Back
Top 29%
22.6%
Top 20%
9.2%
#1
6.1%
Top 18%
6.1%
Top 4%
6.1%
Top 11%
3.9%
Top 2%
3.9%
Top 5%
3.1%
Top 6%
2.0%
Top 3%
2.0%
Top 2%
2.0%
Top 11%
1.6%
Top 5%
1.6%
Top 57%
1.4%
Top 5%
1.4%
Top 5%
1.2%
Top 2%
1.0%
Top 5%
0.7%
Top 24%
0.7%
Top 6%
0.7%
Top 30%
0.7%
Top 26%
0.5%
A quantitative comparison between human experts and AI at estimating tumor-stroma ratio
2025-08-28
pathology
Title + abstract only
View on medRxiv
Show abstract
The tumor-stroma ratio (TSR) is an established prognostic biomarker across several cancer types, yet its manual assessment remains labour-intensive and subject to inter-observer variability. An artificial intelligence (AI)-based estimate could offer an efficient, consistent alternative for this task. In this study, quantitative comparisons were made between expert humans and an AI model for TSR estimation. Using two independent, multi-institutional histopathology datasets, an Attention U-Net was...
Predicted journal destinations
1
PLOS ONE
1737 training papers
2
Scientific Reports
701 training papers
3
The Lancet Digital Health
25 training papers
4
Nature Communications
483 training papers
5
npj Digital Medicine
85 training papers
6
eLife
262 training papers
7
Computers in Biology and Medicine
39 training papers
8
Cancers
57 training papers
9
Cureus
64 training papers
10
iScience
74 training papers
11
BMC Cancer
21 training papers
12
PLOS Digital Health
88 training papers
13
Nature Medicine
88 training papers
14
BMJ Open
553 training papers
15
JAMIA Open
35 training papers
16
Journal of Clinical Microbiology
77 training papers
17
The Journal of Molecular Diagnostics
24 training papers
18
British Journal of Cancer
22 training papers
19
Frontiers in Medicine
99 training papers
20
Frontiers in Oncology
34 training papers
21
JAMA Network Open
125 training papers
22
Journal of Medical Internet Research
81 training papers