Research map / Probabilistic and scientific modeling / Causal discovery
Causal Hidden Markov Model for Time Series Disease Forecasting
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Separate disease-causative latent variables from other latent features (ablation) | Causal-HMM · PPA forecasting test (mean, 10 settings) | ACC: 70.94 ± 2.53 -> 71.92 ± 2.73 | Helped |
| Separate disease-causative latent variables from other latent features (ablation) | Causal-HMM · PPA forecasting test (mean, 10 settings) | AUC: 75.53 ± 1.82 -> 79.55 ± 1.62 | Helped |
| Add clinical measurements and personal attributes to sequential VAE (ablation) | Seq VAE · PPA forecasting test (mean, 10 settings) | ACC: 68.86 ± 2.02 -> 70.94 ± 2.53 | Helped |
| Add clinical measurements and personal attributes to sequential VAE (ablation) | Seq VAE · PPA forecasting test (mean, 10 settings) | AUC: 73.19 ± 1.82 -> 75.53 ± 1.82 | Helped |
| Use sequential VAE modeling instead of image-only CNN (ablation) | PPA forecasting · PPA forecasting test (mean, 10 settings) | ACC: 62.67 ± 2.89 -> 68.86 ± 2.02 | Helped |
| Use sequential VAE modeling instead of image-only CNN (ablation) | PPA forecasting · PPA forecasting test (mean, 10 settings) | AUC: 64.97 ± 1.81 -> 73.19 ± 1.82 | Helped |
| Classify using disease-related s+v instead of non-disease-related z (ablation) | Causal-HMM latent variables · PPA forecasting test (mean, 10 settings) | ACC: 62.82 ± 1.33 -> 71.50 ± 1.82 | Helped |
| Classify using disease-related s+v instead of non-disease-related z (ablation) | Causal-HMM latent variables · PPA forecasting test (mean, 10 settings) | AUC: 59.65 ± 0.99 -> 74.87 ± 0.66 | Helped |
| Use Causal-HMM rather than LogSparse Transformer | PPA forecasting · PPA forecasting test (mean, 10 settings) | ACC: 71.82 ± 3.46 -> 71.92 ± 2.73 | Helped |
| Use Causal-HMM rather than LogSparse Transformer | PPA forecasting · PPA forecasting test (mean, 10 settings) | AUC: 76.76 ± 2.74 -> 79.55 ± 1.53 | Helped |
| Use more recent observations instead of looking further into the past | PPA forecasting methods · PPA forecasting time-series settings | ACC | Helped |
Closest papers by meaning
- MIRA: Medical Time Series Foundation Model for Real-World Health Data (NeurIPS 2025 · similarity 0.83)
- Learning Causal Alignment for Reliable Disease Diagnosis (ICLR 2025 · similarity 0.83)
- Causal Representation Learning from Multimodal Biomedical Observations (ICLR 2025 · similarity 0.83)
- Integrating Sequence and Image Modeling in Irregular Medical Time Series Through Self-Supervised Learning (AAAI 2025 · similarity 0.82)
- Risk Horizons: Structured Hypothesis Spaces for Longitudinal Clinical Prediction (NeurIPS 2026 · similarity 0.82)
- Learning Mixtures of Unknown Causal Interventions (NeurIPS 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.