An effective digitization method for CTG paper report with binary background grids taken by smartphone.

Area of Science:

• Medical Imaging
• Signal Processing
• Obstetrics

Background:

• Cardiotocography (CTG) is vital for prenatal diagnostics, monitoring fetal heart rate (FHR) and uterine contractions (UC).
• Digitizing CTG reports offers improved data management but existing methods struggle with binary grids and image distortion.
• Current algorithms often fail to completely extract signals from binary background grids, a common clinical format.

Purpose of the Study:

• To develop a novel digitization method for CTG paper reports with binary background grids.
• To address limitations of existing methods in signal extraction and image distortion correction.
• To enable accurate electronic conversion of CTG signals from smartphone-captured images.

Main Methods:

• A new digitization approach for CTG paper reports captured by smartphones.
• Utilizes super-pixel-based region merging and improved binary line mask removal for grid background elimination.
• Employs signal extraction based on image column states and a projection map for distortion correction.

Main Results:

• The proposed method achieves high accuracy, with average correlation coefficients of 0.9855 for FHR and 0.9866 for UC.
• Low Root Mean Square Errors (RMSE) and Mean Absolute Errors (MAE) demonstrate the method's precision for FHR and UC signals.
• Digitized CTG signals show no significant difference compared to clinical signals, validating the method's effectiveness.

Conclusions:

• The developed digitization method is effective for converting CTG paper reports into electronic signals.
• This technique offers a promising solution for the electronization of CTG data.
• The method overcomes limitations of existing digitization techniques for binary grid CTG reports.