Multi-modal sequence learning for Alzheimer’s disease progression prediction with incomplete variable-length longitudinal data

Xu, Lei; Wang, Hui; He, Chunming; Jun, Wang; Zhang, Changqing; Nie, Feiping; Chen, Lei

doi:10.1016/j.media.2022.102643

Cited by 21 publications

(3 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(1) Survival analysis could be explored by treating RP as a time-to-event variable, implementing a dynamic prediction model that can be updated over time [ 39 ], and evaluating the models by time-dependent AUROC. (2) More imputation methods could be explored on our baseline and longitudinal data, such as the data-driven missing value imputation approach [ 40 ], HyperImpute [ 41 ], or integrating the data imputation module into the prediction models [ 16 ]. (3) Investigations could be conducted by using adapted supervised machine learning algorithms [ 42 , 43 ] and comparing them to our neural network models in terms of performances and efficiency for longitudinal classification.…”

Section: Discussionmentioning

confidence: 99%

“…Exclusions were based on Modified Hachinski Ischemia Scale scores > 4 or Geriatric Depression Scale scores > 6. EXPEDI-TION and EXPEDITION 2 involved mild-to-moderate AD patients (MMSE scores [16][17][18][19][20][21][22][23][24][25][26] with PET scans and lumbar punctures at various time points. CSF collection and amyloid PET were conducted in subsets of the participants in EXPEDITION and EXPEDITION 2, and neither was required for study eligibility.…”

Section: Data Sourcementioning

confidence: 99%

“…Potential prognostic factors affecting the rate of progression include fluid biomarkers (e.g., amyloid-beta levels [ 8 , 9 ]), neurodegeneration (e.g., brain atrophy [ 10 , 11 ]), and cognitive impairment rate [ 12 – 14 ]. Various data-driven predictive models using different data sources (e.g., ADNI [ 8 , 9 , 11 , 13 , 15 ], CODR-AD [ 14 ], and independent cohorts of investigators [ 10 , 12 ]) and methodologies (e.g., clustering [ 11 , 13 ], dimensionality reduction [ 10 ], conditional restricted Boltzmann machine [ 14 ], gradient boosting decision tree [ 8 ], and deep learning model [ 15 , 16 ]) have achieved limited success in predicting AD patients’ future cognitive scores, as seen in The Alzheimer’s Disease Prediction Of Longitudinal Evolution (TADPOLE) [ 17 ] challenges.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deep learning to predict rapid progression of Alzheimer’s disease from pooled clinical trials: A retrospective study

Ma,

Shyer,

Harris

et al. 2024

PLOS Digit Health

View full text Add to dashboard Cite

The rate of progression of Alzheimer’s disease (AD) differs dramatically between patients. Identifying the most is critical because when their numbers differ between treated and control groups, it distorts the outcome, making it impossible to tell whether the treatment was beneficial. Much recent effort, then, has gone into identifying RPs. We pooled de-identified placebo-arm data of three randomized controlled trials (RCTs), EXPEDITION, EXPEDITION 2, and EXPEDITION 3, provided by Eli Lilly and Company. After processing, the data included 1603 mild-to-moderate AD patients with 80 weeks of longitudinal observations on neurocognitive health, brain volumes, and amyloid-beta (Aβ) levels. RPs were defined by changes in four neurocognitive/functional health measures. We built deep learning models using recurrent neural networks with attention mechanisms to predict RPs by week 80 based on varying observation periods from baseline (e.g., 12, 28 weeks). Feature importance scores for RP prediction were computed and temporal feature trajectories were compared between RPs and non-RPs. Our evaluation and analysis focused on models trained with 28 weeks of observation. The models achieved robust internal validation area under the receiver operating characteristic (AUROCs) ranging from 0.80 (95% CI 0.79–0.82) to 0.82 (0.81–0.83), and the area under the precision-recall curve (AUPRCs) from 0.34 (0.32–0.36) to 0.46 (0.44–0.49). External validation AUROCs ranged from 0.75 (0.70–0.81) to 0.83 (0.82–0.84) and AUPRCs from 0.27 (0.25–0.29) to 0.45 (0.43–0.48). Aβ plasma levels, regional brain volumetry, and neurocognitive health emerged as important factors for the model prediction. In addition, the trajectories were stratified between predicted RPs and non-RPs based on factors such as ventricular volumes and neurocognitive domains. Our findings will greatly aid clinical trialists in designing tests for new medications, representing a key step toward identifying effective new AD therapies.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Data Sourcementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning to predict rapid progression of Alzheimer’s disease from pooled clinical trials: A retrospective study

Ma,

Shyer,

Harris

et al. 2024

PLOS Digit Health

View full text Add to dashboard Cite

show abstract

A multi‐view learning approach with diffusion model to synthesize FDG PET from MRI T1WI for diagnosis of Alzheimer's disease

Chen,

Weng,

Huang

et al. 2024

Alzheimer's & Dementia

View full text Add to dashboard Cite

INTRODUCTIONThis study presents a novel multi‐view learning approach for machine learning (ML)–based Alzheimer's disease (AD) diagnosis.METHODSA diffusion model is proposed to synthesize the fluorodeoxyglucose positron emission tomography (FDG PET) view from the magnetic resonance imaging T1 weighted imaging (MRI T1WI) view and incorporate two synthesis strategies: one‐way synthesis and two‐way synthesis. To assess the utility of the synthesized views, we use multilayer perceptron (MLP)–based classifiers with various combinations of the views.RESULTSThe two‐way synthesis achieves state‐of‐the‐art performance with a structural similarity index measure (SSIM) at 0.9380 and a peak‐signal‐to‐noise ratio (PSNR) at 26.47. The one‐way synthesis achieves an SSIM at 0.9282 and a PSNR at 23.83. Both synthesized FDG PET views have shown their effectiveness in improving diagnostic accuracy.DISCUSSIONThis work supports the notion that ML‐based cross‐domain data synthesis can be a useful approach to improve AD diagnosis by providing additional synthesized disease‐related views for multi‐view learning.Highlights We propose a diffusion model with two strategies to synthesize fluorodeoxyglucose positron emission tomography (FDG PET) from magnetic resonance imaging T1 weighted imaging (MRI T1WI). We raise multi‐view learning with MRl T1Wl and synthesized FDG PET for Alzheimer's disease (AD) diagnosis. We provide a comprehensive experimental comparison for the synthesized FDG PET view. The feasibility of synthesized FDG PET view in AD diagnosis is validated with various experiments. We demonstrate the ability of synthesized FDG PET to enhance the performance of machine learning–based AD diagnosis.

show abstract

Modeling Alzheimers’ Disease Progression from Multi-task and Self-supervised Learning Perspective with Brain Networks

Liang,

Zhang,

Cao

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Multi-modal sequence learning for Alzheimer’s disease progression prediction with incomplete variable-length longitudinal data

Cited by 21 publications

References 45 publications

Deep learning to predict rapid progression of Alzheimer’s disease from pooled clinical trials: A retrospective study

Deep learning to predict rapid progression of Alzheimer’s disease from pooled clinical trials: A retrospective study

A multi‐view learning approach with diffusion model to synthesize FDG PET from MRI T1WI for diagnosis of Alzheimer's disease

Modeling Alzheimers’ Disease Progression from Multi-task and Self-supervised Learning Perspective with Brain Networks

Contact Info

Product

Resources

About