Clinically Applicable Segmentation of Head and Neck Anatomy for Radiotherapy: Deep Learning Algorithm Development and Validation Study.

BACKGROUND: Over half a million individuals are diagnosed with head and neck cancer each year globally. Radiotherapy is an important curative treatment for this disease, but it requires manual time to delineate radiosensitive organs at risk. This planning process can delay treatment while also introducing interoperator variability, resulting in downstream radiation dose differences. Although auto-segmentation algorithms offer a potentially time-saving solution, the challenges in defining, quantifying, and achieving expert performance remain. OBJECTIVE: Adopting a deep learning approach, we aim to demonstrate a 3D U-Net architecture that achieves expert-level performance in delineating 21 distinct head and neck organs at risk commonly segmented in clinical practice. METHODS: The model was trained on a data set of 663 deidentified computed tomography scans acquired in routine clinical practice and with both segmentations taken from clinical practice and segmentations created by experienced radiographers as part of this research, all in accordance with consensus organ at risk definitions. RESULTS: We demonstrated the model's clinical applicability by assessing its performance on a test set of 21 computed tomography scans from clinical practice, each with 21 organs at risk segmented by 2 independent experts. We also introduced surface Dice similarity coefficient, a new metric for the comparison of organ delineation, to quantify the deviation between organ at risk surface contours rather than volumes, better reflecting the clinical task of correcting errors in automated organ segmentations. The model's generalizability was then demonstrated on 2 distinct open-source data sets, reflecting different centers and countries to model training. CONCLUSIONS: Deep learning is an effective and clinically applicable technique for the segmentation of the head and neck anatomy for radiotherapy. With appropriate validation studies and regulatory approvals, this system could improve the efficiency, consistency, and safety of radiotherapy pathways.

The influence of microsatellite polymorphisms in sex steroid receptor genes ESR1, ESR2 and AR on sex differences in brain structure.

The androgen receptor (AR), oestrogen receptor alpha (ESR1) and oestrogen receptor beta (ESR2) play essential roles in mediating the effect of sex hormones on sex differences in the brain. Using Voxel-based morphometry (VBM) and gene sizing in two independent samples (discovery n â€‹= â€‹173, replication â€‹= â€‹61), we determine the common and unique influences on brain sex differences in grey (GM) and white matter (WM) volume between repeat lengths (n) of microsatellite polymorphisms AR(CAG)n, ESR1(TA)n and ESR2(CA)n. In the hypothalamus, temporal lobes, anterior cingulate cortex, posterior insula and prefrontal cortex, we find increased GM volume with increasing AR(CAG)n across sexes, decreasing ESR1(TA)n across sexes and decreasing ESR2(CA)n in females. Uniquely, AR(CAG)n was positively associated with dorsolateral prefrontal and orbitofrontal GM volume and the anterior corona radiata, left superior fronto-occipital fasciculus, thalamus and internal capsule WM volume. ESR1(TA)n was negatively associated with the left superior corona radiata, left cingulum and left inferior longitudinal fasciculus WM volume uniquely. ESR2(CA)n was negatively associated with right fusiform and posterior cingulate cortex uniquely. We thus describe the neuroanatomical correlates of three microsatellite polymorphisms of steroid hormone receptors and their relationship to sex differences.

The Neuro Bureau ADHD-200 Preprocessed repository.

In 2011, the "ADHD-200 Global Competition" was held with the aim of identifying biomarkers of attention-deficit/hyperactivity disorder from resting-state functional magnetic resonance imaging (rs-fMRI) and structural MRI (s-MRI) data collected on 973 individuals. Statisticians and computer scientists were potentially the most qualified for the machine learning aspect of the competition, but generally lacked the specialized skills to implement the necessary steps of data preparation for rs-fMRI. Realizing this barrier to entry, the Neuro Bureau prospectively collaborated with all competitors by preprocessing the data and sharing these results at the Neuroimaging Informatics Tools and Resources Clearinghouse (NITRC) (http://www.nitrc.org/frs/?group_id=383). This "ADHD-200 Preprocessed" release included multiple analytical pipelines to cater to different philosophies of data analysis. The processed derivatives included denoised and registered 4D fMRI volumes, regional time series extracted from brain parcellations, maps of 10 intrinsic connectivity networks, fractional amplitude of low frequency fluctuation, and regional homogeneity, along with grey matter density maps. The data was used by several teams who competed in the ADHD-200 Global Competition, including the winning entry by a group of biostaticians. To the best of our knowledge, the ADHD-200 Preprocessed release was the first large public resource of preprocessed resting-state fMRI and structural MRI data, and remains to this day the only resource featuring a battery of alternative processing paths.

Face-selective regions differ in their ability to classify facial expressions.

Recognition of facial expressions is crucial for effective social interactions. Yet, the extent to which the various face-selective regions in the human brain classify different facial expressions remains unclear. We used functional magnetic resonance imaging (fMRI) and support vector machine pattern classification analysis to determine how well face-selective brain regions are able to decode different categories of facial expression. Subjects participated in a slow event-related fMRI experiment in which they were shown 32 face pictures, portraying four different expressions: neutral, fearful, angry, and happy and belonging to eight different identities. Our results showed that only the amygdala and the posterior superior temporal sulcus (STS) were able to accurately discriminate between these expressions, albeit in different ways: the amygdala discriminated fearful faces from non-fearful faces, whereas STS discriminated neutral from emotional (fearful, angry and happy) faces. In contrast to these findings on the classification of emotional expression, only the fusiform face area (FFA) and anterior inferior temporal cortex (aIT) could discriminate among the various facial identities. Further, the amygdala and STS were better than FFA and aIT at classifying expression, while FFA and aIT were better than the amygdala and STS at classifying identity. Taken together, our findings indicate that the decoding of facial emotion and facial identity occurs in different neural substrates: the amygdala and STS for the former and FFA and aIT for the latter.

Automated analysis of retinal imaging using machine learning techniques for computer vision.

There are almost two million people in the United Kingdom living with sight loss, including around 360,000 people who are registered as blind or partially sighted. Sight threatening diseases, such as diabetic retinopathy and age related macular degeneration have contributed to the 40% increase in outpatient attendances in the last decade but are amenable to early detection and monitoring. With early and appropriate intervention, blindness may be prevented in many cases. Ophthalmic imaging provides a way to diagnose and objectively assess the progression of a number of pathologies including neovascular ("wet") age-related macular degeneration (wet AMD) and diabetic retinopathy. Two methods of imaging are commonly used: digital photographs of the fundus (the 'back' of the eye) and Optical Coherence Tomography (OCT, a modality that uses light waves in a similar way to how ultrasound uses sound waves). Changes in population demographics and expectations and the changing pattern of chronic diseases creates a rising demand for such imaging. Meanwhile, interrogation of such images is time consuming, costly, and prone to human error. The application of novel analysis methods may provide a solution to these challenges. This research will focus on applying novel machine learning algorithms to automatic analysis of both digital fundus photographs and OCT in Moorfields Eye Hospital NHS Foundation Trust patients. Through analysis of the images used in ophthalmology, along with relevant clinical and demographic information, DeepMind Health will investigate the feasibility of automated grading of digital fundus photographs and OCT and provide novel quantitative measures for specific disease features and for monitoring the therapeutic success.

Spatial selectivity in the temporoparietal junction, inferior frontal sulcus, and inferior parietal lobule.

Spatial selectivity, as measured by functional magnetic resonance imaging (fMRI) activity patterns that vary consistently with the location of visual stimuli, has been documented in many human brain regions, notably the occipital visual cortex and the frontal and parietal regions that are active during endogenous, goal-directed attention. We hypothesized that spatial selectivity also exists in regions that are active during exogenous, stimulus-driven attention. To test this hypothesis, we acquired fMRI data while subjects maintained passive fixation. At jittered time intervals, a briefly presented wedge-shaped array of rapidly expanding circles appeared at one of three contralateral or one of three ipsilateral locations. Positive fMRI activations were identified in multiple brain regions commonly associated with exogenous attention, including the temporoparietal junction, the inferior parietal lobule, and the inferior frontal sulcus. These activations were not organized as a map across the cortical surface. However, multivoxel pattern analysis of the fMRI activity correctly classified every pair of stimulus locations, demonstrating that patterns of fMRI activity were correlated with spatial location. These observations held for both contralateral and ipsilateral stimulus pairs as well as for stimuli of different textures (radial checkerboard) and shapes (squares and rings). Permutation testing verified that the obtained accuracies were not due to systematic biases and demonstrated that the findings were statistically significant.

Investigating the use of support vector machine classification on structural brain images of preterm-born teenagers as a biological marker.

Preterm birth has been shown to induce an altered developmental trajectory of brain structure and function. With the aid support vector machine (SVM) classification methods we aimed to investigate whether MRI data, collected in adolescence, could be used to predict whether an individual had been born preterm or at term. To this end we collected T1-weighted anatomical MRI data from 143 individuals (69 controls, mean age 14.6y). The inclusion criteria for those born preterm were birth weight ≤ 1500g and gestational age < 37w. A linear SVM was trained on the grey matter segment of MR images in two different ways. First, all the individuals were used for training and classification was performed by the leave-one-out method, yielding 93% correct classification (sensitivity = 0.905, specificity = 0.942). Separately, a random half of the available data were used for training twice and each time the other, unseen, half of the data was classified, resulting 86% and 91% accurate classifications. Both gestational age (R = -0.24, p<0.04) and birth weight (R = -0.51, p < 0.001) correlated with the distance to decision boundary within the group of individuals born preterm. Statistically significant correlations were also found between IQ (R = -0.30, p < 0.001) and the distance to decision boundary. Those born small for gestational age did not form a separate subgroup in these analyses. The high rate of correct classification by the SVM motivates further investigation. The long-term goal is to automatically and non-invasively predict the outcome of preterm-born individuals on an individual basis using as early a scan as possible.

Quantitative magnetic resonance imaging evidence for altered structural remodeling of the temporal lobe in West syndrome.

OBJECTIVE: To explore the structure-function relation of the temporal lobe in newly diagnosed West syndrome of unknown cause (uWS). METHODS: Quantitative magnetic resonance imaging (three-dimensional [3D] structural MRI and diffusion tensor imaging [DTI]) was analyzed using voxel-based morphometry (VBM) and tract-based spatial statistics (TBSS) in 22 patients and healthy age-matched controls. The electrophysiologic responsiveness of the temporal lobe was measured using the N100 auditory event-related potential (aERP) to a repeated 1,000 Hz tone. Neurocognitive function was assessed using the Bayley Scales of Infant Development, Second Edition (BSID-II). Tests followed first-line treatment with vigabatrin (17 patients) or high-dose oral prednisolone (5 patients). RESULTS: Total temporal lobe volume was similar in patients and controls. Patients had a smaller temporal stem (TS) (p < 0.0001) and planum temporale (PT) (p = 0.029) bilaterally. TS width asymmetry with a larger right-sided width in controls was absent in patients (p = 0.033). PT asymmetry was present in both groups, being larger on the right (p = 0.048). VBM gray matter volume was increased at the left temporal lobe (superior and middle temporal gyri, the peri-rhinal cortex, and medial temporal lobe) (p < 0.005, family wise error-corrected). VBM gray matter volume correlated with the duration of infantile spasms (Pearson's r = -0.630, p = 0.009). DTI metrics did not differ between patients and controls on TBSS. Mean BSID-II scores were lower (p < 0.001) and auditory N100 ERP attenuated less in patients than in controls (p = 0.002). SIGNIFICANCE: The functional networking and white matter development of the temporal lobe are impaired following infantile spasms. Treatment may promote structural plasticity within the temporal lobe following infantile spasms, manifest as increased gray matter volume on VBM. It remains to be investigated further whether this predicts patients' long-term cognitive difficulties.

Foraging under competition: the neural basis of input-matching in humans.

Input-matching is a key mechanism by which animals optimally distribute themselves across habitats to maximize net gains based on the changing input values of food supply rate and competition. To examine the neural systems that underlie this rule in humans, we created a continuous-input foraging task where subjects had to decide to stay or switch between two habitats presented on the left and right of the screen. The subject's decision to stay or switch was based on changing input values of reward-token supply rate and competition density. High density of competition or low-reward token rate was associated with decreased chance of winning. Therefore, subjects attempted to maximize their gains by switching to habitats that possessed low competition density and higher token rate. When it was increasingly disadvantageous to be in a habitat, we observed increased activity in brain regions that underlie preparatory motor actions, including the dorsal anterior cingulate cortex and the supplementary motor area, as well as the insula, which we speculate may be involved in the conscious urge to switch habitats. Conversely, being in an advantageous habitat is associated with activity in the reward systems, namely the striatum and medial prefrontal cortex. Moreover, amygdala and dorsal putamen activity steered interindividual preferences in competition avoidance and pursuing reward. Our results suggest that input-matching decisions are made as a net function of activity in a distributed set of neural systems. Furthermore, we speculate that switching behaviors are related to individual differences in competition avoidance and reward drive.

Effects of image contrast on functional MRI image registration.

Lack of tissue contrast and existing inhomogeneous bias fields from multi-channel coils have the potential to degrade the output of registration algorithms; and consequently degrade group analysis and any attempt to accurately localize brain function. Non-invasive ways to improve tissue contrast in fMRI images include the use of low flip angles (FAs) well below the Ernst angle and longer repetition times (TR). Techniques to correct intensity inhomogeneity are also available in most mainstream fMRI data analysis packages; but are not used as part of the pre-processing pipeline in many studies. In this work, we use a combination of real data and simulations to show that simple-to-implement acquisition/pre-processing techniques can significantly improve the outcome of both functional-to-functional and anatomical-to-functional image registrations. We also emphasize the need of tissue contrast on EPI images to be able to appropriately evaluate the quality of the alignment. In particular, we show that the use of low FAs (e.g., θ≤40°), when physiological noise considerations permit such an approach, significantly improves accuracy, consistency and stability of registration for data acquired at relatively short TRs (TR≤2s). Moreover, we also show that the application of bias correction techniques significantly improves alignment both for array-coil data (known to contain high intensity inhomogeneity) as well as birdcage-coil data. Finally, improvements in alignment derived from the use of the first infinite-TR volumes (ITVs) as targets for registration are also demonstrated. For the purpose of quantitatively evaluating the different scenarios, two novel metrics were developed: Mean Voxel Distance (MVD) to evaluate registration consistency, and Deviation of Mean Voxel Distance (dMVD) to evaluate registration stability across successive alignment attempts.

Decreased premotor cortex volume in victims of urban violence with posttraumatic stress disorder.

BACKGROUND: Studies addressing posttraumatic stress disorder (PTSD) have demonstrated that PTSD patients exhibit structural abnormalities in brain regions that relate to stress regulation and fear responses, such as the hippocampus, amygdala, anterior cingulate cortex, and ventromedial prefrontal cortex. Premotor cortical areas are involved in preparing to respond to a threatening situation and in representing the peripersonal space. Urban violence is an important and pervasive cause of human suffering, especially in large urban centers in the developing world. Violent events, such as armed robbery, are very frequent in certain cities, and these episodes increase the risk of PTSD. Assaultive trauma is characterized by forceful invasion of the peripersonal space; therefore, could this traumatic event be associated with structural alteration of premotor areas in PTSD? METHODOLOGY/PRINCIPAL FINDINGS: Structural magnetic resonance imaging scans were acquired from a sample of individuals that had been exposed to urban violence. This sample consisted of 16 PTSD patients and 16 age- and gender-matched controls. Psychometric questionnaires differentiated PTSD patients from trauma-exposed controls with regard to PTSD symptoms, affective, and resilience predispositions. Voxel-based morphometric analysis revealed that, compared with controls, the PTSD patients presented significant reductions in gray matter volume in the ventral premotor cortex and in the pregenual anterior cingulate cortex. CONCLUSIONS: Volume reduction in the premotor cortex that is observed in victims of urban violence with PTSD may be associated with a disruption in the dynamical modulation of the safe space around the body. The finding that PTSD patients presented a smaller volume of pregenual anterior cingulate cortex is consistent with the results of other PTSD neuroimaging studies that investigated different types of traumatic events.

Does feature selection improve classification accuracy? Impact of sample size and feature selection on classification using anatomical magnetic resonance images.

There are growing numbers of studies using machine learning approaches to characterize patterns of anatomical difference discernible from neuroimaging data. The high-dimensionality of image data often raises a concern that feature selection is needed to obtain optimal accuracy. Among previous studies, mostly using fixed sample sizes, some show greater predictive accuracies with feature selection, whereas others do not. In this study, we compared four common feature selection methods. 1) Pre-selected region of interests (ROIs) that are based on prior knowledge. 2) Univariate t-test filtering. 3) Recursive feature elimination (RFE), and 4) t-test filtering constrained by ROIs. The predictive accuracies achieved from different sample sizes, with and without feature selection, were compared statistically. To demonstrate the effect, we used grey matter segmented from the T1-weighted anatomical scans collected by the Alzheimer's disease Neuroimaging Initiative (ADNI) as the input features to a linear support vector machine classifier. The objective was to characterize the patterns of difference between Alzheimer's disease (AD) patients and cognitively normal subjects, and also to characterize the difference between mild cognitive impairment (MCI) patients and normal subjects. In addition, we also compared the classification accuracies between MCI patients who converted to AD and MCI patients who did not convert within the period of 12 months. Predictive accuracies from two data-driven feature selection methods (t-test filtering and RFE) were no better than those achieved using whole brain data. We showed that we could achieve the most accurate characterizations by using prior knowledge of where to expect neurodegeneration (hippocampus and parahippocampal gyrus). Therefore, feature selection does improve the classification accuracies, but it depends on the method adopted. In general, larger sample sizes yielded higher accuracies with less advantage obtained by using knowledge from the existing literature.

Utilizing temporal information in fMRI decoding: classifier using kernel regression methods.

This paper describes a general kernel regression approach to predict experimental conditions from activity patterns acquired with functional magnetic resonance image (fMRI). The standard approach is to use classifiers that predict conditions from activity patterns. Our approach involves training different regression machines for each experimental condition, so that a predicted temporal profile is computed for each condition. A decision function is then used to classify the responses from the testing volumes into the corresponding category, by comparing the predicted temporal profile elicited by each event, against a canonical hemodynamic response function. This approach utilizes the temporal information in the fMRI signal and maintains more training samples in order to improve the classification accuracy over an existing strategy. This paper also introduces efficient techniques of temporal compaction, which operate directly on kernel matrices for kernel classification algorithms such as the support vector machine (SVM). Temporal compacting can convert the kernel computed from each fMRI volume directly into the kernel computed from beta-maps, average of volumes or spatial-temporal kernel. The proposed method was applied to three different datasets. The first one is a block-design experiment with three conditions of image stimuli. The method outperformed the SVM classifiers of three different types of temporal compaction in single-subject leave-one-block-out cross-validation. Our method achieved 100% classification accuracy for six of the subjects and an average of 94% accuracy across all 16 subjects, exceeding the best SVM classification result, which was 83% accuracy (p=0.008). The second dataset is also a block-design experiment with two conditions of visual attention (left or right). Our method yielded 96% accuracy and SVM yielded 92% (p=0.005). The third dataset is from a fast event-related experiment with two categories of visual objects. Our method achieved 77% accuracy, compared with 72% using SVM (p=0.0006).

Kernel regression for fMRI pattern prediction.

This paper introduces two kernel-based regression schemes to decode or predict brain states from functional brain scans as part of the Pittsburgh Brain Activity Interpretation Competition (PBAIC) 2007, in which our team was awarded first place. Our procedure involved image realignment, spatial smoothing, detrending of low-frequency drifts, and application of multivariate linear and non-linear kernel regression methods: namely kernel ridge regression (KRR) and relevance vector regression (RVR). RVR is based on a Bayesian framework, which automatically determines a sparse solution through maximization of marginal likelihood. KRR is the dual-form formulation of ridge regression, which solves regression problems with high dimensional data in a computationally efficient way. Feature selection based on prior knowledge about human brain function was also used. Post-processing by constrained deconvolution and re-convolution was used to furnish the prediction. This paper also contains a detailed description of how prior knowledge was used to fine tune predictions of specific "feature ratings," which we believe is one of the key factors in our prediction accuracy. The impact of pre-processing was also evaluated, demonstrating that different pre-processing may lead to significantly different accuracies. Although the original work was aimed at the PBAIC, many techniques described in this paper can be generally applied to any fMRI decoding works to increase the prediction accuracy.

Predicting clinical scores from magnetic resonance scans in Alzheimer's disease.

Machine learning and pattern recognition methods have been used to diagnose Alzheimer's disease (AD) and mild cognitive impairment (MCI) from individual MRI scans. Another application of such methods is to predict clinical scores from individual scans. Using relevance vector regression (RVR), we predicted individuals' performances on established tests from their MRI T1 weighted image in two independent data sets. From Mayo Clinic, 73 probable AD patients and 91 cognitively normal (CN) controls completed the Mini-Mental State Examination (MMSE), Dementia Rating Scale (DRS), and Auditory Verbal Learning Test (AVLT) within 3months of their scan. Baseline MRI's from the Alzheimer's disease Neuroimaging Initiative (ADNI) comprised the other data set; 113 AD, 351 MCI, and 122 CN subjects completed the MMSE and Alzheimer's Disease Assessment Scale-Cognitive subtest (ADAS-cog) and 39 AD, 92 MCI, and 32 CN ADNI subjects completed MMSE, ADAS-cog, and AVLT. Predicted and actual clinical scores were highly correlated for the MMSE, DRS, and ADAS-cog tests (P<0.0001). Training with one data set and testing with another demonstrated stability between data sets. DRS, MMSE, and ADAS-Cog correlated better than AVLT with whole brain grey matter changes associated with AD. This result underscores their utility for screening and tracking disease. RVR offers a novel way to measure interactions between structural changes and neuropsychological tests beyond that of univariate methods. In clinical practice, we envision using RVR to aid in diagnosis and predict clinical outcome.

Quantitative prediction of subjective pain intensity from whole-brain fMRI data using Gaussian processes.

Supervised machine learning (ML) algorithms are increasingly popular tools for fMRI decoding due to their predictive capability and their ability to capture information encoded by spatially correlated voxels. In addition, an important secondary outcome is a multivariate representation of the pattern underlying the prediction. Despite an impressive array of applications, most fMRI applications are framed as classification problems and predictions are limited to categorical class decisions. For many applications, quantitative predictions are desirable that more accurately represent variability within subject groups and that can be correlated with behavioural variables. We evaluate the predictive capability of Gaussian process (GP) models for two types of quantitative prediction (multivariate regression and probabilistic classification) using whole-brain fMRI volumes. As a proof of concept, we apply GP models to an fMRI experiment investigating subjective responses to thermal pain and show GP models predict subjective pain ratings without requiring anatomical hypotheses about functional localisation of relevant brain processes. Even in the case of pain perception, where strong hypotheses do exist, GP predictions were more accurate than any region previously demonstrated to encode pain intensity. We demonstrate two brain mapping methods suitable for GP models and we show that GP regression models outperform state of the art support vector- and relevance vector regression. For classification, GP models perform categorical prediction as accurately as a support vector machine classifier and furnish probabilistic class predictions.

Prognostic and diagnostic potential of the structural neuroanatomy of depression.

BACKGROUND: Depression is experienced as a persistent low mood or anhedonia accompanied by behavioural and cognitive disturbances which impair day to day functioning. However, the diagnosis is largely based on self-reported symptoms, and there are no neurobiological markers to guide the choice of treatment. In the present study, we examined the prognostic and diagnostic potential of the structural neural correlates of depression. METHODOLOGY AND PRINCIPAL FINDINGS: Subjects were 37 patients with major depressive disorder (mean age 43.2 years), medication-free, in an acute depressive episode, and 37 healthy individuals. Following the MRI scan, 30 patients underwent treatment with the antidepressant medication fluoxetine or cognitive behavioural therapy (CBT). Of the patients who subsequently achieved clinical remission with antidepressant medication, the whole brain structural neuroanatomy predicted 88.9% of the clinical response, prior to the initiation of treatment (88.9% patients in clinical remission (sensitivity) and 88.9% patients with residual symptoms (specificity), p = 0.01). Accuracy of the structural neuroanatomy as a diagnostic marker though was 67.6% (64.9% patients (sensitivity) and 70.3% healthy individuals (specificity), p = 0.027). CONCLUSIONS AND SIGNIFICANCE: The structural neuroanatomy of depression shows high predictive potential for clinical response to antidepressant medication, while its diagnostic potential is more limited. The present findings provide initial steps towards the development of neurobiological prognostic markers for depression.

Decoding neuronal ensembles in the human hippocampus.

BACKGROUND: The hippocampus underpins our ability to navigate, to form and recollect memories, and to imagine future experiences. How activity across millions of hippocampal neurons supports these functions is a fundamental question in neuroscience, wherein the size, sparseness, and organization of the hippocampal neural code are debated. RESULTS: Here, by using multivariate pattern classification and high spatial resolution functional MRI, we decoded activity across the population of neurons in the human medial temporal lobe while participants navigated in a virtual reality environment. Remarkably, we could accurately predict the position of an individual within this environment solely from the pattern of activity in his hippocampus even when visual input and task were held constant. Moreover, we observed a dissociation between responses in the hippocampus and parahippocampal gyrus, suggesting that they play differing roles in navigation. CONCLUSIONS: These results show that highly abstracted representations of space are expressed in the human hippocampus. Furthermore, our findings have implications for understanding the hippocampal population code and suggest that, contrary to current consensus, neuronal ensembles representing place memories must be large and have an anisotropic structure.

Accuracy of dementia diagnosis: a direct comparison between radiologists and a computerized method.

There has been recent interest in the application of machine learning techniques to neuroimaging-based diagnosis. These methods promise fully automated, standard PC-based clinical decisions, unbiased by variable radiological expertise. We recently used support vector machines (SVMs) to separate sporadic Alzheimer's disease from normal ageing and from fronto-temporal lobar degeneration (FTLD). In this study, we compare the results to those obtained by radiologists. A binary diagnostic classification was made by six radiologists with different levels of experience on the same scans and information that had been previously analysed with SVM. SVMs correctly classified 95% (sensitivity/specificity: 95/95) of sporadic Alzheimer's disease and controls into their respective groups. Radiologists correctly classified 65-95% (median 89%; sensitivity/specificity: 88/90) of scans. SVM correctly classified another set of sporadic Alzheimer's disease in 93% (sensitivity/specificity: 100/86) of cases, whereas radiologists ranged between 80% and 90% (median 83%; sensitivity/specificity: 80/85). SVMs were better at separating patients with sporadic Alzheimer's disease from those with FTLD (SVM 89%; sensitivity/specificity: 83/95; compared to radiological range from 63% to 83%; median 71%; sensitivity/specificity: 64/76). Radiologists were always accurate when they reported a high degree of diagnostic confidence. The results show that well-trained neuroradiologists classify typical Alzheimer's disease-associated scans comparable to SVMs. However, SVMs require no expert knowledge and trained SVMs can readily be exchanged between centres for use in diagnostic classification. These results are encouraging and indicate a role for computerized diagnostic methods in clinical practice.

Automatic classification of MR scans in Alzheimer's disease.

To be diagnostically useful, structural MRI must reliably distinguish Alzheimer's disease (AD) from normal aging in individual scans. Recent advances in statistical learning theory have led to the application of support vector machines to MRI for detection of a variety of disease states. The aims of this study were to assess how successfully support vector machines assigned individual diagnoses and to determine whether data-sets combined from multiple scanners and different centres could be used to obtain effective classification of scans. We used linear support vector machines to classify the grey matter segment of T1-weighted MR scans from pathologically proven AD patients and cognitively normal elderly individuals obtained from two centres with different scanning equipment. Because the clinical diagnosis of mild AD is difficult we also tested the ability of support vector machines to differentiate control scans from patients without post-mortem confirmation. Finally we sought to use these methods to differentiate scans between patients suffering from AD from those with frontotemporal lobar degeneration. Up to 96% of pathologically verified AD patients were correctly classified using whole brain images. Data from different centres were successfully combined achieving comparable results from the separate analyses. Importantly, data from one centre could be used to train a support vector machine to accurately differentiate AD and normal ageing scans obtained from another centre with different subjects and different scanner equipment. Patients with mild, clinically probable AD and age/sex matched controls were correctly separated in 89% of cases which is compatible with published diagnosis rates in the best clinical centres. This method correctly assigned 89% of patients with post-mortem confirmed diagnosis of either AD or frontotemporal lobar degeneration to their respective group. Our study leads to three conclusions: Firstly, support vector machines successfully separate patients with AD from healthy aging subjects. Secondly, they perform well in the differential diagnosis of two different forms of dementia. Thirdly, the method is robust and can be generalized across different centres. This suggests an important role for computer based diagnostic image analysis for clinical practice.