Unsupervised multimodal modeling of cognitive and brain health trajectories for early dementia prediction.

Predicting the course of neurodegenerative disorders early has potential to greatly improve clinical management and patient outcomes. A key challenge for early prediction in real-world clinical settings is the lack of labeled data (i.e., clinical diagnosis). In contrast to supervised classification approaches that require labeled data, we propose an unsupervised multimodal trajectory modeling (MTM) approach based on a mixture of state space models that captures changes in longitudinal data (i.e., trajectories) and stratifies individuals without using clinical diagnosis for model training. MTM learns the relationship between states comprising expensive, invasive biomarkers (ß-amyloid, grey matter density) and readily obtainable cognitive observations. MTM training on trajectories stratifies individuals into clinically meaningful clusters more reliably than MTM training on baseline data alone and is robust to missing data (i.e., cognitive data alone or single assessments). Extracting an individualized cognitive health index (i.e., MTM-derived cluster membership index) allows us to predict progression to AD more precisely than standard clinical assessments (i.e., cognitive tests or MRI scans alone). Importantly, MTM generalizes successfully from research cohort to real-world clinical data from memory clinic patients with missing data, enhancing the clinical utility of our approach. Thus, our multimodal trajectory modeling approach provides a cost-effective and non-invasive tool for early dementia prediction without labeled data (i.e., clinical diagnosis) with strong potential for translation to clinical practice.

Artificial intelligence for diagnostic and prognostic neuroimaging in dementia: A systematic review.

INTRODUCTION: Artificial intelligence (AI) and neuroimaging offer new opportunities for diagnosis and prognosis of dementia. METHODS: We systematically reviewed studies reporting AI for neuroimaging in diagnosis and/or prognosis of cognitive neurodegenerative diseases. RESULTS: A total of 255 studies were identified. Most studies relied on the Alzheimer's Disease Neuroimaging Initiative dataset. Algorithmic classifiers were the most commonly used AI method (48%) and discriminative models performed best for differentiating Alzheimer's disease from controls. The accuracy of algorithms varied with the patient cohort, imaging modalities, and stratifiers used. Few studies performed validation in an independent cohort. DISCUSSION: The literature has several methodological limitations including lack of sufficient algorithm development descriptions and standard definitions. We make recommendations to improve model validation including addressing key clinical questions, providing sufficient description of AI methods and validating findings in independent datasets. Collaborative approaches between experts in AI and medicine will help achieve the promising potential of AI tools in practice. HIGHLIGHTS: There has been a rapid expansion in the use of machine learning for diagnosis and prognosis in neurodegenerative disease Most studies (71%) relied on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset with no other individual dataset used more than five times There has been a recent rise in the use of more complex discriminative models (e.g., neural networks) that performed better than other classifiers for classification of AD vs healthy controls We make recommendations to address methodological considerations, addressing key clinical questions, and validation We also make recommendations for the field more broadly to standardize outcome measures, address gaps in the literature, and monitor sources of bias.

Predicting incident dementia in cerebral small vessel disease: comparison of machine learning and traditional statistical models.

Background: Cerebral small vessel disease (SVD) contributes to 45% of dementia cases worldwide, yet we lack a reliable model for predicting dementia in SVD. Past attempts largely relied on traditional statistical approaches. Here, we investigated whether machine learning (ML) methods improved prediction of incident dementia in SVD from baseline SVD-related features over traditional statistical methods. Methods: We included three cohorts with varying SVD severity (RUN DMC, n = 503; SCANS, n = 121; HARMONISATION, n = 265). Baseline demographics, vascular risk factors, cognitive scores, and magnetic resonance imaging (MRI) features of SVD were used for prediction. We conducted both survival analysis and classification analysis predicting 3-year dementia risk. For each analysis, several ML methods were evaluated against standard Cox or logistic regression. Finally, we compared the feature importance ranked by different models. Results: We included 789 participants without missing data in the survival analysis, amongst whom 108 (13.7%) developed dementia during a median follow-up of 5.4 years. Excluding those censored before three years, we included 750 participants in the classification analysis, amongst whom 48 (6.4%) developed dementia by year 3. Comparing statistical and ML models, only regularised Cox/logistic regression outperformed their statistical counterparts overall, but not significantly so in survival analysis. Baseline cognition was highly predictive, and global cognition was the most important feature. Conclusions: When using baseline SVD-related features to predict dementia in SVD, the ML survival or classification models we evaluated brought little improvement over traditional statistical approaches. The benefits of ML should be evaluated with caution, especially given limited sample size and features.

The Discriminative Kalman Filter for Bayesian Filtering with Nonlinear and Nongaussian Observation Models.

The Kalman filter provides a simple and efficient algorithm to compute the posterior distribution for state-space models where both the latent state and measurement models are linear and gaussian. Extensions to the Kalman filter, including the extended and unscented Kalman filters, incorporate linearizations for models where the observation model p(observation|state) is nonlinear. We argue that in many cases, a model for p(state|observation) proves both easier to learn and more accurate for latent state estimation. Approximating p(state|observation) as gaussian leads to a new filtering algorithm, the discriminative Kalman filter (DKF), which can perform well even when p(observation|state) is highly nonlinear and/or nongaussian. The approximation, motivated by the Bernstein-von Mises theorem, improves as the dimensionality of the observations increases. The DKF has computational complexity similar to the Kalman filter, allowing it in some cases to perform much faster than particle filters with similar precision, while better accounting for nonlinear and nongaussian observation models than Kalman-based extensions. When the observation model must be learned from training data prior to filtering, off-the-shelf nonlinear and nonparametric regression techniques can provide a gaussian model for p(observation|state) that cleanly integrates with the DKF. As part of the BrainGate2 clinical trial, we successfully implemented gaussian process regression with the DKF framework in a brain-computer interface to provide real-time, closed-loop cursor control to a person with a complete spinal cord injury. In this letter, we explore the theory underlying the DKF, exhibit some illustrative examples, and outline potential extensions.

Robust Closed-Loop Control of a Cursor in a Person with Tetraplegia using Gaussian Process Regression.

Intracortical brain computer interfaces can enable individuals with paralysis to control external devices through voluntarily modulated brain activity. Decoding quality has been previously shown to degrade with signal nonstationarities-specifically, the changes in the statistics of the data between training and testing data sets. This includes changes to the neural tuning profiles and baseline shifts in firing rates of recorded neurons, as well as nonphysiological noise. While progress has been made toward providing long-term user control via decoder recalibration, relatively little work has been dedicated to making the decoding algorithm more resilient to signal nonstationarities. Here, we describe how principled kernel selection with gaussian process regression can be used within a Bayesian filtering framework to mitigate the effects of commonly encountered nonstationarities. Given a supervised training set of (neural features, intention to move in a direction)-pairs, we use gaussian process regression to predict the intention given the neural data. We apply kernel embedding for each neural feature with the standard radial basis function. The multiple kernels are then summed together across each neural dimension, which allows the kernel to effectively ignore large differences that occur only in a single feature. The summed kernel is used for real-time predictions of the posterior mean and variance under a gaussian process framework. The predictions are then filtered using the discriminative Kalman filter to produce an estimate of the neural intention given the history of neural data. We refer to the multiple kernel approach combined with the discriminative Kalman filter as the MK-DKF. We found that the MK-DKF decoder was more resilient to nonstationarities frequently encountered in-real world settings yet provided similar performance to the currently used Kalman decoder. These results demonstrate a method by which neural decoding can be made more resistant to nonstationarities.

Rapid calibration of an intracortical brain-computer interface for people with tetraplegia.

OBJECTIVE: Brain-computer interfaces (BCIs) can enable individuals with tetraplegia to communicate and control external devices. Though much progress has been made in improving the speed and robustness of neural control provided by intracortical BCIs, little research has been devoted to minimizing the amount of time spent on decoder calibration. APPROACH: We investigated the amount of time users needed to calibrate decoders and achieve performance saturation using two markedly different decoding algorithms: the steady-state Kalman filter, and a novel technique using Gaussian process regression (GP-DKF). MAIN RESULTS: Three people with tetraplegia gained rapid closed-loop neural cursor control and peak, plateaued decoder performance within 3 min of initializing calibration. We also show that a BCI-naïve user (T5) was able to rapidly attain closed-loop neural cursor control with the GP-DKF using self-selected movement imagery on his first-ever day of closed-loop BCI use, acquiring a target 37 s after initiating calibration. SIGNIFICANCE: These results demonstrate the potential for an intracortical BCI to be used immediately after deployment by people with paralysis, without the need for user learning or extensive system calibration.

Involvement of the C-terminal disulfide-bonded loop of murine leukemia virus SU protein in a postbinding step critical for viral entry.

A role for the C-terminal domain (CTD) of murine leukemia virus (MuLV) Env protein in viral fusion was indicated by the potent inhibition of MuLV-induced fusion, but not receptor binding, by two rat monoclonal antibodies (MAbs) specific for epitopes in the CTD. Although these two MAbs, 35/56 and 83A25, have very different patterns of reactivity with viral isolates, determinants of both epitopes were mapped to the last C-terminal disulfide-bonded loop of SU (loop 10), and residues in this loop responsible for the different specificities of these MAbs were identified. Both MAbs reacted with a minor fraction of a truncated SU fragment terminating four residues after loop 10, indicating that while the deleted C-terminal residues were not part of these epitopes, they promoted their formation. Neither MAb recognized the loop 10 region expressed in isolated form, suggesting that these epitopes were not completely localized within loop 10 but required additional sequences located N terminal to the loop. Direct support for a role for loop 10 in fusion was provided by the demonstration that Env mutants containing an extra serine or threonine residue between the second and third positions of the loop were highly attenuated for infectivity and defective in fusion assays, despite wild-type levels of expression, processing, and receptor binding. Other mutations at positions 1 to 3 of loop 10 inhibited processing of the gPr80 precursor protein or led to increased shedding of SU, suggesting that loop 10 also affects Env folding and the stability of the interaction between SU and TM.

Distinct mechanisms of neutralization by monoclonal antibodies specific for sites in the N-terminal or C-terminal domain of murine leukemia virus SU.

The epitope specificities and functional activities of monoclonal antibodies (MAbs) specific for the murine leukemia virus (MuLV) SU envelope protein subunit were determined. Neutralizing antibodies were directed towards two distinct sites in MuLV SU: one overlapping the major receptor-binding pocket in the N-terminal domain and the other involving a region that includes the most C-terminal disulfide-bonded loop. Two other groups of MAbs, reactive with distinct sites in the N-terminal domain or in the proline-rich region (PRR), did not neutralize MuLV infectivity. Only the neutralizing MAbs specific for the receptor-binding pocket were able to block binding of purified SU and MuLV virions to cells expressing the ecotropic MuLV receptor, mCAT-1. Whereas the neutralizing MAbs specific for the C-terminal domain did not interfere with the SU-mCAT-1 interaction, they efficiently inhibited cell-to-cell fusion mediated by MuLV Env, indicating that they interfered with a postattachment event necessary for fusion. The C-terminal domain MAbs displayed the highest neutralization titers and binding activities. However, the nonneutralizing PRR-specific MAbs bound to intact virions with affinities similar to those of the neutralizing receptor-binding pocket-specific MAbs, indicating that epitope exposure, while necessary, is not sufficient for viral neutralization by MAbs. These results identify two separate neutralization domains in MuLV SU and suggest a role for the C-terminal domain in a postattachment step necessary for viral fusion.

Efficient isolation of novel human monoclonal antibodies with neutralizing activity against HIV-1 from transgenic mice expressing human Ig loci.

Despite considerable interest in the isolation of mAbs with potent neutralization activity against primary HIV-1 isolates, both for identifying useful targets for vaccine development and for the development of therapeutically useful reagents against HIV-1 infection, a relatively limited number of such reagents have been isolated to date. Human mAbs (hu-mAbs) are preferable to rodent mAbs for treatment of humans, but isolation of hu-mAbs from HIV-infected subjects by standard methods of EBV transformation of B cells or phage display of Ig libraries is inefficient and limited by the inability to control or define the original immunogen. An alternative approach for the isolation of hu-mAbs has been provided by the development of transgenic mice that produce fully hu-mAbs. In this report, we show that immunizing the XenoMouse G2 strain with native recombinant gp120 derived from HIV(SF162) resulted in robust humoral Ab responses against gp120 and allowed the efficient isolation of hybridomas producing specific hu-mAbs directed against multiple regions and epitopes of gp120. hu-mAbs possessing strong neutralizing activity against the autologous HIV(SF162) strain were obtained. The epitopes recognized were located in three previously described neutralization domains, the V2-, V3- and CD4-binding domains, and in a novel neutralization domain, the highly variable C-terminal region of the V1 loop. This is the first report of neutralizing mAbs directed at targets in the V1 region. Furthermore, the V2 and V3 epitopes recognized by neutralizing hu-mAbs were distinct from those of previously described human and rodent mAbs and included an epitope requiring a full length V3 loop peptide for effective presentation. These results further our understanding of neutralization targets for primary, R5 HIV-1 viruses and demonstrate the utility of the XenoMouse system for identifying new and interesting epitopes on HIV-1.