Rule-Based Algorithm for Intrapartum Cardiotocograph Pattern Features Extraction and Classification

Computerized FHR features extraction and Cardiotocograph (CTG) classification would support obstetricians in CTG analysis and enhance their interpretation. The main objective of this paper is to develop a simple reliable algorithm for FHR feature extraction and systematic CTG classification based on MATLAB rule based functions and the Royal College of Obstetricians and Gynecologists (RCOG) guideline. An application was developed to extract the basic features of FHR, such as: baseline, baseline variability, acceleration and deceleration based on (RCOG) guideline. A classification system using MATLAB was then introduced to classify the CTG signal patterns into normal, suspicious or pathological based on the RCOG guideline. The computerized interpretations of 80 CTG signals were compared with the visual interpretation by five obstetricians. The obtained results show slight difference of about ±2 beats per minutes (b.p.m.) and 5 b.p.m. for variability. The algorithm and obstetricians were in agreement on number and type of decelerations but differed on the number of accelerations by up to (±4), results in classifying CTG signals. These results are considered promising for the use of computerised CTG interpretation in the hospital as well as a component homecare facilities for pregnant women.

Author(s):

Shahad Alyousif, MA Mohd, Bilal B, M Sheikh and M Algunaidi

Abstract | Full-Text | PDF

Share this