Research map / Time series / Long-term forecasting
Attend and Diagnose: Clinical Time Series Analysis Using Attention Models
Read the paperCheck 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 |
|---|---|---|---|
| replace single-task training with joint four-task training (ablation) | SAnD · MIMIC-III phenotyping | Macro AUC: 0.766 -> 0.771 | Helped |
| replace single-task training with joint four-task training (ablation) | SAnD · MIMIC-III phenotyping | Micro AUC: 0.816 -> 0.819 | No clear effect |
| replace single-task training with joint four-task training (ablation) | SAnD · MIMIC-III in-hospital mortality | AUROC: 0.857 -> 0.859 | No clear effect |
| replace single-task training with joint four-task training (ablation) | SAnD · MIMIC-III in-hospital mortality | min(Se, P+): 0.5 -> 0.504 | Helped |
| replace single-task training with joint four-task training (ablation) | SAnD · MIMIC-III decompensation | AUROC: 0.895 -> 0.908 | Helped |
| replace single-task training with joint four-task training (ablation) | SAnD · MIMIC-III decompensation | AUPRC: 0.316 -> 0.327 | Helped |
| replace single-task training with joint four-task training (ablation) | SAnD · MIMIC-III length of stay | Kappa: 0.429 -> 0.429 | No clear effect |
| replace single-task training with joint four-task training (ablation) | SAnD · MIMIC-III length of stay | MAPE: 167.3 -> 157.8 | Helped |
| increase the self-attention mask range (ablation) | SAnD · MIMIC-III phenotyping | AUC | Helped |
| increase the self-attention mask range beyond r = 96 (ablation) | SAnD · MIMIC-III phenotyping | AUC | No clear effect |
| increase the dense interpolation factor beyond the best setting (ablation) | SAnD · MIMIC-III in-hospital mortality | AUROC | Hurt |
| increase model capacity (ablation) | SAnD · MIMIC-III decompensation | AUROC | Hurt |
Closest papers by meaning
- RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism (NeurIPS 2016 · similarity 0.88)
- Understanding the Differences in Foundation Models: Attention, State Space Models, and Recurrent Neural Networks (NeurIPS 2024 · similarity 0.84)
- Revitalizing Medical Time Series with Vision-Informed Retrieval: A Vision-Language Perspective (NeurIPS 2026 · similarity 0.83)
- MAESTRO : Adaptive Sparse Attention and Robust Learning for Multimodal Dynamic Time Series (NeurIPS 2025 · similarity 0.83)
- JoLT: Jointly Learned Representations of Language and Time-Series for Clinical Time-Series Interpretation (Student Abstract) (AAAI 2024 · similarity 0.83)
- Integrating Sequence and Image Modeling in Irregular Medical Time Series Through Self-Supervised Learning (AAAI 2025 · similarity 0.83)
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.