Back

Distance and Colour-based Scores for Blood Test Risk Stratification

Zenil, H.; Hernandez-Quiroz, F.; Hernandez-Orozco, S.; Saeb-Parsy, K.; Hernandez de la Cerda, H.; Riedel, J.

2020-02-10 bioinformatics
10.1101/2020.02.09.941096 bioRxiv
Show abstract

We introduce a numerical and a colour-based risk stratification score to quantify abnormal blood analyte values. The score indicates how removed values of an individual are from considered healthy ranges from the literature or derived from empirical data such as medical surveys. The scores behaviour can be adjusted to incorporate medical knowledge by assigning multipliers or weights to individual components and is rooted on a numerical and a colour-based scheme. We test the score against real and synthetic data from medically relevant cases, extremes cases, and empirical blood cell count data from the CDC NHANES survey spanning 13 years, from 2003 to 2016. We find that both the numerical and colour-based scores are informative in distinguishing healthy individuals from those with diseases manifested with abnormal blood results.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

1
PLOS ONE
● 5266 papers in training set
Top 11%
17.3%
Similar papers in this journal
2
Computers in Biology and Medicine
○ 128 papers in training set
Top 0.2%
8.0%
Similar papers in this journal
3
PLOS Computational Biology
● 1863 papers in training set
Top 5%
6.9%
Similar papers in this journal
4
BMC Bioinformatics
○ 457 papers in training set
Top 2%
5.6%
Similar papers in this journal
5
Scientific Reports
○ 3612 papers in training set
Top 19%
4.9%
Similar papers in this journal
6
Computational and Structural Biotechnology Journal
● 242 papers in training set
Top 1%
3.3%
7
Bioinformatics Advances
◐ 203 papers in training set
Top 2%
3.3%
8
Briefings in Bioinformatics
◐ 354 papers in training set
Top 3%
3.3%
50% of probability mass above
9
Cytometry Part A
○ 33 papers in training set
Top 0.1%
3.3%
10
Bioinformatics
◐ 1204 papers in training set
Top 5%
3.2%
11
Frontiers in Immunology
○ 638 papers in training set
Top 4%
2.5%
12
BioData Mining
○ 22 papers in training set
Top 0.1%
2.5%
13
Forensic Science International: Genetics
○ 26 papers in training set
Top 0.1%
1.8%
14
Journal of Immunological Methods
○ 24 papers in training set
Top 0.2%
1.5%
15
BMC Medical Genomics
○ 50 papers in training set
Top 0.6%
1.5%
16
Frontiers in Bioinformatics
○ 49 papers in training set
Top 0.6%
1.5%
17
Royal Society Open Science
● 214 papers in training set
Top 4%
1.4%
18
Patterns
○ 78 papers in training set
Top 2%
1.1%
19
Sensors
○ 43 papers in training set
Top 0.9%
1.1%
20
Heliyon
○ 152 papers in training set
Top 6%
1.0%
21
Communications Biology
○ 993 papers in training set
Top 28%
0.9%
22
GigaScience
◐ 212 papers in training set
Top 4%
0.9%
23
Life
○ 29 papers in training set
Top 0.8%
0.9%
24
International Journal of Molecular Sciences
○ 494 papers in training set
Top 14%
0.9%
25
Frontiers in Artificial Intelligence
○ 20 papers in training set
Top 0.7%
0.9%
26
Applied Sciences
○ 25 papers in training set
Top 1%
0.6%
27
Cancers
○ 213 papers in training set
Top 5%
0.6%
28
BMC Medical Informatics and Decision Making
○ 43 papers in training set
Top 2%
0.6%
29
Biology
○ 45 papers in training set
Top 1%
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.