# Forecasting Clinical Risk from Textual Time Series: Structuring Narratives for Temporal AI in Healthcare

arXiv: https://arxiv.org/abs/2504.10340 · HTML: https://mlsanity.com/paper/2504.10340/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 29 references that are in the map.
- First posted on arXiv 14 Apr 2025.

## Builds on (most cited first, 9 of 9)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing](https://mlsanity.com/paper/2111.09543/) (2021)
- [Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing](https://mlsanity.com/paper/2007.15779/) (2020)
- [Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference](https://mlsanity.com/paper/2412.13663/) (2024)
- [An Empirical Study of Multi-Task Learning on BERT for Biomedical Text Mining](https://mlsanity.com/paper/2005.02799/) (2020)
- [Temporal Supervised Contrastive Learning for Modeling Patient Risk Progression](https://mlsanity.com/paper/2312.05933/) (2023)
- [Retrieval-augmented systems can be dangerous medical communicators](https://mlsanity.com/paper/2502.14898/) (2025)
- [Reconstructing Sepsis Trajectories from Clinical Case Reports using LLMs: the Textual Time Series Corpus for Sepsis](https://mlsanity.com/paper/2504.12326/) (2025)

## Built on by (most cited first, 5 of 5)

- [Temporally Phenotyping GLP-1RA Case Reports with Large Language Models: A Textual Time Series Corpus and Risk Modeling](https://mlsanity.com/paper/2604.06197/) (2026)
- [Text Knows What, Tables Know When: Clinical Timeline Reconstruction via Retrieval-Augmented Multimodal Alignment](https://mlsanity.com/paper/2605.15168/) (2026)
- [Building Safe and Deployable Clinical Natural Language Processing under Temporal Leakage Constraints](https://mlsanity.com/paper/2602.15852/) (2026)
- [SCENARIODIFF: A Scenario-level Guidance Framework for Multimodal Time Series Forecasting--Extended Version](https://mlsanity.com/paper/2608.17164/) (2026)
- [Anchoring Clinical Events in Time: UID-Preserving Multimodal Reconstruction and Source-Grounded Adjudication](https://mlsanity.com/paper/2609.13062/) (2026)

## Abstract (arXiv)

Clinical case reports encode temporal patient trajectories that are often underexploited by traditional machine learning methods relying on structured data. In this work, we introduce the forecasting problem from textual time series, where timestamped clinical findings -- extracted via an LLM-assisted annotation pipeline -- serve as the primary input for prediction. We systematically evaluate a diverse suite of models, including fine-tuned decoder-based large language models and encoder-based transformers, on tasks of event occurrence prediction, temporal ordering, and survival analysis. Our experiments reveal that encoder-based models consistently achieve higher F1 scores and superior temporal concordance for short- and long-horizon event forecasting, while fine-tuned masking approaches enhance ranking performance. In contrast, instruction-tuned decoder models demonstrate a relative advantage in survival analysis, especially in early prognosis settings. Our sensitivity analyses further demonstrate the importance of time ordering, which requires clinical time series construction, as compared to text ordering, the format of the text inputs that LLMs are classically trained on. This highlights the additional benefit that can be ascertained from time-ordered corpora, with implications for temporal tasks in the era of widespread LLM use.