Reliable Extraction of Clinical Follow-Up Instructions: A Hybrid Neural-Symbolic Pipeline
Summary
This paper presents a hybrid neural-symbolic pipeline for extracting follow-up instructions from clinical notes, using BioBERT and deterministic date arithmetic. It achieves high performance (Pair F1 ~0.99) compared to generative baselines.
View Cached Full Text
Cached at: 05/27/26, 09:07 AM
# Reliable Extraction of Clinical Follow-Up Instructions: A Hybrid Neural-Symbolic Pipeline
Source: [https://arxiv.org/abs/2605.26560](https://arxiv.org/abs/2605.26560)
[View PDF](https://arxiv.org/pdf/2605.26560)
> Abstract:Objective\. Outpatient notes carry follow\-up instructions pairing actions with future times \("MRI brain in two weeks"\)\. Extracting \(action, date\) pairs supports scheduling and audit, but generative extractors miss the date because linking and arithmetic are implicit in decoding\. We test a hybrid neural\-symbolic pipeline against direct generation\. Methods\. We define TestSpecification and TimeSpecification entities and a ScheduledFor relation\. BioBERT feeds BIO tagging and a biaffine linker; entities are canonicalized via a 28\-action ontology and times normalized to day offsets deterministically\. We evaluate on a 2,000\-note synthetic outpatient corpus with action\-disjoint splits \(18 train, 6 OOV\-test\) against zero\-shot GPT\-4o\-mini and LoRA\-fine\-tuned LLaMA\-3 8B with note\-level bootstrap 95% CIs\. Results\. On 259\-note seen and OOV splits the hybrid pipeline achieves Test\-Time Pair F1 of 0\.997 and 0\.986 with 0\.00\-day MAE\. Baselines reach high action F1 \(LLaMA\-3 0\.992; GPT\-4o\-mini 0\.963 seen\) but Pair F1 stays at 0\.51\-0\.57 \(LLaMA\-3\) and 0\.53 \(GPT\-4o\-mini\), CIs non\-overlapping with the hybrid\. Conclusion\. Separating learned entity extraction from deterministic date arithmetic outperforms generation on this benchmark, generalizes to held\-out actions, and exposes failure modes\. Transfer to real EHR notes is the next validation; a first\-pass realism check is in Limitations\.
## Submission history
From: Yehudit Aperstein \[[view email](https://arxiv.org/show-email/0d2b913b/2605.26560)\] **\[v1\]**Tue, 26 May 2026 05:14:33 UTC \(939 KB\)Similar Articles
How Compliant is Sepsis Treatment? An Expert-Guided Neuro-symbolic Pipeline for Generating Clinical Compliance Insights
This paper introduces an expert-guided neuro-symbolic pipeline combining LLMs for semantic normalization and fuzzy logic to assess compliance with sepsis treatment protocols, providing graded insights from clinical data.
Retrieval-Augmented Large Language Models for Schema-Constrained Clinical Information Extraction
This paper presents a modular retrieval-augmented generation (RAG) pipeline for extracting structured clinical observations from conversational nurse-patient transcripts, using schema-constrained prompting and second-pass auditing with Llama and GPT backbones, achieving 80.36% F1 score.
AI-assisted Protocol Information Extraction For Improved Accuracy and Efficiency in Clinical Trial Workflows
Researchers from Banting Health AI present an AI system using generative LLMs with Retrieval-Augmented Generation (RAG) for automated clinical trial protocol information extraction, achieving 89% accuracy compared to 62.6% for standalone LLMs, with AI-assisted workflows completing tasks 40% faster and reducing cognitive demand.
Neuro-Symbolic AI for LEED compliance: Document-Centric Benchmarking, Deterministic Numeric Checking, and When Multimodal Hurts
This paper introduces a neuro-symbolic pipeline for automating LEED v4.1 BD+C compliance verification using small locally deployed language models and deterministic numeric checking. Experiments on four university buildings show that a 4B model outperforms an 8B model, and the deterministic checker corrects arithmetic errors on key credits, though multimodal inputs reduce accuracy.
Training Large Language Models to Predict Clinical Events
This paper extends Foresight Learning to clinical event prediction by converting time-ordered clinical notes into prediction examples. A LoRA adapter on a 120B model improves calibration and outperforms GPT-5 on held-out questions.