ctg: Data on automatic analysis of cardiotocograms

In diversityForest: Innovative Complex Split Procedures in Random Forests Through Candidate Split Sampling

Data on automatic analysis of cardiotocograms

Description

This data set contains measurements from 2126 fetal cardiotocograms (CTGs). The CTGs were automatically processed and the respective diagnostic features measured. The CTGs were also classified by three expert obstetricians and a consensus classification label assigned to each of them. This description is taken from the UC Irvine Machine Learning Repository, where this data set was downloaded from. The outcome CLASS is categorical with ten classes that correspond to different fetal heart rate patterns. See the 'Details' section below for further information.

Format

A data frame with 2126 observations, 24 covariates and one 10-class outcome variable

Details

The variables are as follows:

• b. numeric. Start instant

• e. numeric. End instant

• LBE. numeric. Fetal heart rate (FHR) baseline value assessed by medical expert (beats per minute)

• LB. numeric. FHR baseline value assessed by SisPorto (beats per minute)

• AC. numeric. Number of accelerations per second

• FM. numeric. Number of fetal movements per second

• UC. numeric. Number of uterine contractions per second

• DL. numeric. Number of light decelerations per second

• DS. numeric. Number of severe decelerations per second

• DP. numeric. Number of prolonged decelerations per second

• ASTV. numeric. Percentage of time with abnormal short term variability

• MSTV. numeric. Mean value of short term variability

• ALTV. numeric. Percentage of time with abnormal long term variability

• MLTV. numeric. Mean value of long term variability

• Width. numeric. Width of FHR histogram

• Min. numeric. Minimum value of FHR histogram

• Max. numeric. Maximum value of FHR histogram

• Nmax. numeric. Number of histogram peaks

• Nzeros. numeric. Number of histogram zeros

• Mode. numeric. Mode of the histogram

• Mean. numeric. Mean of the histogram

• Median. numeric. Median of the histogram

• Variance. numeric. Variance of the histogram

• Tendency. factor. Histogram tendency (-1 for left asymmetric; 0 for symmetric; 1 for right asymmetric)

• CLASS. factor. FHR pattern class

⁠ ⁠
The classes of the outcome CLASS are as follows:

• A. Calm sleep

• B. REM sleep

• C. Calm vigilance

• D. Active vigilance

• SH. Shift pattern (A or SUSP with shifts)

• AD. Accelerative/decelerative pattern (stress situation)

• DE. Decelerative pattern (vagal stimulation)

• LD. Largely decelerative pattern

• FS. Flat-sinusoidal pattern (pathological state)

• SUSP. Suspect pattern

This is a pre-processed version of the "Cardiotocography" data set published in the UC Irvine Machine Learning Repository. The raw data contained the four additional variables Date, FileName, SegFile, and NSP, which were removed in this version of the data. Moreover, the variable DR, representing the number of repetitive decelerations per second was removed as well because it was 0 for all observations.

Source

UC Irvine Machine Learning Repository, link: https://archive.ics.uci.edu/dataset/193/cardiotocography/ (Accessed: 29/08/2024)

References

• Ayres-de Campos, D., Bernardes, J., Garrido, A., Marques-de-Sá, J., Pereira-Leite, L. (2000). SisPorto 2.0: a program for automated analysis of cardiotocograms. J Matern Fetal Med. 9(5):311-318, <\Sexpr[results=rd]{tools:::Rd_expr_doi("10.1002/1520-6661(200009/10)9:5<311::AID-MFM12>3.0.CO;2-9")}>.

• Dua, D., Graff, C. (2019). UCI Machine Learning Repository. Irvine, CA: University of California, School of Information and Computer Science. https://archive.ics.uci.edu/ml/.

Examples

# Load data:
data(ctg)

# Numbers of observations per outcome class:
table(ctg$CLASS)

# Dimension of data:
dim(ctg)

# First rows of data:
head(ctg) 

diversityForest documentation built on June 8, 2025, 1:23 p.m.