Using deep-learning to obtain calibrated individual disease and ADL damage transition probabilities between successive ELSA waves
Dil, E.; Rutenberg, A.
Show abstract
We predictively model damage transition probabilities for binary health outputs of 19 diseases and 25 activities of daily living states (ADLs) between successive waves of the English Longitudinal Study of Aging (ELSA). Model selection between deep neural networks (DNN), random forests, and logistic regression found that a simple one-hidden layer 128-node DNN was best able to predict future health states (AUC [≥] 0.91) and average damage probabilities (R2 [≥] 0.92). Feature selection from 134 explanatory variables found that 33 variables are sufficient to predict all disease and ADL states well. Deciles of predicted damage transition probabilities were well calibrated, but correlations between predicted health states were stronger than observed. The hazard ratios (HRs) between high-risk deciles and the average were between 3 and 10; high prevalence damage transitions typically had smaller HRs. Model predictions were good across all individual ages. A simple one-hidden layer DNN predicts multiple binary diseases and ADLs with well calibrated damage and repair transition probabilities.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- AI reveals insights into link between CD33 and cognitive impairment in Alzheimer's Disease 93%
- A Deep Survival EWAS approach estimating risk profile based on pre-diagnostic DNA methylation: an application to Breast Cancer time to diagnosis 93%
- Explainable deep transfer learning model for disease risk prediction using high-dimensional genomic data 93%
Similar papers in this journal
Similar papers in this journal
- Identification of a blood test-based biomarker of aging through deep learning of aging trajectories in large phenotypic datasets of mice 94%
- Deep representation learning for clustering longitudinal survival data from electronic health records 94%
- Using deep learning to predict age from liver and pancreas magnetic resonance images allows the identification of genetic and non-genetic factors associated with abdominal aging 93%
Similar papers in this journal
- Compressive Big Data Analytics: An Ensemble Meta-Algorithm for High-dimensional Multisource Datasets 93%
- A physically inspired approach to coarse-graining transcriptomes reveals the dynamics of aging: multiscale description of gene expressions and a spectral view of aging dynamics 92%
- Proteome-scale prediction of molecular mechanisms underlying dominant genetic diseases 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.