Back

Development and Validation of a Highly Generalizable Deep Learning Pulmonary Embolism Detection Algorithm

2020-10-11 radiology and imaging Title + abstract only
View on medRxiv
Show abstract

Several algorithms have been developed for the detection of pulmonary embolism, though generalizability and bias remain potential weaknesses due to small sample size and sample homogeneity. We developed and validated a highly generalizable deep-learning algorithm, Emboleye, for the detection of PE by using a large and diverse dataset, which included 30,574 computed tomography (CT) exams sourced from over 2,000 hospital sites. On angiography exams, Emboleye demonstrates an AUROC of 0.79 with a sp...

Predicted journal destinations

1
Scientific Reports
701 training papers
#1 40.3%
2
PLOS ONE
1737 training papers
Top 37% 18.0%
3
Nature Communications
483 training papers
Top 31% 3.4%
4
PLOS Digital Health
88 training papers
Top 8% 2.4%
5
npj Digital Medicine
85 training papers
Top 8% 2.4%
6
Journal of Clinical Medicine
77 training papers
Top 9% 1.9%
7
Diagnostics
36 training papers
Top 2% 1.7%
8
BMC Medical Informatics and Decision Making
36 training papers
Top 6% 1.5%
9
Computers in Biology and Medicine
39 training papers
Top 5% 1.5%
10
Journal of the American Heart Association
92 training papers
Top 10% 1.2%
11
Medicine
29 training papers
Top 3% 1.2%
12
Heliyon
57 training papers
Top 10% 0.9%
13
BMJ Open
553 training papers
Top 64% 0.9%
14
Imaging Neuroscience
18 training papers
Top 2% 0.9%
15
Human Brain Mapping
53 training papers
Top 9% 0.9%
16
eLife
262 training papers
Top 48% 0.9%
17
NeuroImage
36 training papers
Top 6% 0.9%
18
Nature Medicine
88 training papers
Top 15% 0.9%
19
Frontiers in Neurology
74 training papers
Top 15% 0.6%
20
The Lancet Digital Health
25 training papers
Top 3% 0.6%
21
Frontiers in Oncology
34 training papers
Top 7% 0.6%
22
NeuroImage: Clinical
77 training papers
Top 14% 0.6%
23
Scientific Data
30 training papers
Top 5% 0.6%
24
PeerJ
46 training papers
Top 10% 0.6%
25
Cureus
64 training papers
Top 25% 0.6%
26
Cancers
57 training papers
Top 9% 0.6%