Training recurrent neural networks robust to incomplete data: Application to Alzheimer’s disease progression modeling

Ghazi, Mostafa Mehdipour; Nielsen, Mads Eggert; Pai, Akshay; Cardoso, M. Jorge; Modat, Marc; Ourselin, Sébastien; Sørensen, Lauge; Initiative, Alzheimer’s Disease Neuroimaging

doi:10.1016/j.media.2019.01.004

Cited by 101 publications

(33 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Functional connectivity analysis found decreased connectivity in the superior temporal gyrus in dementia patients including AD (Hafkemeijer et al, 2015; Schwab et al, 2018). The middle temporal gyrus has been shown to atrophy significantly in both MCI and AD patients when compared to controls in longitudinal studies (Ghazi et al, 2019) and research that combined multi-modal data types (Convit et al, 2000; Korolev et al, 2016). Finally, Guo et al found significant gray matter volume reductions in the superior and middle temporal gyrus (2010).…”

Section: Discussionmentioning

confidence: 99%

Quantifying Neurodegenerative Progression With DeepSymNet, an End-to-End Data-Driven Approach

et al. 2019

View full text Add to dashboard Cite

Alzheimer's disease (AD) is the most common neurodegenerative disorder worldwide and is one of the leading sources of morbidity and mortality in the aging population. There is a long preclinical period followed by mild cognitive impairment (MCI). Clinical diagnosis and the rate of decline is variable. Progression monitoring remains a challenge in AD, and it is imperative to create better tools to quantify this progression. Brain magnetic resonance imaging (MRI) is commonly used for patient assessment. However, current approaches for analysis require strong a priori assumptions about regions of interest used and complex preprocessing pipelines including computationally expensive non-linear registrations and iterative surface deformations. These preprocessing steps are composed of many stacked processing layers. Any error or bias in an upstream layer will be propagated throughout the pipeline. Failures or biases in the non-linear subject registration and the subjective choice of atlases of specific regions are common in medical neuroimaging analysis and may hinder the translation of many approaches to the clinical practice. Here we propose a data-driven method based on an extension of a deep learning architecture, DeepSymNet, that identifies longitudinal changes without relying on prior brain regions of interest, an atlas, or non-linear registration steps. Our approach is trained end-to-end and learns how a patient's brain structure dynamically changes between two-time points directly from the raw voxels. We compare our approach with Freesurfer longitudinal pipelines and voxel-based methods using the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Our model can identify AD progression with comparable results to existing Freesurfer longitudinal pipelines without the need of predefined regions of interest, non-rigid registration algorithms, or iterative surface deformation at a fraction of the processing time. When compared to other voxel-based methods which share some of the same benefits, our model showed a statistically significant performance improvement. Additionally, we show that our model can differentiate between healthy subjects and patients with MCI. The model's decision was investigated using the epsilon layer-wise propagation algorithm. We found that the predictions were driven by the pallidum, putamen, and the superior temporal gyrus. Our novel longitudinal based, deep learning approach has the potential to diagnose patients earlier and enable new computational tools to monitor neurodegeneration in clinical practice.

show abstract

Section: Discussionmentioning

confidence: 99%

Quantifying Neurodegenerative Progression With DeepSymNet, an End-to-End Data-Driven Approach

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Here, we considered recurrent neural networks (RNNs), which allow an individual's latent state to be represented by a vector of numbers, thus providing a richer encoding of an individual's "disease state" beyond a single integer (as in the case of discrete state hidden Markov models). In the context of medical applications, RNNs have been used to model electronic health records (Lipton et al, 2016a;Choi et al, 2016;Esteban et al, 2016;Pham et al, 2017;Rajkomar et al, 2018;Suo et al, 2018) and AD disease progression (Nguyen et al, 2018;Ghazi et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Predicting Alzheimer’s disease progression using deep recurrent neural networks

Nguyen

et al. 2019

Preprint

View full text Add to dashboard Cite

Early identification of individuals at risk of developing Alzheimer's disease (AD) dementia is important for developing disease-modifying therapies. In this study, given multimodal AD markers and clinical diagnosis of an individual from one or more timepoints, we seek to predict the clinical diagnosis, cognition and ventricular volume of the individual for every month (indefinitely) into the future. We proposed a recurrent neural network (RNN) model and applied it to data from The Alzheimer's Disease Prediction Of Longitudinal Evolution (TADPOLE) challenge, comprising longitudinal data of 1677 participants (Marinescu et al. 2018) from the Alzheimer's Disease Neuroimaging Initiative (ADNI). We compared the performance of the RNN model and three baseline algorithms up to 6 years into the future.Most previous work on predicting AD progression ignore the issue of missing data, which is a prevalent issue in longitudinal data. Here, we explored three different strategies to handle missing data. Two of the strategies treated the missing data as a "preprocessing" issue, by imputing the missing data using the previous timepoint ("forward filling") or linear interpolation ("linear filling). The third strategy utilized the RNN model itself to fill in the missing data both during training and testing ("model filling"). Our analyses suggest that the RNN with "model filling" was better than baseline algorithms, including support vector machine/regression and linear state space (LSS) models. However, there was no statistical difference between the RNN and LSS for predicting cognition and ventricular volume.Importantly, although the training procedure utilized longitudinal data, we found that the trained RNN model exhibited similar performance, when using only 1 input timepoint or 4 input timepoints, suggesting that our approach might work well with just cross-sectional data.An earlier version of our approach was ranked 5th (out of 53 entries) in the TADPOLE challenge in 2019. The current approach is ranked 2nd out of 56 entries as of August 12th, 2019.

show abstract

“…Other models are also based on differential equation models [3]. Non parametric models using deep learning were also explored [16].…”

Section: Related Workmentioning

confidence: 99%

Mixture Modeling for Identifying Subtypes in Disease Course Mapping

Poulet

Durrleman

2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Disease modeling techniques summarize the possible trajectories of progression from multimodal and longitudinal data. These techniques often assume that individuals form a homogeneous cluster, thus ignoring possible disease subtypes within the population. We extend a non-linear mixed-effect model used for disease course mapping with a mixture framework. We jointly estimate model parameters and subtypes with a tempered version of a stochastic approximation of the Expectation Maximisation algorithm. We show that our model recovers the ground truth parameters from synthetic data, in contrast to the naive solution consisting in post hoc clustering of individual parameters from a one-class model. Applications to Alzheimer's disease data allows the unsupervised identification of disease subtypes associated with distinct relationship between cognitive decline and progression of imaging and biological biomarkers.

show abstract

Training recurrent neural networks robust to incomplete data: Application to Alzheimer’s disease progression modeling

Cited by 101 publications

References 25 publications

Quantifying Neurodegenerative Progression With DeepSymNet, an End-to-End Data-Driven Approach

Quantifying Neurodegenerative Progression With DeepSymNet, an End-to-End Data-Driven Approach

Predicting Alzheimer’s disease progression using deep recurrent neural networks

Mixture Modeling for Identifying Subtypes in Disease Course Mapping

Contact Info

Product

Resources

About