Back

AI/ML Models to Aid in the Diagnosis of COVID-19 Illness from Forced Cough Vocalizations: Good Machine Learning Practice and Good Clinical Practices from Concept to Consumer for AI/ML Software Devices

Kelley, K.; Kelley, M.; Kelley, S. C.; Sakara, A. A.; Ramirez, M. A.

2021-11-15 health policy
10.1101/2021.11.13.21266289 medRxiv
Show abstract

From a comprehensive and systematic search of the relevant literature on signal data signature (SDS)-based artificial intelligence/machine learning (AI/ML) systems designed to aid in the diagnosis of COVID-19 illness, we identified the highest quality articles with statistically significant data sets for a head-to-head comparison to our own model in development. Further comparisons were made to the recently released "Good Machine Learning Practice (GMLP) for Medical Device Development: Guiding Principles" and, in conclusions, we proposed supplemental principles aimed at bringing AI/ML technologies in closer alignment GMLP and Good Clinical Practices (GCP).

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

1
PLOS ONE
● 5266 papers in training set
Top 4%
31.7%
Similar papers in this journal
2
npj Digital Medicine
○ 118 papers in training set
Top 0.8%
6.9%
Similar papers in this journal
3
BMJ Health & Care Informatics
● 15 papers in training set
Top 0.2%
5.6%
Similar papers in this journal
4
Journal of the American Medical Informatics Association
◐ 71 papers in training set
Top 0.6%
5.6%
Similar papers in this journal
5
PLOS Digital Health
● 106 papers in training set
Top 1%
5.0%
Similar papers in this journal
50% of probability mass above
6
BMC Medical Informatics and Decision Making
○ 43 papers in training set
Top 0.5%
4.1%
7
Trials
○ 29 papers in training set
Top 0.4%
2.4%
8
BMJ Open
● 601 papers in training set
Top 8%
2.2%
9
JMIR Medical Informatics
◐ 18 papers in training set
Top 0.3%
2.2%
10
F1000Research
○ 88 papers in training set
Top 1%
1.8%
11
JAMA Network Open
● 130 papers in training set
Top 2%
1.8%
12
Scientific Reports
○ 3612 papers in training set
Top 56%
1.5%
13
Frontiers in Public Health
○ 148 papers in training set
Top 4%
1.5%
14
Journal of Medical Internet Research
◐ 87 papers in training set
Top 2%
1.5%
15
eLife
● 5828 papers in training set
Top 61%
1.1%
16
Computer Methods and Programs in Biomedicine
○ 28 papers in training set
Top 0.8%
1.0%
17
International Journal of Medical Informatics
○ 26 papers in training set
Top 1%
1.0%
18
Clinical and Translational Science
○ 22 papers in training set
Top 0.6%
0.9%
19
Diagnostics
○ 50 papers in training set
Top 2%
0.9%
20
JMIRx Med
◐ 32 papers in training set
Top 2%
0.9%
21
JAMIA Open
◐ 42 papers in training set
Top 1%
0.9%
22
BMC Cardiovascular Disorders
○ 18 papers in training set
Top 0.9%
0.9%
23
BMC Medical Education
○ 21 papers in training set
Top 0.6%
0.9%
24
Cureus
○ 68 papers in training set
Top 4%
0.9%
25
BMC Medicine
○ 176 papers in training set
Top 4%
0.9%
26
Clinical Trials
○ 11 papers in training set
Top 0.3%
0.6%

"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.