Radiological risk stratification and management planning are essential downstream tasks in medical imaging. Automating these tasks is challenging because clinical guideline criteria are distributed across prose, tables, and cross-referenced definitions. We present Guideline-derived Decision Graph (GDGraph), a framework that decouples guideline compilation from case evaluation. During compilation, a large language model partitions guidelines into section-based chunks, extracts local decision subgraphs with linked definitions, and aggregates related branches into an executable decision graph. During inference, a fast System One model answers localized clinical questions while deterministic traversal resolves Boolean logic, turning opaque single-pass predictions into a traceable decision path without task-specific training. Across 347 cases spanning renal cyst, adnexal mass, and thyroid nodule guidelines, GDGraph improves System One accuracy from 57.3% to 90.8%, outperforming retrieval-augmented GPT-4o (80.1%) and graph-guided GPT-4o with chain-of-thought (82.7%) at substantially lower latency and cost. The source code will be available.
Figures & tables
Figure 1: Overview of the GDGraph framework. (a) LLM-based Graph Construction: Offline, an LLM partitions clinical guidelines into m section-based chunks {Ci}i=1m , extracts local subgraphs {Gi}i=1m with TF-IDF-retrieved definitions from a global memory, and aggregates related branches into a unified decision graph G . (b) Graph-Guided Selection: Online, a System One model answers localized questions using case findings and attached definitions, guiding traversal along the selected path to reach a traceable decision.
1:
Partition D into section-based chunks C={C1,…,Cm}
2:
Build definition memory M←⋃i=1mExtractTerms(Ci)
3:
for each chunk Ci∈C do
4:
Retrieve relevant definitions di⊂M via TF-IDF
5:
Extract subgraph Gi=(Si,Ai,Ei,Yi) from (Ci,di)
6:
Render questions for each a∈Ai with Da⊂M
7:
end for
Algorithm 1 GDGraph Construction
Guideline
Clinical Task
Cases
CUA (2023) [ 11 ]
Renal cyst management
127
O-RADS US v2022 [ 15 ]
Adnexal mass stratification
117
K-TIRADS (2021) [ 4 ]
Thyroid nodule stratification
103
Total
347
Table 1: Summary of evaluated clinical guidelines, target tasks, and case counts.
Guideline
Node Prec.
Rule Prec.
CUA
137/137 (100.0%)
51/55 (92.7%)
O-RADS
136/137 (99.3%)
92/98 (93.9%)
K-TIRADS
216/216 (100.0%)
123/123 (100.0%)
Overall
489/490 (99.8%)
266/276 (96.4%)
Table 2: Precision of constructed decision graphs against source guidelines, reported as supported/total (%).
Clinical practice guidelines are long, multimodal documents whose branching recommendations are difficult to convert into executable clinical decision support (CDS), and one-shot parsing often breaks cross-page continuity. Recent LLM/VLM extractors are mostly local or text-centric, under-specifying section interfaces and failing to consolidate cross-page control flow across full documents into one coherent decision graph. We present a decomposition-first pipeline that converts full-guideline evidence into an executable clinical decision graph through topology-aware chunking, interface-constrained chunk graph generation, and provenance-preserving global aggregation. Rather than relying on single-pass generation, the pipeline uses explicit entry/terminal interfaces and semantic deduplication to preserve cross-page continuity while keeping the induced control flow auditable and structurally consistent. We evaluate on an adjudicated prostate-guideline benchmark with matched inputs and the same underlying VLM backbone across compared methods. On the complete merged graph, our approach improves edge and triplet precision/recall from 19.6%/16.1% in existing models to 69.0%/87.5%, while node recall rises from 78.1% to 93.8%. These results support decomposition-first, auditable guideline-to-CDS conversion on this benchmark, while current evidence remains limited to one adjudicated prostate guideline and motivates broader multi-guideline validation.
Onur Selim Kilic, Yeti Z. Gurbuz, Cem O. Yaldiz +4
Georgia Institute of Technology · MetaDialog · Infuse Inc.
Clinical practice guidelines (CPGs) encode diagnostic criteria, but LLM systems typically retrieve guideline text or absorb it through training rather than execute its rules. We introduce GuideSkill, an external reasoning layer that compiles disease-specific criteria into executable functions returning ordinal diagnostic-support scores. GuideSkill-Zero is initialized from guidelines, while GuideSkill-Evo uses case--diagnosis pairs to refine covered skills and add missing diagnoses. At inference, an LLM proposes a differential diagnosis, grounds the features required by each matched skill, and fuses its ranking with the executed skill scores. Across four benchmarks and four backbones, GuideSkill-Zero improves macro-average accuracy over guideline RAG by 13.45% on average. GuideSkill-Evo achieves the highest macro-average for every backbone, improves over direct inference by 18.49% relatively, and increases gold-label skill coverage from 56.5% to 99.5%. On Qwen3.5-9B, it also exceeds the strongest parameter-update baseline by 11.16% without updating the backbone. Expert evaluation further indicates that GuideSkill produces clinically sound and broadly acceptable skills, suggesting that its initialized and evolved rules are reliable and practically meaningful. These results support executable skills as a model-agnostic mechanism for combining guideline-derived procedures with case-derived diagnostic patterns.
Lang Cao, Yuhao Shen, Tianyang Luo +3
University of Illinois Urbana-Champaign · ♦The Chinese University of Hong Kong, Shenzhen · ⋂Albert Einstein College of Medicine
Clinical practice guidelines (CPGs) encode evidence-based decision logic that clinicians apply by evaluating patient variables, conditional criteria, and recommendation rules. However, existing methods often use CPGs as free-text training data or retrieval sources, underutilizing their procedural decision structure. To better exploit this structure, we introduce a guideline-derived training pipeline that transforms CPG recommendations into executable clinical decision logic and uses it to generate factual and counterfactual question-answering data. Theses data teach models both guideline-supported decisions and how decisions change under different patient conditions. Post-training a medical LLM on the generated data yields MedGuideX. Across four clinical reasoning benchmarks, MedGuideX achieves a 10.28% relative improvement in average accuracy. Physician evaluation further shows that MedGuideX better recovers clinician authored reasoning steps and produces physician-preferred rationales in faithfulness, validity, completeness, and clarity. Overall, our results show that executable decision logic from CPGs can be transformed into scalable supervision for building reliable medical LLMs.
Yuhao Shen, Lang Cao, Simo Du +4
University of Illinois Urbana-Champaign · 2The Chinese University of Hong Kong, Shenzhen · 3Albert Einstein College of Medicine