Back

OncoRAG: Graph-Based Retrieval Enabling Clinical Phenotyping from Oncology Notes Using Local Mid-Size Language Models

2026-03-06 oncology Title + abstract only
View on medRxiv
Show abstract

Introduction: Manual data extraction from unstructured clinical notes is labor-intensive and impractical for large-scale clinical and research operations. Existing automated approaches typically require large language models, dedicated computational infrastructure, and/or task-specific fine-tuning that depends on curated data. The objective of this study is to enable accurate extraction with smaller locally deployed models using a disease-site specific pipeline and prompt configuration that are ...

Predicted journal destinations

1
Cancers
57 training papers
Top 3% 8.7%
2
PLOS ONE
1737 training papers
Top 62% 8.7%
3
Nature Communications
483 training papers
Top 13% 7.6%
4
Frontiers in Oncology
34 training papers
Top 3% 6.2%
5
Scientific Reports
701 training papers
Top 41% 6.2%
6
Clinical Cancer Research
22 training papers
Top 2% 5.8%
7
Cancer Medicine
17 training papers
Top 0.4% 4.9%
8
eLife
262 training papers
Top 11% 4.0%
9
British Journal of Cancer
22 training papers
Top 1% 4.0%
10
npj Digital Medicine
85 training papers
Top 6% 4.0%
11
JAMA Network Open
125 training papers
Top 5% 3.7%
12
BMJ Open
553 training papers
Top 40% 3.7%
13
BMC Cancer
21 training papers
Top 1% 2.7%
14
PLOS Computational Biology
141 training papers
Top 7% 1.9%
15
Computers in Biology and Medicine
39 training papers
Top 7% 1.1%
16
Communications Medicine
63 training papers
Top 3% 1.1%
17
Journal of Medical Internet Research
81 training papers
Top 16% 1.0%
18
PLOS Digital Health
88 training papers
Top 16% 0.9%
19
Proceedings of the National Academy of Sciences
100 training papers
Top 17% 0.9%
20
Journal of the American Medical Informatics Association
53 training papers
Top 6% 0.9%
21
Diagnostics
36 training papers
Top 7% 0.7%
22
iScience
74 training papers
Top 14% 0.7%
23
International Journal of Molecular Sciences
39 training papers
Top 7% 0.7%
24
Scientific Data
30 training papers
Top 5% 0.7%
25
PeerJ
46 training papers
Top 9% 0.7%
26
International Journal of Cancer
18 training papers
Top 2% 0.7%