Back
Top 48%
14.2%
Top 22%
8.7%
Top 4%
6.2%
Top 17%
6.2%
Top 4%
5.8%
Top 3%
4.9%
Top 3%
4.9%
Top 0.1%
4.0%
Top 3%
4.0%
Top 4%
4.0%
Top 2%
3.1%
Top 50%
1.9%
Top 23%
1.9%
Top 4%
1.5%
Top 9%
1.5%
Top 4%
1.3%
Top 14%
1.1%
Top 4%
1.1%
Top 14%
1.1%
Top 11%
0.7%
Top 14%
0.7%
Top 9%
0.7%
Top 19%
0.7%
Top 7%
0.7%
An agentic AI system enhances clinical detection of immunotherapy toxicities: a multi-phase validation study
2026-03-02
oncology
Title + abstract only
View on medRxiv
Show abstract
Immune-related adverse events (irAEs) affect up to 40% of patients receiving immune checkpoint inhibitors, yet their identification depends on laborious and inconsistent manual chart review. Here we developed and evaluated an agentic large language model system to extract the presence, temporality, severity grade, attribution, and certainty of six irAE types from clinical notes. Retrospectively (263 notes), the system achieved macro-averaged F1 of 0.92 for detection and 0.66 for multi-class seve...
Predicted journal destinations
1
PLOS ONE
1737 training papers
2
Scientific Reports
701 training papers
3
Cancers
57 training papers
4
Nature Communications
483 training papers
5
npj Digital Medicine
85 training papers
6
JAMA Network Open
125 training papers
7
Frontiers in Oncology
34 training papers
8
Scientific Data
30 training papers
9
Clinical Cancer Research
22 training papers
10
Journal of the American Medical Informatics Association
53 training papers
11
Cancer Medicine
17 training papers
12
BMJ Open
553 training papers
13
eLife
262 training papers
14
British Journal of Cancer
22 training papers
15
PLOS Computational Biology
141 training papers
16
BMC Cancer
21 training papers
17
Journal of Medical Internet Research
81 training papers
18
Communications Medicine
63 training papers
19
Proceedings of the National Academy of Sciences
100 training papers
20
Computers in Biology and Medicine
39 training papers
21
iScience
74 training papers
22
PeerJ
46 training papers
23
Nature Medicine
88 training papers
24
Diagnostics
36 training papers