Search

Women with an increased life-time risk of breast cancer undergo supplemental annual screening MRI. We propose to predict the risk of developing breast cancer within one year based on the current MRI, with the objective of reducing screening burden and facilitating early detection. An AI algorithm was developed on 53,858 breasts from 12,694 patients who underwent screening or diagnostic MRI and accrued over 12 years, with 2,331 confirmed cancers. A first U-Net was trained to segment lesions and identify regions of concern. A second convolutional network was trained to detect malignant cancer using features extracted by the U-Net. This network was then fine-tuned to estimate the risk of developing cancer within a year in cases that radiologists considered normal or likely benign. Risk predictions from this AI were evaluated with a retrospective analysis of 9,183 breasts from a high-risk screening cohort, which were not used for training. Statistical analysis focused on the tradeoff between number of omitted exams versus negative predictive value, and number of potential early detections versus positive predictive value. The AI algorithm identified regions of concern that coincided with future tumors in 52% of screen-detected cancers. Upon directed review, a radiologist found that 71.3% of cancers had a visible correlate on the MRI prior to diagnosis, 65% of these correlates were identified by the AI model. Reevaluating these regions in 10% of all cases with higher AI-predicted risk could have resulted in up to 33% early detections by a radiologist. Additionally, screening burden could have been reduced in 16% of lower-risk cases by recommending a later follow-up without compromising current interval cancer rate. With increasing datasets and improving image quality we expect this new AI-aided, adaptive screening to meaningfully reduce screening burden and improve early detection., Comment: Major revisions and rewriting in progress

Purpose: To develop a deep network architecture that would achieve fully automated radiologist-level segmentation of cancers at breast MRI. Materials and Methods: In this retrospective study, 38229 examinations (composed of 64063 individual breast scans from 14475 patients) were performed in female patients (age range, 12-94 years; mean age, 52 years +/- 10 [standard deviation]) who presented between 2002 and 2014 at a single clinical site. A total of 2555 breast cancers were selected that had been segmented on two-dimensional (2D) images by radiologists, as well as 60108 benign breasts that served as examples of noncancerous tissue; all these were used for model training. For testing, an additional 250 breast cancers were segmented independently on 2D images by four radiologists. Authors selected among several three-dimensional (3D) deep convolutional neural network architectures, input modalities, and harmonization methods. The outcome measure was the Dice score for 2D segmentation, which was compared between the network and radiologists by using the Wilcoxon signed rank test and the two one-sided test procedure. Results: The highest-performing network on the training set was a 3D U-Net with dynamic contrast-enhanced MRI as input and with intensity normalized for each examination. In the test set, the median Dice score of this network was 0.77 (interquartile range, 0.26). The performance of the network was equivalent to that of the radiologists (two one-sided test procedures with radiologist performance of 0.69-0.84 as equivalence bounds, P <= .001 for both; n = 250). Conclusion: When trained on a sufficiently large dataset, the developed 3D U-Net performed as well as fellowship-trained radiologists in detailed 2D segmentation of breast cancers at routine clinical MRI.

Purpose: Conventional automated segmentation of the head anatomy in MRI distinguishes different brain and non-brain tissues based on image intensities and prior tissue probability maps (TPM). This works well for normal head anatomies, but fails in the presence of unexpected lesions. Deep convolutional neural networks leverage instead spatial patterns and can learn to segment lesions, but often ignore prior probabilities. Approach: We add three sources of prior information to a three-dimensional convolutional network, namely, spatial priors with a TPM, morphological priors with conditional random fields, and spatial context with a wider field-of-view at lower resolution. We train and test these networks on 3D images of 43 stroke patients and 4 healthy individuals which have been manually segmented. Results: We demonstrate the benefits of each sources of prior information, and we show that the new architecture, which we call Multiprior network, improves the performance of existing segmentation software, such as SPM, FSL, and DeepMedic for abnormal anatomies. The relevance of the different priors was compared and the TPM was found to be most beneficial. The benefit of adding a TPM is generic in that it can boost the performance of established segmentation networks such as the DeepMedic and a UNet. We also provide an out-of-sample validation and clinical application of the approach on an additional 47 patients with disorders of consciousness. We make the code and trained networks freely available. Conclusions: Biomedical images follow imaging protocols that can be leveraged as prior information into deep convolutional neural networks to improve performance. The network segmentations match human manual corrections performed in 3D, and are comparable in performance to human segmentations obtained from scratch in 2D for abnormal brain anatomies.

Neural development is generally marked by an increase in the efficiency and diversity of neural processes. In a large sample ( = 114) of human children and adults with ages ranging from 5 to 44 yr, we investigated the neural responses to naturalistic video stimuli. Videos from both real-life classroom settings and Hollywood feature films were used to probe different aspects of attention and engagement. For all stimuli, older ages were marked by more variable neural responses. Variability was assessed by the intersubject correlation of evoked electroencephalographic responses. Young males also had less-variable responses than young females. These results were replicated in an independent cohort ( = 303). When interpreted in the context of neural maturation, we conclude that neural function becomes more variable with maturity, at least during the passive viewing of real-world stimuli.

Global integration of information in the brain results from complex interactions of segregated brain networks. Identifying the most influential neuronal populations that efficiently bind these networks is a fundamental problem of systems neuroscience. Here, we apply optimal percolation theory and pharmacogenetic interventions in vivo to predict and subsequently target nodes that are essential for global integration of a memory network in rodents. The theory predicts that integration in the memory network is mediated by a set of low-degree nodes located in the nucleus accumbens. This result is confirmed with pharmacogenetic inactivation of the nucleus accumbens, which eliminates the formation of the memory network, while inactivations of other brain areas leave the network intact. Thus, optimal percolation theory predicts essential nodes in brain networks. This could be used to identify targets of interventions to modulate brain function.

Global integration of information in the brain results from complex interactions of segregated brain networks. Identifying the most influential neuronal populations that efficiently bind these networks is a fundamental problem of systems neuroscience. Here, we apply optimal percolation theory and pharmacogenetic interventions in vivo to predict and subsequently target nodes that are essential for global integration of a memory network in rodents. The theory predicts that integration in the memory network is mediated by a set of low-degree nodes located in the nucleus accumbens. This result is confirmed with pharmacogenetic inactivation of the nucleus accumbens, which eliminates the formation of the memory network, while inactivations of other brain areas leave the network intact. Thus, optimal percolation theory predicts essential nodes in brain networks. This could be used to identify targets of interventions to modulate brain function.

Brain data recorded with electroencephalography (EEG), magnetoencephalography (MEG) and related techniques often have poor signal-to-noise ratios due to the presence of multiple competing sources and artifacts. A common remedy is to average responses over repeats of the same stimulus, but this is not applicable for temporally extended stimuli that are presented only once (speech, music, movies, natural sound). An alternative is to average responses over multiple subjects that were presented with identical stimuli, but differences in geometry of brain sources and sensors reduce the effectiveness of this solution. Multiway canonical correlation analysis (MCCA) brings a solution to this problem by allowing data from multiple subjects to be fused in such a way as to extract components common to all. This paper reviews the method, offers application examples that illustrate its effectiveness, and outlines the caveats and risks entailed by the method.

How does one find dimensions in multivariate data that are reliably expressed across repetitions? For example, in a brain imaging study one may want to identify combinations of neural signals that are reliably expressed across multiple trials or subjects. For a behavioral assessment with multiple ratings, one may want to identify an aggregate score that is reliably reproduced across raters. Correlated Components Analysis (CorrCA) addresses this problem by identifying components that are maximally correlated between repetitions (e.g. trials, subjects, raters). Here we formalize this as the maximization of the ratio of between-repetition to within-repetition covariance. We show that this criterion maximizes repeat-reliability, defined as mean over variance across repeats, and that it leads to CorrCA or to multi-set Canonical Correlation Analysis, depending on the constraints. Surprisingly, we also find that CorrCA is equivalent to Linear Discriminant Analysis for zero-mean signals, which provides an unexpected link between classic concepts of multivariate analysis. We present an exact parametric test of statistical significance based on the F-statistic for normally distributed independent samples, and present and validate shuffle statistics for the case of dependent samples. Regularization and extension to non-linear mappings using kernels are also presented. The algorithms are demonstrated on a series of data analysis applications, and we provide all code and data required to reproduce the results.

We performed simultaneous recordings of electroencephalography (EEG) from multiple students in a classroom, and measured the inter-subject correlation (ISC) of activity evoked by a common video stimulus. The neural reliability, as quantified by ISC, has been linked to engagement and attentional modulation in earlier studies that used high-grade equipment in laboratory settings. Here we reproduce many of the results from these studies using portable low-cost equipment, focusing on the robustness of using ISC for subjects experiencing naturalistic stimuli. The present data shows that stimulus-evoked neural responses, known to be modulated by attention, can be tracked for groups of students with synchronized EEG acquisition. This is a step towards real-time inference of engagement in the classroom.

We present a dataset combining electrophysiology and eye tracking intended as a resource for the investigation of information processing in the developing brain. The dataset includes high-density task-based and task-free EEG, eye tracking, and cognitive and behavioral data collected from 126 individuals (ages: 6-44). The task battery spans both the simple/complex and passive/active dimensions to cover a range of approaches prevalent in modern cognitive neuroscience. The active task paradigms facilitate principled deconstruction of core components of task performance in the developing brain, whereas the passive paradigms permit the examination of intrinsic functional network activity during varying amounts of external stimulation. Alongside these neurophysiological data, we include an abbreviated cognitive test battery and questionnaire-based measures of psychiatric functioning. We hope that this dataset will lead to the development of novel assays of neural processes fundamental to information processing, which can be used to index healthy brain development as well as detect pathologic processes.

We performed simultaneous recordings of electroencephalography (EEG) from multiple students in a classroom, and measured the inter-subject correlation (ISC) of activity evoked by a common video stimulus. The neural reliability, as quantified by ISC, has been linked to engagement and attentional modulation in earlier studies that used high-grade equipment in laboratory settings. Here we reproduce many of the results from these studies using portable low-cost equipment, focusing on the robustness of using ISC for subjects experiencing naturalistic stimuli. The present data shows that stimulus-evoked neural responses, known to be modulated by attention, can be tracked in for groups of students with synchronized EEG acquisition. This is a step towards real-time inference of engagement in the classroom., Comment: 14 pages, 5 figures, 3 tables. Preprint version. Revision of original preprint. Supplementary materials added as ancillary file

We present a dataset combining electrophysiology and eye tracking intended as a resource for the investigation of information processing in the developing brain. The dataset includes high-density task-based and task-free EEG, eye tracking, and cognitive and behavioral data collected from 126 individuals (ages: 6-44). The task battery spans both the simple/complex and passive/active dimensions to cover a range of approaches prevalent in modern cognitive neuroscience. The active task paradigms facilitate principled deconstruction of core components of task performance in the developing brain, whereas the passive paradigms permit the examination of intrinsic functional network activity during varying amounts of external stimulation. Alongside these neurophysiological data, we include an abbreviated cognitive test battery and questionnaire-based measures of psychiatric functioning. We hope that this dataset will lead to the development of novel assays of neural processes fundamental to information processing, which can be used to index healthy brain development as well as detect pathologic processes.

Recent efforts have aimed to characterize clinical pediatric populations by using neurophysiological tests in addition to behavioral assays. Here we report on a data collection effort in which electroencephalography (EEG) was recorded in both juveniles and adults (N=114 participants, ages 6-44 years of age) during various stimulation protocols. The present analysis focuses on how neural responses during passive viewing of naturalistic videos vary with age and sex, and in particular, how similar they are within developmental groups. Similarity of neural responses was measured as the inter-subject correlation of the EEG. Stimulus-evoked neural responses are more similar among children and decrease in similarity with age. Among children, males respond more similarly to each other than females. This was uniformly true for a variety of videos. The decrease in group similarity with age may result from an overall decline in the magnitude of evoked responses, but this cannot explain the sex differences found in the young. We therefore propose that as children mature, neural function may become more variable.

The probability distribution of number of ties of an individual in a social network follows a scale-free power-law. However, how this distribution arises has not been conclusively demonstrated in direct analyses of people's actions in social networks. Here, we perform a causal inference analysis and find an underlying cause for this phenomenon. Our analysis indicates that heavy-tailed degree distribution is causally determined by similarly skewed distribution of human activity. Specifically, the degree of an individual is entirely random - following a "maximum entropy attachment" model - except for its mean value which depends deterministically on the volume of the users' activity. This relation cannot be explained by interactive models, like preferential attachment, since the observed actions are not likely to be caused by interactions with other people., Comment: 23 pages, 5 figures

During the past decades, much attention has been given to the separation of mixed sources, in particular for the blind case where both the sources and the mixing process are unknown and only recordings of the mixtures are available. In several situations it is desirable to recover all sources from the recorded mixtures, or at least to segregate a particular source. Furthermore, it may be useful to identify the mixing process itself to reveal information about the physical mixing system. In some simple mixing models each recording consists of a sum of differently weighted source signals. However, in many real-world applications, such as in acoustics, the mixing process is more complex. In such systems, the mixtures are weighted and delayed, and each source contributes to the sum with multiple delays corresponding to the multiple paths by which an acoustic signal propagates to a microphone. Such filtered sums of different sources are called convolutive mixtures. There are already a number of partial reviews available on this topic so the purpose of this chapter is to provide a complete survey of convolutive BSS and identify a taxonomy that can organize the large number of available algorithms. This may help practitioners and researchers new to the area of convolutive source separation obtain a complete overview of the field. Hopefully those with more experience in the field can identify useful tools, or find inspiration for new algorithms.

During the past decades, much attention has been given to the separation of mixed sources, in particular for the blind case where both the sources and the mixing process are unknown and only recordings of the mixtures are available. In several situations it is desirable to recover all sources from the recorded mixtures, or at least to segregate a particular source. Furthermore, it may be useful to identify the mixing process itself to reveal information about the physical mixing system. In some simple mixing models each recording consists of a sum of differently weighted source signals. However, in many real-world applications, such as in acoustics, the mixing process is more complex. In such systems, the mixtures are weighted and delayed, and each source contributes to the sum with multiple delays corresponding to the multiple paths by which an acoustic signal propagates to a microphone. Such filtered sums of different sources are called convolutive mixtures. There are already a number of partial reviews available on this topic so the purpose of this chapter is to provide a complete survey of convolutive BSS and identify a taxonomy that can organize the large number of available algorithms. This may help practitioners and researchers new to the area of convolutive source separation obtain a complete overview of the field. Hopefully those with more experience in the field can identify useful tools, or find inspiration for new algorithms.

During the past decades, much attention has been given to the separation of mixed sources, in particular for the blind case where both the sources and the mixing process are unknown and only recordings of the mixtures are available. In several situations it is desirable to recover all sources from the recorded mixtures, or at least to segregate a particular source. Furthermore, it may be useful to identify the mixing process itself to reveal information about the physical mixing system. In some simple mixing models each recording consists of a sum of differently weighted source signals. However, in many real-world applications, such as in acoustics, the mixing process is more complex. In such systems, the mixtures are weighted and delayed, and each source contributes to the sum with multiple delays corresponding to the multiple paths by which an acoustic signal propagates to a microphone. Such filtered sums of different sources are called convolutive mixtures. There are already a number of partial reviews available on this topic so the purpose of this chapter is to provide a complete survey of convolutive BSS and identify a taxonomy that can organize the large number of available algorithms. This may help practitioners and researchers new to the area of convolutive source separation obtain a complete overview of the field. Hopefully those with more experience in the field can identify useful tools, or find inspiration for new algorithms.

Phenomena resembling tinnitus and Zwicker phantom tone are seen to result from an auditory gain adaptation mechanism that attempts to make full use of a fixed-capacity channel. In the case of tinnitus, the gain adaptation enhances internal noise of a frequency band otherwise silent due to damage. This generates a percept of a phantom sound as a consequence of hearing loss. In the case of Zwicker tone, a frequency band is temporarily silent during the presentation of a notched broad-band sound, resulting in a percept of a tone at the notched frequency. The model suggests a link between tinnitus and the Zwicker tone percept, in that it predicts different results for normal and tinnitus subjects due to a loss of instantaneous nonlinear compression. Listening experiments on 44 subjects show that tinnitus subjects (11 of 44) are significantly more likely to hear the Zwicker tone. This psychoacoustic experiment establishes the first empirical link between the Zwicker tone percept and tinnitus. Together with the modeling results, this supports the hypothesis that the phantom percept is a consequence of a central adaptation mechanism confronted with a degraded sensory apparatus.

Phenomena resembling tinnitus and Zwicker phantom tone are seen to result from an auditory gain adaptation mechanism that attempts to make full use of a fixed-capacity channel. In the case of tinnitus, the gain adaptation enhances internal noise of a frequency band otherwise silent due to damage. This generates a percept of a phantom sound as a consequence of hearing loss. In the case of Zwicker tone, a frequency band is temporarily silent during the presentation of a notched broad-band sound, resulting in a percept of a tone at the notched frequency. The model suggests a link between tinnitus and the Zwicker tone percept, in that it predicts different results for normal and tinnitus subjects due to a loss of instantaneous nonlinear compression. Listening experiments on 44 subjects show that tinnitus subjects (11 of 44) are significantly more likely to hear the Zwicker tone. This psychoacoustic experiment establishes the first empirical link between the Zwicker tone percept and tinnitus. Together with the modeling results, this supports the hypothesis that the phantom percept is a consequence of a central adaptation mechanism confronted with a degraded sensory apparatus.

Conventional electroencephalography (EEG) and magnetoencephalography (MEG) analysis often rely on averaging over multiple trials to extract statistically relevant di7erences between two or more experimental conditions. We demonstrate that by linearly integrating information over multiple spatially distributed sensors within a prede9ned time window, one can discriminate conditions on a trial-by-trial basis with high accuracy. We restrict ourselves to a linear integration as it allows the computation of a spatial distribution of the discriminating source activity. In the present set of experiments the resulting source activity distributions correspond to functional neuroanatomy consistent with the task (e.g. contralateral sensory-motor cortex and anterior cingulate).

Conventional electroencephalography (EEG) and magnetoencephalography (MEG) analysis often rely on averaging over multiple trials to extract statistically relevant di7erences between two or more experimental conditions. We demonstrate that by linearly integrating information over multiple spatially distributed sensors within a prede9ned time window, one can discriminate conditions on a trial-by-trial basis with high accuracy. We restrict ourselves to a linear integration as it allows the computation of a spatial distribution of the discriminating source activity. In the present set of experiments the resulting source activity distributions correspond to functional neuroanatomy consistent with the task (e.g. contralateral sensory-motor cortex and anterior cingulate).

Source separation arises in a surprising number of signal processing applications, from speech recognition to EEG analysis. In the square linear blind source separation problem without time delays, one must find an unmixing matrix which can detangle the result of mixing n unknown independent sources through an unknown n x n mixing matrix. The recently introduced ICA blind source separation algorithm (Baram and Roth 1994; Bell and Sejnowski 1995) is a powerful and surprisingly simple technique for solving this problem. ICA is all the more remarkable for performing so well despite making absolutely no use of the temporal structure of its input! This paper presents a new algorithm, contextual ICA, which derives from a maximum likelihood density estimation formulation of the problem. cICA can incorporate arbitrarily complex adaptive history-sensitive source models, and thereby make use of the temporal structure of its input. This allows it to separate in a number of situations where standard ICA cannot, including sources with low kurtosis, colored gaussian sources, and sources which have gaussian histograms. Since ICA is a special case of cICA, the MLE derivation provides as a corollary a rigorous derivation of classic ICA.

Source separation arises in a surprising number of signal processing applications, from speech recognition to EEG analysis. In the square linear blind source separation problem without time delays, one must find an unmixing matrix which can detangle the result of mixing n unknown independent sources through an unknown n x n mixing matrix. The recently introduced ICA blind source separation algorithm (Baram and Roth 1994; Bell and Sejnowski 1995) is a powerful and surprisingly simple technique for solving this problem. ICA is all the more remarkable for performing so well despite making absolutely no use of the temporal structure of its input! This paper presents a new algorithm, contextual ICA, which derives from a maximum likelihood density estimation formulation of the problem. cICA can incorporate arbitrarily complex adaptive history-sensitive source models, and thereby make use of the temporal structure of its input. This allows it to separate in a number of situations where standard ICA cannot, including sources with low kurtosis, colored gaussian sources, and sources which have gaussian histograms. Since ICA is a special case of cICA, the MLE derivation provides as a corollary a rigorous derivation of classic ICA.