Large Language Models for Structured Information Extraction from German Histopathology Reports - Hepatocellular Carcinoma
Koch, N.; Hartmann, N.; Maennel, A.; Baretton, G.; Aust, D.; Hoenscheid, P.
Show abstract
Background: Clinical registries and institutional quality control (QC) in pathology heavily depend on complex histopathological data. However, critical tumor characteristics are mostly locked in unstructured free-text reports. We hypothesize that locally deployed, resource-efficient language models operating within a secure institutional network can reliably extract relevant variables from German reports of hepatocellular carcinoma (HCC), thereby enabling automated data curation without transferring sensitive patient data. Methods: We developed a local, privacy-preserving open-weight Large Language Model (LLM) pipeline consisting of two consecutive stages: Stage 1 utilizes a lightweight gatekeeper model to isolate the target cohort of confirmed Hepatocellular Carcinoma cases from a screening pool of N = 2,864 reports. Stage 2 deploys a high-capacity model to extract 15 predefined histopathological variables, including tumor-node-metastasis staging, histological grading, vascular invasion, and selected immunohistochemical features into structured JSON format. For both Stages, pipeline performance was evaluated using Accuracy, Macro F1-scores, and Cohen's kappa compared to expert-curated reference annotations. Results: The top-performing Stage 1 gatekeeper isolated the target patient cohort with zero classification errors, effectively reaching complete agreement with the reference standard (ground truth) and correcting inaccurate administrative metadata. For Stage 2 feature extraction, current-generation models demonstrated robust structural reliability, reducing the burden of manual post-processing. The 31B model configuration (Gemma 4) virtually eliminated objective factual errors, achieving near-perfect agreement with the reference standard (Macro F1 > 0.99). Conclusion: Local two-stage pipelines offer a highly accurate, resource-efficient, and secure solution for clinical text processing. This framework serves as a privacy-preserving engine for institutional quality control and the scalable curation of pathology registries.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Large language models improve transferability of electronic health record-based predictions across countries and coding systems 94%
- Enhancing Privacy-Preserving Deployable Large Language Models for Perioperative Complication Detection: A Targeted Strategy with LoRA Fine-tuning 93%
- High-Sensitivity Pan-Cancer AI Assessment of Lymph Node Metastasis via Uncertainty Quantification 92%
Similar papers in this journal
- Development and validation of AI-based pre-screening of large bowel biopsies 92%
- CARDBiomedBench: A Benchmark for Evaluating Large Language Model Performance in Biomedical Research 91%
- Novel deep learning algorithm predicts the status of molecular pathways and key mutations in colorectal cancer from routine histology images 89%
Similar papers in this journal
Similar papers in this journal
- Comparison of local large language models for extraction of signs and symptoms data from electronic health records 94%
- GOcats: A tool for categorizing Gene Ontology into subgraphs of user-defined concepts 91%
- tbiExtractor: A framework for Extracting Traumatic Brain Injury Common Data Elements from Radiology Reports 90%
Similar papers in this journal
- Large Language Models for Zero-Shot Procedure Extraction in Orthopedic Surgery: A Comparative Evaluation 94%
- Systematic identification of rare disease patients in electronic health records enables evaluation of clinical outcomes 92%
- Leveraging Temporal Learning with Dynamic Range (TLDR) for Enhanced Prediction of Outcomes in Recurrent Exposure and Treatment Settings in Electronic Health Records 91%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.