[A new lossy compression method for fetal heart rate signals-Convolutional Codec Network].

Area of Science:

• Biomedical Engineering
• Signal Processing
• Data Compression

Background:

• Internet of Things (IoT) systems face energy loss challenges during data transmission and storage.
• Efficient real-time monitoring of fetal heart rate is crucial for prenatal care.
• Existing fetal heart rate data compression methods may compromise signal integrity.

Purpose of the Study:

• To develop a convolutional codec network (CC-Net) for compressing fetal heart rate data.
• To reduce energy consumption in IoT systems by improving data transmission efficiency.
• To enable reliable real-time fetal heart rate monitoring.

Main Methods:

• Utilized a convolutional codec network (CC-Net) comprising encoding and decoding modules.
• Compressed fetal heart rate data using the encoding module.
• Reconstructed data in the decoding module, minimizing mean square error between original and reconstructed signals.

Main Results:

• Achieved a compression ratio of 12.07% for fetal heart rate signals.
• Maintained a low error rate of 0.03 between reconstructed and original signals.
• Demonstrated high similarity between original and reconstructed signals.

Conclusions:

• The proposed CC-Net effectively compresses fetal heart rate signals.
• CC-Net ensures minimal information loss, preserving crucial data for accurate monitoring.
• This method enhances the efficiency of fetal heart rate data transmission and storage in IoT systems.