Search

Introduction and objective Several prediction models for falls/near falls in Parkinson’s disease (PD) have been proposed. However, longitudinal predictors of frequency of falls/near falls are poorly investigated. Therefore, we aimed to identify short- and long-term predictors of the number of falls/near falls in PD. Methods A prospective cohort of 58 persons with PD was assessed at baseline (mean age and PD duration, 65 and 3.2 years, respectively) and 3.5 years later. Potential predictors were history of falls and near falls, comfortable gait speed, freezing of gate, dyskinesia, retropulsion, tandem gait (TG), pain, and cognition (Mini-Mental State Exam, MMSE). After each assessment, the participants registered a number of falls/near falls during the following 6 months. Multivariate Poisson regression was used to identify short- and long-term predictors of a number of falls/near falls. Results Baseline median (q1–q3) motor (UPDRS) and MMSE scores were 10 (6.75–14) and 28.5 (27–29), respectively. History of falls was the only significant short-time predictor [incidence rate ratio (IRR), 15.17] for the number of falls/near falls during 6 months following baseline. Abnormal TG (IRR, 3.77) and lower MMSE scores (IRR, 1.17) were short-term predictors 3.5 years later. Abnormal TG (IRR, 7.79) and lower MMSE scores (IRR, 1.49) at baseline were long-term predictors of the number of falls/near falls 3.5 years later. Conclusion Abnormal TG and MMSE scores predict the number of falls/near falls in short and long term, and may be indicative of disease progression. Our observations provide important additions to the evidence base for clinical fall prediction in PD.

Missing data is a common problem in longitudinal studies due to subject dropouts and failed scans. We present a graph-based convolutional neural network to predict missing diffusion MRI data. In particular, we consider the relationships between sampling points in the spatial domain and the diffusion wave-vector domain to construct a graph. We then use a graph convolutional network to learn the non-linear mapping from available data to missing data. Our method harnesses a multi-scale residual architecture with adversarial learning for prediction with greater accuracy and perceptual quality. Experimental results show that our method is accurate and robust in the longitudinal prediction of infant brain diffusion MRI data.

Contains fulltext : 284388.pdf (Publisher’s version ) (Open Access) Motricity is the most commonly affected ability after a stroke. While many clinical studies attempt to predict motor symptoms at different chronic time points after a stroke, longitudinal acute-to-chronic studies remain scarce. Taking advantage of recent advances in mapping brain disconnections, we predict motor outcomes in 62 patients assessed longitudinally two weeks, three months, and one year after their stroke. Results indicate that brain disconnection patterns accurately predict motor impairments. However, disconnection patterns leading to impairment differ between the three-time points and between left and right motor impairments. These results were cross-validated using resampling techniques. In sum, we demonstrated that while some neuroplasticity mechanisms exist changing the structure-function relationship, disconnection patterns prevail when predicting motor impairment at different time points after stroke. 14 p.

The infant brain undergoes a remarkable period of neural development that is crucial for the development of cognitive and behavioral capacities (Hasegawa et al., 2018). Longitudinal magnetic resonance imaging (MRI) is able to characterize the developmental trajectories and is critical in neuroimaging studies of early brain development. However, missing data at different time points is an unavoidable occurrence in longitudinal studies owing to participant attrition and scan failure. Compared to dropping incomplete data, data imputation is considered a better solution to address such missing data in order to preserve all available samples. In this paper, we adapt generative adversarial networks (GAN) to a new application: longitudinal image prediction of structural MRI in the first year of life. In contrast to existing medical image-to-image translation applications of GANs, where inputs and outputs share a very close anatomical structure, our task is more challenging as brain size, shape and tissue contrast vary significantly between the input data and the predicted data. Several improvements over existing GAN approaches are proposed to address these challenges in our task. To enhance the realism, crispness, and accuracy of the predicted images, we incorporate both a traditional voxel-wise reconstruction loss as well as a perceptual loss term into the adversarial learning scheme. As the differing contrast changes in T1w and T2w MR images in the first year of life, we incorporate multi-contrast images leading to our proposed 3D multi-contrast perceptual adversarial network (MPGAN). Extensive evaluations are performed to assess the qualityand fidelity of the predicted images, including qualitative and quantitative assessments of the image appearance, as well as quantitative assessment on two segmentation tasks. Our experimental results show that our MPGAN is an effective solution for longitudinal MR image data imputation in the infant brain. We further apply our predicted/imputed images to two practical tasks, a regression task and a classification task, in order to highlight the enhanced task-related performance following image imputation. The results show that the model performance in both tasks is improved by including the additional imputed data, demonstrating the usability of the predicted images generated from our approach.

We examined the longitudinal predictors of electrophysiological and behavioral markers of inhibitory control in adolescence. Participants were 63 adolescent boys who have been followed since birth as part of a prospective longitudinal study on the developmental pathways to attention-deficit hyperactivity disorder (ADHD). At 17 years of age, they completed the stop-signal task (SST) while electroencephalography (EEG) was continuously recorded. Inhibitory control was evaluated by the stop-signal reaction time (SSRT) as well as by the amplitude of the event-related potential (ERP) component of N2 during successful inhibition. We found that higher inattention symptoms throughout childhood predicted reduced amplitude (i.e., less negative) of the N2 in adolescence. Furthermore, the N2 amplitude was longitudinally predicted by the early precursors of child familial risk for ADHD and early childhood temperament. Specifically, father’s inattention symptoms (measured in the child’s early infancy) and child’s effortful control at 36 months of age directly predicted the N2 amplitude in adolescence, even beyond the consistency of inattention symptoms throughout development. The SSRT was predicted by ADHD symptoms throughout childhood but not by the early precursors. Our findings emphasize the relevance of early familial and temperamental risk for ADHD to the prediction of a later dysfunction in inhibitory control.

The prognostic significance of serial echocardiography and its rate of change in children with dilated cardiomyopathy (DCM) is incompletely defined.We retrospectively analysed up to 4 serial echocardiograms. Associations between mortality/transplant and echocardiographic parameters over time and between outcomes and the rate of change of echocardiographic parameters were analysed. Estimation of patient-specific intercepts and slopes was done using linear regression models.Fifty-seven DCM children were studied (50% male; median age, 0.6 year; average follow-up, 2.1 ± 2.4 years). The median time to transplant or death was 2.0 years. Increased left ventricular (LV) diastolic (LVEDD) and systolic (LVESD) dimensions and myocardial performance index (MPI) were associated with increased mortality and transplant risk. Increased LV ejection fraction, mitral E-deceleration time, right ventricular (RV) fractional area change, and tricuspid annular plane systolic excursion were associated with reduced mortality and transplant risk. Transplant/mortality likelihood increased by 41.6% and 19.8% for each unit increase in LVEDD and LVESD z scores, respectively (LVEDD: hazard ratio [HR], 1.416; 95% confidence interval [CI], 1.285-1.560; P0.001; LVESD: HR, 1.198; 95% CI, 1.147-1.251; P0.001). A higher monthly change in LVESD z score increased transplant/mortality likelihood by 85.6% (HR, 1.856; 95% CI, 1.572-2.191; P = 0.015). Greater changes in mitral E/e' (HR, 0.707; 95% CI, 0.636-0.786; P0.001) and RV MPI (HR, 0.412; 95% CI, 0.277-0.613; P0.001) were associated with reduced mortality and transplant risk.LV and RV systolic and diastolic dimensions and function over time and their rate of change are associated with risk for transplant and mortality in childhood DCM. Serial changes in these parameters may be useful to predict clinical outcomes.

Longitudinal magnetic resonance imaging (MRI) is essential in neuroimaging studies of early brain development. However, incomplete data is an inevitable problem in longitudinal studies because of participant attrition and scan failure. Data imputation is a possible way to address such missing data. Here, we propose a novel 3D multi-modal perceptual adversarial network (MPGAN) to predict a missing MR image from an existing longitudinal image of the same subject. To the best of our knowledge, this is the first application of deep generative methods for longitudinal image prediction of structural MRI in the first year of life, where brain volume and image intensities are changing dramatically. In order to produce sharper and more realistic images, we incorporate the perceptual loss into the adversarial training process. To leverage complementary information contained in the multi-modality data, MPGAN predicts T1w and T2w images jointly in the prediction process. We evaluated MPGAN versus six alternative approaches based on visual as well as quantitative assessment. The results indicate that our MPGAN predicts missing MR images in an accurate and visually realistic fashion, and shows better performance than the alternative methods.

Death in amyotrophic lateral sclerosis (ALS) patients is related to respiratory failure, which is assessed in clinical settings by measuring vital capacity. We developed ALS-VC, a modeling tool for longitudinal prediction of vital capacity in ALS patients.A gradient boosting machine (GBM) model was trained using the PRO-ACT (Pooled Resource Open-access ALS Clinical Trials) database of over 10,000 ALS patient records. We hypothesized that a reliable vital capacity predictive model could be developed using PRO-ACT.The model was used to compare FVC predictions with a 30-day run-in period to predictions made from just baseline. The internal root mean square deviations (RMSD) of the run-in and baseline models were 0.534 and 0.539, respectively, across the 7L FVC range captured in PRO-ACT. The RMSDs of the run-in and baseline models using an unrelated, contemporary external validation dataset (0.553 and 0.538, respectively) were comparable to the internal validation. The model was shown to have similar accuracy for predicting SVC (RMSD = 0.562). The most important features for both run-in and baseline models were "Baseline forced vital capacity" and "Days since baseline."We developed ALS-VC, a GBM model trained with the PRO-ACT ALS dataset that provides vital capacity predictions generalizable to external datasets. The ALS-VC model could be helpful in advising and counseling patients, and, in clinical trials, it could be used to generate virtual control arms against which observed outcomes could be compared, or used to stratify patients into slowly, average, and rapidly progressing subgroups.

In establishing prognostic models, often aided by machine learning methods, much effort is concentrated in identifying good predictors. However, the same level of rigor is often absent in improving the outcome side of the models. In this study, we focus on this rather neglected aspect of model development. We are particularly interested in the use of longitudinal information as a way of improving the outcome side of prognostic models. This involves optimally characterizing individuals’ outcome status, classifying them, and validating the formulated prediction targets. None of these tasks are straightforward, which may explain why longitudinal prediction targets are not commonly used in practice despite their compelling benefits. As a way of improving this situation, we explore the joint use of empirical model fitting, clinical insights, and cross-validation based on how well formulated targets are predicted by clinically relevant baseline characteristics (antecedent validators). The idea here is that all these methods are imperfect but can be used together to triangulate valid prediction targets. The proposed approach is illustrated using data from the longitudinal assessment of manic symptoms study.

It is highly desirable to predict the progression of Alzheimer's disease (AD) of patients [e.g., to predict conversion of mild cognitive impairment (MCI) to AD], especially longitudinal prediction of AD is important for its early diagnosis. Currently, most existing methods predict different clinical scores using different models, or separately predict multiple scores at different future time points. Such approaches prevent coordinated learning of multiple predictions that can be used to jointly predict clinical scores at multiple future time points. In this paper, we propose a joint learning method for predicting clinical scores of patients using multiple longitudinal prediction models for various future time points. Three important relationships among training samples, features, and clinical scores are explored. The relationship among different longitudinal prediction models is captured using a common feature set among the multiple prediction models at different time points. Our experimental results based on the Alzheimer's disease neuroimaging initiative (ADNI) database shows that our method achieves considerable improvement over competing methods in predicting multiple clinical scores.

Dramatic changes of the human brain during the first year of postnatal development are poorly understood due to their multifold complexity. In this paper, we present the first attempt to jointly predict, using neonatal data, the dynamic growth pattern of brain cortical surfaces (collection of 3D triangular faces) and fiber tracts (collection of 3D lines). These two entities are modeled jointly as a multishape (a set of interlinked shapes). We propose a hybrid learning-based multishape prediction framework that captures both the diffeomorphic evolution of the cortical surfaces and the non-diffeomorphic growth of fiber tracts. In particular, we learn a set of geometric and dynamic cortical features and fiber connectivity features that characterize the relationships between cortical surfaces and fibers at different timepoints (0, 3, 6, and 9 months of age). Given a new neonatal multishape at 0 month of age, we hierarchically predict, at 3, 6 and 9 months, the postnatal cortical surfaces vertex-by-vertex along with fibers connected to adjacent faces to these vertices. This is achieved using a new fiber-to-face metric that quantifies the similarity between multishapes. For validation, we propose several evaluation metrics to thoroughly assess the performance of our framework. The results confirm that our framework yields good prediction accuracy of complex neonatal multishape development within a few seconds.

Indirect calorimetry is the criterion method for assessment of energy expenditure in critically ill patients but is decidedly uncommon. Thus, calculation methods proliferate. Even if indirect calorimetry is available, it usually is not repeated more than weekly on the same patient, creating potential for error. The purpose of the current study was to quantify estimation errors against indirect calorimetry measurements in critically ill patients over time.In mechanically ventilated, critical care patients, indirect calorimetry was used to measure resting metabolic rate for 7 days. Three estimation methods were compared with the cumulative measurement: the Penn State equations, the American College of Chest Physicians (ACCP) standard (25 kcal/kg body weight), and an extrapolated value based on the first measurement multiplied by 7 days.The cumulative difference between measured resting metabolic rate and the rate predicted by the Penn State equations was -468 ± 642 kcal (-3.7% ± 5.1% of the measured value). The difference for the ACCP was smaller, but variation was much wider (-387 ± 1597 kcal or -2.2% ± 11.9% of the measured value). The extrapolated value was -684 ± 1731 kcal (-4.1% ± 11.4% of measured expenditure).On average, the Penn State equations predict resting metabolic rate over time within 5% of the measured value. This performance is similar to the practice of making 1 measurement and extrapolating it over 1 week. The ACCP method has an unacceptably wide limit of agreement.

Purpose: To do a follow-up sonography assessment of the thymic size in infants at an age of 24 months, and to create a longitudinal prediction model for the thymic index covering all ages from birth to 24 months. Material and Methods: of 37 infants examined in an earlier investigation, 34 attended a 24-month follow-up examination. The thymic index, a volume estimate, was assessed by sonography and compared to clinical variables, breastfeeding status and illness. The longitudinal prediction model was based on data throughout 2 years. Results: There was no significant relation between the thymic index and the clinical variables, breast-feeding status or illness at 24 months. An overall test for the effect of breast-feeding status at 4 months for infants from 0-24 months was significant, as was the actual body length of the infants from 0-8 months. Prediction models were estimated. Conclusion: Based on a 24-month longitudinal sonography study, prediction models are presented whereby the thymic size, as an index, can be predicted at all times from birth to 24 months of age.