Deep learning to detect anterior cruciate ligament tear on knee MRI: multi-continental external validation.

OBJECTIVES: To develop a deep-learning algorithm for anterior cruciate ligament (ACL) tear detection and to compare its accuracy using two external datasets. METHODS: A database of 19,765 knee MRI scans (17,738 patients) issued from different manufacturers and magnetic fields was used to build a deep learning-based ACL tear detector. Fifteen percent showed partial or complete ACL rupture. Coronal and sagittal fat-suppressed proton density or T2-weighted sequences were used. A Natural Language Processing algorithm was used to automatically label reports associated with each MRI exam. We compared the accuracy of our model on two publicly available external datasets: MRNet, Bien et al, USA (PLoS Med 15:e1002699, 2018); and KneeMRI, Stajduhar et al, Croatia (Comput Methods Prog Biomed 140:151-164, 2017). Receptor operating characteristics (ROC) curves, area under the curve (AUC), sensitivity, specificity, and accuracy were used to evaluate our model. RESULTS: Our neural networks achieved an AUC value of 0.939 for detection of ACL tears, with a sensitivity of 87% (0.875) and a specificity of 91% (0.908). After retraining our model on Bien dataset and Stajduhar dataset, our algorithm achieved AUC of 0.962 (95% CI 0.930-0.988) and 0.922 (95% CI 0.875, 0.962) respectively. Sensitivity, specificity, and accuracy were respectively 85% (95% CI 75-94%, 0.852), 89% (95% CI 82-97%, 0.894), 0.875 (95% CI 0.817-0.933) for Bien dataset, and 68% (95% CI 54-81%, 0.681), 93% (95% CI 89-97%, 0.934), and 0.870 (95% CI 0.821-0.913) for Stajduhar dataset. CONCLUSION: Our algorithm showed high performance in the detection of ACL tears with AUC on two external datasets, demonstrating its generalizability on different manufacturers and populations. This study shows the performance of an algorithm for detecting anterior cruciate ligament tears with an external validation on populations from countries and continents different from the study population. KEY POINTS: • An algorithm for detecting anterior cruciate ligament ruptures was built from a large dataset of nearly 20,000 MRI with AUC values of 0.939, sensitivity of 87%, and specificity of 91%. • This algorithm was tested on two external populations from different other countries: a dataset from an American population and a dataset from a Croatian population. Performance remains high on these two external validation populations (AUC of 0.962 and 0.922 respectively).

Neuroimaging evidence for structural correlates in adolescents resilient to polysubstance use: A five-year follow-up study.

Early initiation of polysubstance use (PSU) is a strong predictor of subsequent addiction, however scarce individuals present resilience capacity. This neuroimaging study aimed to investigate structural correlates associated with cessation or reduction of PSU and determine the extent to which brain structural features accounted for this resilient outcome. Participants from a European community-based cohort self-reported their alcohol, tobacco and cannabis use frequency at ages 14, 16 and 19 and had neuroimaging sessions at ages 14 and 19. We included three groups in the study: the resilient-to-PSU participants showed PSU at 16 and/or 14 but no more at 19 (n = 18), the enduring polysubstance users at 19 displayed PSU continuation from 14 or 16 (n = 193) and the controls were abstinent or low drinking participants (n = 460). We conducted between-group comparisons of grey matter volumes on whole brain using voxel-based morphometry and regional fractional anisotropy using tract-based spatial statistics. Random-forests machine-learning approach generated individual-level PSU-behavior predictions based on personality and neuroimaging features. Adolescents resilient to PSU showed significant larger grey matter volumes in the bilateral cingulate gyrus compared with enduring polysubstance users and controls at ages 19 and 14 (p<0.05 corrected) but no difference in fractional anisotropy. The larger cingulate volumes and personality trait "openness to experience" were the best precursors of resilience to PSU. Early in adolescence, a larger cingulate gyrus differentiated adolescents resilient to PSU, and this feature was critical in predicting this outcome. This study encourages further research into the neurobiological bases of resilience to addictive behaviors.

Refined measure of functional connectomes for improved identifiability and prediction.

Brain functional connectome analysis is commonly based on population-wise inference. However, in this way precious information provided at the individual subject level may be overlooked. Recently, several studies have shown that individual differences contribute strongly to the functional connectivity patterns. In particular, functional connectomes have been proven to offer a fingerprint measure, which can reliably identify a given individual from a pool of participants. In this work, we propose to refine the standard measure of individual functional connectomes using dictionary learning. More specifically, we rely on the assumption that each functional connectivity is dominated by stable group and individual factors. By subtracting population-wise contributions from connectivity patterns facilitated by dictionary representation, intersubject variability should be increased within the group. We validate our approach using several types of analyses. For example, we observe that refined connectivity profiles significantly increase subject-specific identifiability across functional magnetic resonance imaging (fMRI) session combinations. Besides, refined connectomes can also improve the prediction power for cognitive behaviors. In accordance with results from the literature, we find that individual distinctiveness is closely linked with differences in neurocognitive activity within the brain. In summary, our results indicate that individual connectivity analysis benefits from the group-wise inferences and refined connectomes are indeed desirable for brain mapping.

Fast and Accurate Detection of Complex Imaging Genetics Associations Based on Greedy Projected Distance Correlation.

Recent advances in imaging genetics produce large amounts of data including functional MRI images, single nucleotide polymorphisms (SNPs), and cognitive assessments. Understanding the complex interactions among these heterogeneous and complementary data has the potential to help with diagnosis and prevention of mental disorders. However, limited efforts have been made due to the high dimensionality, group structure, and mixed type of these data. In this paper we present a novel method to detect conditional associations between imaging genetics data. We use projected distance correlation to build a conditional dependency graph among high-dimensional mixed data, then use multiple testing to detect significant group level associations (e.g., ROI-gene). In addition, we introduce a scalable algorithm based on orthogonal greedy algorithm, yielding the greedy projected distance correlation (G-PDC). This can reduce the computational cost, which is critical for analyzing large-volume of imaging genomics data. The results from our simulations demonstrate a higher degree of accuracy with GPDC than distance correlation, Pearson's correlation and partial correlation, especially when the correlation is nonlinear. Finally, we apply our method to the Philadelphia Neurodevelopmental data cohort with 866 samples including fMRI images and SNP profiles. The results uncover several statistically significant and biologically interesting interactions, which are further validated with many existing studies. The Matlab code is available at https://sites.google.com/site/jianfang86/gPDC.

FDR-Corrected Sparse Canonical Correlation Analysis With Applications to Imaging Genomics.

Reducing the number of false discoveries is presently one of the most pressing issues in the life sciences. It is of especially great importance for many applications in neuroimaging and genomics, where data sets are typically high-dimensional, which means that the number of explanatory variables exceeds the sample size. The false discovery rate (FDR) is a criterion that can be employed to address that issue. Thus it has gained great popularity as a tool for testing multiple hypotheses. Canonical correlation analysis (CCA) is a statistical technique that is used to make sense of the cross-correlation of two sets of measurements collected on the same set of samples (e.g., brain imaging and genomic data for the same mental illness patients), and sparse CCA extends the classical method to high-dimensional settings. Here, we propose a way of applying the FDR concept to sparse CCA, and a method to control the FDR. The proposed FDR correction directly influences the sparsity of the solution, adapting it to the unknown true sparsity level. Theoretical derivation as well as simulation studies show that our procedure indeed keeps the FDR of the canonical vectors below a user-specified target level. We apply the proposed method to an imaging genomics data set from the Philadelphia Neurodevelopmental Cohort. Our results link the brain connectivity profiles derived from brain activity during an emotion identification task, as measured by functional magnetic resonance imaging, to the corresponding subjects' genomic data.

Estimation of Dynamic Sparse Connectivity Patterns From Resting State fMRI.

Functional connectivity (FC) estimated from functional magnetic resonance imaging (fMRI) time series, especially during resting state periods, provides a powerful tool to assess human brain functional architecture in health, disease, and developmental states. Recently, the focus of connectivity analysis has shifted toward the subnetworks of the brain, which reveals co-activating patterns over time. Most prior works produced a dense set of high-dimensional vectors, which are hard to interpret. In addition, their estimations to a large extent were based on an implicit assumption of spatial and temporal stationarity throughout the fMRI scanning session. In this paper, we propose an approach called dynamic sparse connectivity patterns (dSCPs), which takes advantage of both matrix factorization and time-varying fMRI time series to improve the estimation power of FC. The feasibility of analyzing dynamic FC with our model is first validated through simulated experiments. Then, we use our framework to measure the difference between young adults and children with real fMRI data set from the Philadelphia Neurodevelopmental Cohort (PNC). The results from the PNC data set showed significant FC differences between young adults and children in four different states. For instance, young adults had reduced connectivity between the default mode network and other subnetworks, as well as hyperconnectivity within the visual system in states 1 and 3, and hypoconnectivity in state 2. Meanwhile, they exhibited temporal correlation patterns that changed over time within functional subnetworks. In addition, the dSCPs model indicated that older people tend to spend more time within a relatively connected FC pattern. Overall, the proposed method provides a valid means to assess dynamic FC, which could facilitate the study of brain networks.

Enforcing Co-Expression Within a Brain-Imaging Genomics Regression Framework.

Among the challenges arising in brain imaging genetic studies, estimating the potential links between neurological and genetic variability within a population is key. In this paper, we propose a multivariate, multimodal formulation for variable selection that leverages co-expression patterns across various data modalities. Our approach is based on an intuitive combination of two widely used statistical models: sparse regression and canonical correlation analysis (CCA). While the former seeks multivariate linear relationships between a given phenotype and associated observations, the latter searches to extract co-expression patterns between sets of variables belonging to different modalities. In the following, we propose to rely on a "CCA-type" formulation in order to regularize the classical multimodal sparse regression problem (essentially incorporating both CCA and regression models within a unified formulation). The underlying motivation is to extract discriminative variables that are also co-expressed across modalities. We first show that the simplest formulation of such model can be expressed as a special case of collaborative learning methods. After discussing its limitation, we propose an extended, more flexible formulation, and introduce a simple and efficient alternating minimization algorithm to solve the associated optimization problem. We explore the parameter space and provide some guidelines regarding parameter selection. Both the original and extended versions are then compared on a simple toy data set and a more advanced simulated imaging genomics data set in order to illustrate the benefits of the latter. Finally, we validate the proposed formulation using single nucleotide polymorphisms data and functional magnetic resonance imaging data from a population of adolescents ( subjects, age 16.9 ± 1.9 years from the Philadelphia Neurodevelopmental Cohort) for the study of learning ability. Furthermore, we carry out a significance analysis of the resulting features that allow us to carefully extract brain regions and genes linked to learning and cognitive ability.

Fused Estimation of Sparse Connectivity Patterns From Rest fMRI-Application to Comparison of Children and Adult Brains.

In this paper, we consider the problem of estimating multiple sparse, co-activated brain regions from functional magnetic resonance imaging (fMRI) observations belonging to different classes. More precisely, we propose a method to analyze similarities and differences in functional connectivity between children and young adults. Often, analysis is conducted on each class separately, and differences across classes are identified with an additional postprocessing step using adequate statistical tools. Here, we propose to rely on a generalized fused Lasso penalty, which allows us to make use of the entire data set in order to estimate connectivity patterns that are either shared across classes, or specific to a given group. By using the entire population during the estimation, we hope to increase the power of our analysis. The proposed model falls in the category of population-wise matrix decomposition, and a simple and efficient alternating direction method of multipliers algorithm is introduced to solve the associated optimization problem. After validating our approach on simulated data, experiments are performed on resting-state fMRI imaging from the Philadelphia neurodevelopmental cohort data set, comprised of normally developing children from ages 8 to 21. Developmental differences were observed in various brain regions, as a total of three class-specific resting-state components were identified. Statistical analysis of the estimated subject-specific features, as well as classification results (based on age groups, up to 81% accuracy, samples) related to these components demonstrate that the proposed method is able to properly extract meaningful shared and class-specific sub-networks.

ENFORCING CO-EXPRESSION IN MULTIMODAL REGRESSION FRAMEWORK.

We consider the problem of multimodal data integration for the study of complex neurological diseases (e.g. schizophrenia). Among the challenges arising in such situation, estimating the link between genetic and neurological variability within a population sample has been a promising direction. A wide variety of statistical models arose from such applications. For example, Lasso regression and its multitask extension are often used to fit a multivariate linear relationship between given phenotype(s) and associated observations. Other approaches, such as canonical correlation analysis (CCA), are widely used to extract relationships between sets of variables from different modalities. In this paper, we propose an exploratory multivariate method combining these two methods. More Specifically, we rely on a 'CCA-type' formulation in order to regularize the classical multimodal Lasso regression problem. The underlying motivation is to extract discriminative variables that display are also co-expressed across modalities. We first evaluate the method on a simulated dataset, and further validate it using Single Nucleotide Polymorphisms (SNP) and functional Magnetic Resonance Imaging (fMRI) data for the study of schizophrenia.

Observation model based on scale interactions for optical flow estimation.

In this paper, an original observation model for multiresolution optical flow estimation is introduced. Multiresolution frameworks, often based on coarse-to-fine warping strategies, are widely used by state-of-the-art optical flow methods. They allow the recovery of large motions by successive estimations of the flow field at several resolution levels. Although such approaches perform very efficiently and usually lead to faster minimizations, they generally consider independent problems at each resolution levels and do not exploit the existing interactions between scales (especially the influences of fine scales on larger ones). In this paper, we tackle this issue by proposing a flexible framework, inspired from fluid mechanics, able to partly counter these limitations. For each resolution level, our process filters the equations of interest and decomposes the key variables into resolved (i.e., at a given resolution) and unresolved (i.e., at finer resolutions) components. This enables to derive a new data term that takes into account, at coarse resolutions, the influence of their unresolved parts. From this new term, we propose two different estimation strategies, depending on whether we explicitly know the type of relations between the different scales (as for physical processes) or not. In order to test the efficiency of this new observation model, we have embedded it in a simple multiresolution Lucas-Kanade estimator. Comparing the usual optical flow constraint equation with this new term in the same motion estimation procedure, it clearly appears that the proposed term leads to more consistent estimates and prevents from errors propagation apparition during the estimation. In all situations (synthetic, real, physical images or not), our new term is able to greatly improve the results compared with usual conservation constraints.