MRI provides assessment of functional and structural parameters in kidney allografts. It offers a non-invasive alternative to the current reference standard of kidney biopsy. PURPOSE: To evaluate the diagnostic and prognostic utility of MRI parameters in the assessment of allograft function in the first 3-months post-transplantation. STUDY TYPE: Prospective. SUBJECTS: 32 transplant recipients (54 ± 17 years, 20 females), divided into two groups according to estimated glomerular filtration rate (eGFR) at 3-months post-transplantation: inferior graft function (IGF; eGFR<45 mL/min/1.73 m2 , n = 10) and superior graft function (SGF; eGFR ≥ 45 mL/min/1.73 m2 , n = 22). Further categorization was based on the need for hemodialysis (C1) and decrease in s-creatinine (C2) at 1-week post-transplantation: delayed-graft-function (DGF: n = 4 C1, n = 10 C2) and early graft-function (EGF: n = 28 C1, n = 22 C2). FIELD STRENGTH/SEQUENCE: 3-T, pseudo-continuous arterial spin labeling, T1-mapping, and diffusion-weighted imaging. ASSESSMENT: Multiparametric MRI was evaluated at 1-week in all patients and 3-months after transplantation in 28 patients. Renal blood flow (RBF), diffusion coefficients (ADC, ΔADC, D, ∆ $$ \Delta $$ D, D*, flowing fraction f), T1 and ∆ $$ \Delta $$ T1 were calculated in cortex and medulla. The diagnostic and prognostic value of these parameters, obtained at 3-months and 1-week post-transplantation, respectively, was evaluated in the cortex to discriminate between DGF and EGF, and between SGF and IGF. STATISTICAL TESTS: Logistic regression, receiver-operating-characteristics, area-under-the-curve (AUC), confidence intervals (CIs), analysis-of-variance, t-test, Wilcoxon-Mann-Whitney test, Fisher's exact test, Pearson's correlation. P-value<0.05 was considered significant. RESULTS: DGF patients exhibited significantly lower cortical RBF and f and higher D*. The diagnostic value of MRI for detecting DGF was excellent (AUC = 100%). Significant differences between patients with IGF and SGF were found in RBF, ∆T1 , and ∆D. Multiparametric MRI showed higher diagnostic (AUC = 95.32%; CI: 88%-100%) and prognostic (AUC = 97.47%, CI: 92%-100%) values for detecting IGF than eGFR (AUC = 89.50%, CI: 79%-100%). DATA CONCLUSION: Multiparametric MRI may show high diagnostic and prognostic value in transplanted patients, yielding better results compared to eGFR measurements. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 1.

Efficacy of different 8 h time-restricted eating schedules on visceral adipose tissue and cardiometabolic health: A study protocol.

BACKGROUND AND AIMS: To investigate the efficacy and feasibility of three different 8 h time-restricted eating (TRE) schedules (i.e., early, late, and self-selected) compared to each other and to a usual-care (UC) intervention on visceral adipose tissue (VAT) and cardiometabolic health in men and women. METHODS AND RESULTS: Anticipated 208 adults (50% women) aged 30-60 years, with overweight/obesity (25 ≤ BMI<40 kg/m2) and with mild metabolic impairments will be recruited for this parallel-group, multicenter randomized controlled trial. Participants will be randomly allocated (1:1:1:1) to one of four groups for 12 weeks: UC, early TRE, late TRE or self-selected TRE. The UC group will maintain their habitual eating window and receive, as well as the TRE groups, healthy lifestyle education for weight management. The early TRE group will start eating not later than 10:00, and the late TRE group not before 13:00. The self-selected TRE group will select an 8 h eating window before the intervention and maintain it over the intervention. The primary outcome is changes in VAT, whereas secondary outcomes include body composition and cardiometabolic risk factors. CONCLUSION: This study will determine whether the timing of the eating window during TRE impacts its efficacy on VAT, body composition and cardiometabolic risk factors and provide insights about its feasibility.

Multiparametric renal magnetic resonance imaging: A reproducibility study in renal allografts with stable function.

Monitoring renal allograft function after transplantation is key for the early detection of allograft impairment, which in turn can contribute to preventing the loss of the allograft. Multiparametric renal MRI (mpMRI) is a promising noninvasive technique to assess and characterize renal physiopathology; however, few studies have employed mpMRI in renal allografts with stable function (maintained function over a long time period). The purposes of the current study were to evaluate the reproducibility of mpMRI in transplant patients and to characterize normal values of the measured parameters, and to estimate the labeling efficiency of Pseudo-Continuous Arterial Spin Labeling (PCASL) in the infrarenal aorta using numerical simulations considering experimental measurements of aortic blood flow profiles. The subjects were 20 transplant patients with stable kidney function, maintained over 1 year. The MRI protocol consisted of PCASL, intravoxel incoherent motion, and T1 inversion recovery. Phase contrast was used to measure aortic blood flow. Renal blood flow (RBF), diffusion coefficient (D), pseudo-diffusion coefficient (D*), flowing fraction ( f ), and T1 maps were calculated and mean values were measured in the cortex and medulla. The labeling efficiency of PCASL was estimated from simulation of Bloch equations. Reproducibility was assessed with the within-subject coefficient of variation, intraclass correlation coefficient, and Bland-Altman analysis. Correlations were evaluated using the Pearson correlation coefficient. The significance level was p less than 0.05. Cortical reproducibility was very good for T1, D, and RBF, moderate for f , and low for D*, while medullary reproducibility was good for T1 and D. Significant correlations in the cortex between RBF and f (r = 0.66), RBF and eGFR (r = 0.64), and D* and eGFR (r = -0.57) were found. Normal values of the measured parameters employing the mpMRI protocol in kidney transplant patients with stable function were characterized and the results showed good reproducibility of the techniques.

Differences in specific abdominal fat depots between metabolically healthy and unhealthy children with overweight/obesity: The role of cardiorespiratory fitness.

OBJECTIVES: Fat depots localization has a critical role in the metabolic health status of adults. Nevertheless, whether that is also the case in children remains under-studied. Therefore, the aims of this study were: (i) to examine the differences between metabolically healthy (MHO) and unhealthy (MUO) overweight/obesity phenotypes on specific abdominal fat depots, and (ii) to further explore whether cardiorespiratory fitness plays a major role in the differences between metabolic phenotypes among children with overweight/obesity. METHODS: A total of 114 children with overweight/obesity (10.6 ± 1.1 years, 62 girls) were included. Children were classified as MHO (n = 68) or MUO. visceral (VAT), abdominal subcutaneous (ASAT), intermuscular abdominal (IMAAT), psoas, hepatic, pancreatic, and lumbar bone marrow adipose tissues were measured by magnetic resonance imaging. Cardiorespiratory fitness was assessed using the 20 m shuttle run test. RESULTS: MHO children had lower VAT and ASAT contents and psoas fat fraction compared to MUO children (difference = 12.4%-25.8%, all p < 0.035). MUO-unfit had more VAT and ASAT content than those MUO-fit and MHO-fit (difference = 34.8%-45.3%, all p < 0.044). MUO-unfit shows also greater IMAAT fat fraction than those MUO-fit and MHO-fit peers (difference = 16.4%-13.9% respectively, all p ≤ 0.001). In addition, MHO-unfit presented higher IMAAT fat fraction than MHO-fit (difference = 13.4%, p < 0.001). MUO-unfit presented higher psoas fat fraction than MHO-fit (difference = 29.1%, p = 0.008). CONCLUSIONS: VAT together with ASAT and psoas fat fraction, were lower in MHO than in MUO children. Further, we also observed that being fit, regardless of metabolic phenotype, has a protective role over the specific abdominal fat depots among children with overweight/obesity.

Associations of fitness and physical activity with specific abdominal fat depots in children with overweight/obesity.

OBJECTIVES: To examine the relationship between physical fitness and physical activity (PA) with specific abdominal fat depots and their potential implications for cardiometabolic risk and insulin resistance (IR) in children with overweight/obesity. MATERIALS AND METHODS: A total of 116 children with overweight/obesity (10.7 ± 1.1 year, 54% girls) participated in the study. Abdominal visceral (VAT), subcutaneous (ASAT), and intermuscular abdominal adipose tissue (IMAAT) were assessed by magnetic resonance imaging. The cardiometabolic risk (MetS) score and the insulin resistance homeostasis model assessment (HOMA-IR) were calculated. Health-related physical fitness components (treadmill test, and 20 m shuttle run, handgrip, standing broad jump and 4 × 10 m tests) were evaluated, and PA was measured (accelerometry). Children were categorized as fit or unfit for each specific fitness test, and as active or inactive. RESULTS: Higher VAT, ASAT, and IMAAT were associated with higher MetS score and HOMA-IR (all p < 0.02). A better performance in all fitness tests and total and vigorous PA were strongly associated with lower VAT (all p < 0.04), ASAT (all p < 0.005), and IMAAT (all p < 0.005). Fit or active children had lower VAT, ASAT, and IMAAT (all p < 0.03) than their unfit or inactive counterparts. CONCLUSION: These results reinforce the importance of having adequate fitness and PA levels to reduce abdominal fat accumulation in children. Given that VAT, ASAT, and IMAAT are associated with higher risk of developing cardiovascular diseases and type 2 diabetes, the improvement of physical fitness by the promotion of PA should be goals of lifestyle interventions for improving health in children with overweight/obesity.

Accurate Pupil Center Detection in Off-the-Shelf Eye Tracking Systems Using Convolutional Neural Networks.

Remote eye tracking technology has suffered an increasing growth in recent years due to its applicability in many research areas. In this paper, a video-oculography method based on convolutional neural networks (CNNs) for pupil center detection over webcam images is proposed. As the first contribution of this work and in order to train the model, a pupil center manual labeling procedure of a facial landmark dataset has been performed. The model has been tested over both real and synthetic databases and outperforms state-of-the-art methods, achieving pupil center estimation errors below the size of a constricted pupil in more than 95% of the images, while reducing computing time by a 8 factor. Results show the importance of use high quality training data and well-known architectures to achieve an outstanding performance.

Low-Cost Eye Tracking Calibration: A Knowledge-Based Study.

Subject calibration has been demonstrated to improve the accuracy in high-performance eye trackers. However, the true weight of calibration in off-the-shelf eye tracking solutions is still not addressed. In this work, a theoretical framework to measure the effects of calibration in deep learning-based gaze estimation is proposed for low-resolution systems. To this end, features extracted from the synthetic U2Eyes dataset are used in a fully connected network in order to isolate the effect of specific user's features, such as kappa angles. Then, the impact of system calibration in a real setup employing I2Head dataset images is studied. The obtained results show accuracy improvements over 50%, probing that calibration is a key process also in low-resolution gaze estimation scenarios. Furthermore, we show that after calibration accuracy values close to those obtained by high-resolution systems, in the range of 0.7°, could be theoretically obtained if a careful selection of image features was performed, demonstrating significant room for improvement for off-the-shelf eye tracking systems.

A novel chaperone-effector-immunity system identified in uropathogenic Escherichia coli UMN026.

Background: Urinary tract infections (UTIs) are very common worldwide. According to their symptomatology, these infections are classified as pyelonephritis, cystitis, or asymptomatic bacteriuria (AB). Approximately 75-95% of UTIs are caused by uropathogenic Escherichia coli (UPEC), which is an extraintestinal bacterium that possesses virulence factors for bacterial adherence and invasion in the urinary tract. In addition, UPEC possesses type 6 secretion systems (T6SS) as virulence mechanisms that can participate in bacterial competition and in bacterial pathogenicity. UPEC UMN026 carries three genes, namely, ECUMN_0231, ECUMN_0232, and ECUMN_0233, which encode three uncharacterized proteins related to the T6SS that are conserved in strains from phylogroups B2 and D and have been proposed as biomarkers of UTIs. Aim: To analyze the frequency of the ECUMN_0231, ECUMN_0232, ECUMN_0233, and vgrG genes in UTI isolates, as well as their expression in Luria Bertani (LB) medium and urine; to determine whether these genes are related to UTI symptoms or bacterial competence and to identify functional domains on the putative proteins. Methods: The frequency of the ECUMN and vgrG genes in 99 clinical isolates from UPEC was determined by endpoint PCR. The relationship between gene presence and UTI symptomatology was determined using the chi2 test, with p < 0.05 considered to indicate statistical significance. The expression of the three ECUMN genes and vgrG was analyzed by RT-PCR. The antibacterial activity of strain UMN026 was determined by bacterial competence assays. The identification of functional domains and the docking were performed using bioinformatic tools. Results: The ECUMN genes are conserved in 33.3% of clinical isolates from patients with symptomatic and asymptomatic UTIs and have no relationship with UTI symptomatology. Of the ECUMN+ isolates, only five (15.15%, 5/33) had the three ECUMN and vgrG genes. These genes were expressed in LB broth and urine in UPEC UMN026 but not in all the clinical isolates. Strain UMN026 had antibacterial activity against UPEC clinical isolate 4014 (ECUMN-) and E. faecalis but not against isolate 4012 (ECUMN+). Bioinformatics analysis suggested that the ECUMN genes encode a chaperone/effector/immunity system. Conclusions: The ECUMN genes are conserved in clinical isolates from symptomatic and asymptomatic patients and are not related to UTI symptoms. However, these genes encode a putative chaperone/effector/immunity system that seems to be involved in the antibacterial activity of strain UMN026.

Liver Fat, Bone Marrow Adipose Tissue, and Bone Mineral Density in Children With Overweight.

CONTEXT: Hepatic steatosis is associated with decreased bone mineral density (BMD). Bone marrow fat fraction (BMFF) could play a role in this relationship in children with obesity. OBJECTIVE: The objectives of this work were (i) to examine the relationship between the lumbar spine (LS) BMFF and BMD, and (ii) to explore the mediating role of LS-BMFF on the relationship between percentage hepatic fat with LS-BMD in preadolescent children with overweight/obesity. METHODS: Hepatic fat and LS-BMFF (magnetic resonance imaging) and areal LS-BMD (LS-aBMD, dual-energy x-ray absorptiometry) were measured in 106 children (aged 10.6 ± 1.1 years, 53.8% girls) with overweight/obesity. RESULTS: LS-BMFF was inversely associated with LS-aBMD (r = -0.313; P = .001) and directly related with percentage hepatic fat (r = 0.276; P = .005). LS-BMFF was significantly greater in children with than without hepatic steatosis (P = .003; Cohen's d: 0.61; 95% CI, -0.21 to 1.0), while no significant difference was seen between children with overweight and those with obesity (P = .604; Cohen's d: 0.16; 95% CI, -0.21-0.55). Mediating analysis indicated that LS-BMFF is an important mediator (50%) in the association of hepatic fat with lower LS-aBMD (indirect effect: ß = -.076; 95% CI, -0.143 to -0.015). CONCLUSION: These findings suggest that hepatic steatosis, rather than overall excess adiposity, is associated with greater bone marrow adipose tissue in preadolescent children with overweight/obesity, which in turn, is related to lower BMD. Hepatic steatosis could be a potential biomarker of osteoporosis risk, and a therapeutic target for interventions that aim to reduce not only hepatic steatosis, but for those designed to improve bone health in such children.

A deep learning image analysis method for renal perfusion estimation in pseudo-continuous arterial spin labelling MRI.

Accurate segmentation of renal tissues is an essential step for renal perfusion estimation and postoperative assessment of the allograft. Images are usually manually labeled, which is tedious and prone to human error. We present an image analysis method for the automatic estimation of renal perfusion based on perfusion magnetic resonance imaging. Specifically, non-contrasted pseudo-continuous arterial spin labeling (PCASL) images are used for kidney transplant evaluation and perfusion estimation, as a biomarker of the status of the allograft. The proposed method uses machine/deep learning tools for the segmentation and classification of renal cortical and medullary tissues and automates the estimation of perfusion values. Data from 16 transplant patients has been used for the experiments. The automatic analysis of differentiated tissues within the kidney, such as cortex and medulla, is performed by employing the time-intensity-curves of non-contrasted T1-weighted MRI series. Specifically, using the Dice similarity coefficient as a figure of merit, results above 93%, 92% and 82% are obtained for whole kidney, cortex, and medulla, respectively. Besides, estimated cortical and medullary perfusion values are considered to be within the acceptable ranges within clinical practice.

Development of a prediction protocol for the screening of metabolic associated fatty liver disease in children with overweight or obesity.

BACKGROUND: The early detection and management of children with metabolic associated fatty liver disease (MAFLD) is challenging. OBJECTIVE: To develop a non-invasive and accurate prediction protocol for the identification of MAFLD among children with overweight/obesity candidates to confirmatory diagnosis. METHODS: A total of 115 children aged 8-12 years with overweight/obesity, recruited at a primary care, were enrolled in this cross-sectional study. The external validation was performed using a cohort of children with overweight/obesity (N = 46) aged 8.5-14.0 years. MAFLD (≥5.5% hepatic fat) was diagnosed by magnetic resonance imaging (MRI). Fasting blood biochemical parameters were measured, and 25 candidates' single nucleotide polymorphisms (SNPs) were determined. Variables potentially associated with the presence of MAFLD were included in a multivariate logistic regression. RESULTS: Children with MAFLD (36%) showed higher plasma triglycerides (TG), insulin, homeostasis model assessment of insulin resistance (HOMA-IR), alanine aminotransferase (ALT), aspartate transaminase (AST), glutamyl-transferase (GGT) and ferritin (p < 0.05). The distribution of the risk-alleles of PPARGrs13081389, PPARGrs1801282, HFErs1800562 and PNLPLA3rs4823173 was significantly different between children with and without MAFLD (p < 0.05). Three biochemical- and/or SNPs-based predictive models were developed, showing strong discriminatory capacity (AUC-ROC: 0.708-0.888) but limited diagnostic performance (sensitivity 67%-82% and specificity 63%-69%). A prediction protocol with elevated sensitivity (72%) and specificity (84%) based on two consecutive steps was developed. The external validation showed similar results: sensitivity of 70% and specificity of 85%. CONCLUSIONS: The HEPAKID prediction protocol is an accurate, easy to implant, minimally invasive and low economic cost tool useful for the early identification and management of paediatric MAFLD in primary care.

Effect of a Multicomponent Intervention on Hepatic Steatosis Is Partially Mediated by the Reduction of Intermuscular Abdominal Adipose Tissue in Children With Overweight or Obesity: The EFIGRO Project.

OBJECTIVE: In adults, there is evidence that improvement of metabolic-associated fatty liver disease (MAFLD) depends on the reduction of myosteatosis. In children, in whom the prevalence of MAFLD is alarming, this muscle-liver crosstalk has not been tested. Therefore, we aimed to explore whether the effects of a multicomponent intervention on hepatic fat is mediated by changes in intermuscular abdominal adipose tissue (IMAAT) in children with overweight/obesity. RESEARCH DESIGN AND METHODS: A total of 116 children with overweight/obesity were allocated to a 22-week family-based lifestyle and psychoeducational intervention (control group, n = 57) or the same intervention plus supervised exercise (exercise group, n = 59). Hepatic fat percentage and IMAAT were acquired by MRI at baseline and at the end of the intervention. RESULTS: Changes in IMAAT explained 20.7% of the improvements in hepatic steatosis (P < 0.05). Only children who meaningfully reduced their IMAAT (i.e., responders) had improved hepatic steatosis at the end of the intervention (within-group analysis: responders -20% [P = 0.005] vs. nonresponders -1.5% [P = 0.803]). Between-group analysis showed greater reductions in favor of IMAAT responders compared with nonresponders (18.3% vs. 0.6%, P = 0.018), regardless of overall abdominal fat loss. CONCLUSIONS: The reduction of IMAAT plays a relevant role in the improvement of hepatic steatosis after a multicomponent intervention in children with overweight/obesity. Indeed, only children who achieved a meaningful reduction in IMAAT at the end of the intervention had a reduced percentage of hepatic fat independent of abdominal fat loss. Our findings suggest that abdominal muscle fat infiltration could be a therapeutic target for the treatment of MAFLD in childhood.

Effects of a Family-Based Lifestyle Intervention Plus Supervised Exercise Training on Abdominal Fat Depots in Children With Overweight or Obesity: A Secondary Analysis of a Nonrandomized Clinical Trial.

Importance: Excess abdominal fat is a major determinant in the development of insulin resistance and other metabolic disorders. Increased visceral adipose tissue (VAT) seems to precede the development of insulin resistance and is therefore a prime target of childhood lifestyle interventions aimed at preventing diabetes. Objectives: To examine the effect of added exercise to a family-based lifestyle intervention program designed to reduce VAT plus subcutaneous (ASAT), intermuscular (IMAAT), and pancreatic (PAT) adipose tissue in children with overweight or obesity and to explore the effect of changes in VAT on insulin resistance. Design, Setting, and Participants: This 2-group, parallel-design clinical trial was conducted in Vitoria-Gasteiz, Spain. A total of 116 children with overweight or obesity participated and were assigned to a 22-week family-based lifestyle program (control group [n = 57]) or the same program plus an exercise intervention (exercise group [n = 59]). Data were collected between September 1, 2014, and June 30, 2017, and imaging processing for fat depot assessments and data analysis were performed between May 1, 2019, and February 12, 2021. Interventions: The compared interventions consisted of a family-based lifestyle and psychoeducation program (two 90-minute sessions per month) and the same program plus supervised exercise (three 90-minute sessions per week). Main Outcomes and Measures: The primary outcome of this study was the change in VAT between baseline and 22 weeks as estimated by magnetic resonance imaging. The secondary outcomes were changes in ASAT, IMAAT, and PAT. The effect of changes in VAT area on insulin resistance was also recorded. Results: The 116 participants included in the analysis (62 girls [53.4%]) had a mean (SD) age of 10.6 (1.1) years, and 67 (57.8%) presented with obesity. Significantly greater reductions were recorded for the exercise group in terms of reduction in VAT (-18.1% vs -8.5% for the control group; P = .004), ASAT (-9.9% vs -3.0%; P = .001), and IMAAT (-6.0% vs -2.6%; P = .02) fat fractions compared with the control group. Changes in VAT explained 87.6% of the improvement seen in insulin resistance (ß = -0.102 [95% CI, -0.230 to -0.002]). Conclusions and Relevance: These findings suggest that the addition of exercise to a lifestyle intervention program substantially enhanced the positive effects on abdominal fat depots in children with overweight or obesity. In addition, the reduction in VAT seemed to largely mediate the improvement of insulin sensitivity. These results highlight the importance of including exercise as part of lifestyle therapies aimed at treating childhood obesity and preventing the development of type 2 diabetes. Trial Registration: ClinicalTrials.gov Identifier: NCT02258126.

Low cost gaze estimation: knowledge-based solutions.

Eye tracking technology in low resolution scenarios is not a completely solved issue to date. The possibility of using eye tracking in a mobile gadget is a challenging objective that would permit to spread this technology to non-explored fields. In this paper, a knowledge based approach is presented to solve gaze estimation in low resolution settings. The understanding of the high resolution paradigm permits to propose alternative models to solve gaze estimation. In this manner, three models are presented: a geometrical model, an interpolation model and a compound model, as solutions for gaze estimation for remote low resolution systems. Since this work considers head position essential to improve gaze accuracy, a method for head pose estimation is also proposed. The methods are validated in an optimal framework, I2Head database, which combines head and gaze data. The experimental validation of the models demonstrates their sensitivity to image processing inaccuracies, critical in the case of the geometrical model. Static and extreme movement scenarios are analyzed showing the higher robustness of compound and geometrical models in the presence of user's displacement. Accuracy values of about 3° have been obtained, increasing to values close to 5° in extreme displacement settings, results fully comparable with the state-of-the-art.

Ultrasound Image Discrimination between Benign and Malignant Adnexal Masses Based on a Neural Network Approach.

The discrimination between benign and malignant adnexal masses in ultrasound images represents one of the most challenging problems in gynecologic practice. In the study described here, a new method for automatic discrimination of adnexal masses based on a neural networks approach was tested. The proposed method first calculates seven different types of characteristics (local binary pattern, fractal dimension, entropy, invariant moments, gray level co-occurrence matrix, law texture energy and Gabor wavelet) from ultrasound images of the ovary, from which several features are extracted and collected together with the clinical patient age. The proposed technique was validated using 106 benign and 39 malignant images obtained from 145 patients, corresponding to its probability of appearance in general population. On evaluation of the classifier, an accuracy of 98.78%, sensitivity of 98.50%, specificity of 98.90% and area under the curve of 0.997 were calculated.

Generalized multiresolution hierarchical shape models via automatic landmark clusterization.

Point Distribution Models (PDM) are some of the most popular shape description techniques in medical imaging. However, to create an accurate shape model it is essential to have a representative sample of the underlying population, which is often challenging. This problem is particularly relevant as the dimensionality of the modeled structures increases, and becomes critical when dealing with complex 3D shapes. In this paper, we introduce a new generalized multiresolution hierarchical PDM (GMRH-PDM) able to efficiently address the high-dimension-low-sample-size challenge when modeling complex structures. Unlike previous approaches, our new and general framework extends hierarchical modeling to any type of structure (multi- and single-object shapes) allowing to describe efficiently the shape variability at different levels of resolution. Importantly, the configuration of the algorithm is automatized thanks to the new agglomerative landmark clustering method presented here. Our new and automatic GMRH-PDM framework performed significantly better than classical approaches, and as well as the state-of-the-art with the best manual configuration. Evaluations have been studied for two different cases, the right kidney, and a multi-object case composed of eight subcortical structures.

Multiresolution hierarchical shape models in 3D subcortical brain structures.

Point distribution models (PDM) are one of the most extended methods to characterize the underlying population of set of samples, whose usefulness has been demonstrated in a wide variety of applications, including medical imaging. However, one important issue remains unsolved: the large number of training samples required. This problem becomes critical as the complexity of the problem increases, and the modeling of 3D multiobjects/organs represents one of the most challenging cases. Based on the 3D wavelet transform, this paper introduces a multiresolution hierarchical variant of PDM (MRH-PDM) able to efficiently characterize the different inter-object relationships, as well as the particular locality of each element separately. The significant advantage of this new method over two previous approaches in terms of accuracy has been successfully verified for the particular case of 3D subcortical brain structures.

Hierarchical statistical shape models of multiobject anatomical structures: application to brain MRI.

The accurate segmentation of subcortical brain structures in magnetic resonance (MR) images is of crucial importance in the interdisciplinary field of medical imaging. Although statistical approaches such as active shape models (ASMs) have proven to be particularly useful in the modeling of multiobject shapes, they are inefficient when facing challenging problems. Based on the wavelet transform, the fully generic multiresolution framework presented in this paper allows us to decompose the interobject relationships into different levels of detail. The aim of this hierarchical decomposition is twofold: to efficiently characterize the relationships between objects and their particular localities. Experiments performed on an eight-object structure defined in axial cross sectional MR brain images show that the new hierarchical segmentation significantly improves the accuracy of the segmentation, and while it exhibits a remarkable robustness with respect to the size of the training set.

Near real-time stereo matching using geodesic diffusion.

Adaptive-weight algorithms currently represent the state of the art in local stereo matching. However, due to their computational requirements, these types of solutions are not suitable for real-time implementation. Here, we present a novel aggregation method inspired by the anisotropic diffusion technique used in image filtering. The proposed aggregation algorithm produces results similar to adaptive-weight solutions while reducing the computational requirements. Moreover, near real-time performance is demonstrated with a GPU implementation of the algorithm.