Automatic Detection of Fetal QRS Complex using Time-Frequency Image Based Features and Deep Learning Architecture

Krupa, Abel Jaba Deva; Dhanalakshmi, Samiappan

doi:10.1109/icesc54411.2022.9885294

Cited by 1 publication

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in Table I, values of Tn presented. Parameters of classification can be calculated, as shown in equations (27)(28)(29)(30)(31)(32)(33). Where t method is MMFCC using MST instead of Where p equals 1 in equation ( 6).…”

Section: Figure 1 Steps Of Feature Extraction Usingmentioning

confidence: 99%

See 1 more Smart Citation

Gender of Fetus Identification Using Modified Mel-Frequency Cepstral Coefficients Based on Fractional Discrete Cosine Transform

Azmy

2024

IEEE Access

View full text Add to dashboard Cite

Fetal heart sounds is measured to follow the developing status of fetus. The used database of fetal heart sounds is obtained from Physionet challenge. In this paper, novel models are created to extract features from fetal heart sounds; to identify the gender of fetus (male or female), using modified Mel-Frequency Cepstral Coefficients (MMFCC). The high pass filter which is used in the pre-emphasis step of MMFCC, used to remove noise from fetal heart sounds. These models are compared; to obtain high classification parameters (i.e. the accuracy rate and the area under curve). Classification method is based on deep learning, using recurrent neural networks (RNN) based on bidirectional long short-term memory (BiLSTM). The programming code of 13 models is created by the author, using Matlab software environment. Both accuracy rate and area under curve (AUC) are obtained using MMFCC based on fractional discrete cosine transform (FRDCT), overperform the same results of other models. So, this model is selected to extract features of fetal heart sounds. The contribution here is using novel model. FRDCT is new fractional transform, which is decomposed from fractional Fourier transform (FRFT) and discrete cosine transform (DCT). The secret behind selecting FRDCT based on MMFCC as a model; FRDCT decomposes signals perfectly in time-frequency domain. The best model can be programmed on a portable device; to know the gender of fetus by a caregiver in rural areas, without needing to use ultrasound.

show abstract

Section: Figure 1 Steps Of Feature Extraction Usingmentioning

confidence: 99%

“…A.J.D. Krupa and S. Dhanalakshmi made researches about automatic detection of fetal QRS complex using ST based on deep learning [33]. A.J.D.…”

Section: Oefficients Of Mfcc Based On Frdctmentioning

confidence: 99%

Gender of Fetus Identification Using Modified Mel-Frequency Cepstral Coefficients Based on Fractional Discrete Cosine Transform

Azmy

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

Automatic Detection of Fetal QRS Complex using Time-Frequency Image Based Features and Deep Learning Architecture

Cited by 1 publication

References 24 publications

Gender of Fetus Identification Using Modified Mel-Frequency Cepstral Coefficients Based on Fractional Discrete Cosine Transform

Gender of Fetus Identification Using Modified Mel-Frequency Cepstral Coefficients Based on Fractional Discrete Cosine Transform

Contact Info

Product

Resources

About