This paper proposes a big data Doppler scan image classification system that uses empirical mode decomposition (EMD) and dual sigmoid activation function-based faster region convolutional neural network (R-CNN). This approach initially arranges the pixels of the Doppler scan images in a zig-zag order sequence. This one-dimensional sequence is decomposed to L number of intrinsic mode function (IMF) using the EMD algorithm. The spectrums of the decomposed IMF are estimated using the one-dimensional Fourier transforms. Required IMFs are then selected based on the frequency spacing estimated on the Fourier spectrum. The resultant image is then reconstructed using the selected IMF’s. The resultant Doppler scan image has less redundant information and is trained using a faster RCNN algorithm. Instead of using the traditional activation functions, the proposed faster RCNN uses a dual sigmoid activation function that classifies the Doppler images into five classes. The classes are namely Maternal cervix, Thorax, Femur, Brain, Abdomen, and other regions. The experimental evaluation uses the parametersnamelyF1 score, specificity, accuracy, sensitivity, precision, and time complexity with the big data Doppler ultrasound scan image dataset that contains 12400 images collected from 1792 patients.

Ms.S. Sandhya Kumari,Dr.K.Sandhya Rani."Empirical mode Decomposition and Dual Sigmoid Activation Function-Based Faster RCNN for Big Data Doppler Scan Image Classification". International Journal of Computer Engineering In Research Trends (IJCERT) ,ISSN:2349-7084 ,Vol.8, Issue 09,pp.151-165, September - 2021, URL :https://ijcert.org/ems/ijcert_papers/V8I901.pdf,

[1]LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nat. 521, 436–444 (2015).
[2]Hosny, A., Parmar, C., Quackenbush, J., Schwartz, L. H. &Aerts, H. J. Artificial intelligencein radiology. Nat. Rev. Cancer 18, 500 (2018).
[3] Das, Sraddha, et al. "Deep learning architecture based on segmented fundus image features for classification of diabetic retinopathy." Biomedical Signal Processing and Control 68 (2021): 102600.
[4] Suchetha, M., Rajiv Raman, and Edwin Dhas. "Region of Interest based Predictive Algorithm for SubretinalHemorrhage Detection using Faster R-CNN." (2021).
[5] Litjens, G. et al. A survey on deep learning in medical image analysis. Med. image analysis 42, 60–88 (2017).
[6] Esteva, A. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nat. 542, 115 (2017).
[7] Maraci, M., Bridge, C., Napolitano, R., Papageorghiou, A. & Noble, J. A framework for analysis of linear ultrasound videos to detect fetal presentation and heartbeat. Med. Image Analysis 37, 22–36 (2017).
[8] Ryou, H. et al. Automated 3d ultrasound biometry planes extraction for first trimester fetal assessment. In Machine Learning inMedical Imaging, 196–204 (2016).
[9] Li, Y. et al. Standard plane detection in 3d fetal ultrasound using an iterative transformation network. Medical Image Computing andComputer Assisted Intervention – MICCAI 2018, 392–400 (2018).
[10] Lee, LokHin, Yuan Gao, and J. Alison Noble. "Principled Ultrasound Data Augmentation for Classification of Standard Planes." International Conference on Information Processing in Medical Imaging. Springer, Cham, 2021.
[11]Attallah, Omneya, Maha A. Sharkas, and HebaGadelkarim. "Fetal brain abnormality classification from MRI images of different gestational age." Brain sciences 9.9 (2019): 231.
[12]Sridar, Pradeeba, et al. "Decision fusion-based fetal ultrasound image plane classification using convolutional neural networks." Ultrasound in medicine & biology 45.5 (2019): 1259-1273.
[13] Xie, H. N., et al. "Using deep?learning algorithms to classify fetal brain ultrasound images as normal or abnormal." Ultrasound in Obstetrics &Gynecology 56.4 (2020): 579-587.
[14] Sushma, T. V., et al. "Classification of Fetal Heart Ultrasound Images for the Detection of CHD." Innovative Data Communication Technologies and Application. Springer, Singapore, 2021. 489-505.
[15]Qiao, Sibo, et al. "RLDS: An explainable residual learning diagnosis system for fetal congenital heart disease." Future Generation Computer Systems 128 (2022): 205-218.
[16]Subba, Basant, and Prakriti Gupta. "A tfidfvectorizer and singular value decomposition based host intrusion detection system framework for detecting anomalous system processes." Computers & Security 100 (2021): 102084.
[17]Dubey, Rahul, et al. "Automated diagnosis of muscle diseases from EMG signals using empirical mode decomposition based method." Biomedical Signal Processing and Control 71 (2022): 103098.
[18]Gonzales-Martínez, Rosa, et al. "Hyperparameters Tuning of Faster R-CNN Deep Learning Transfer for Persistent Object Detection in Radar Images." IEEE Latin America Transactions 20.4 (2022): 677-685.
[19] Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition. CoRR abs/1409.1556 (2014).
[20] Sandler, M., Howard, A., Zhu, M., Zhmoginov, A. & Chen, L.-C. Mobilenetv2: Inverted residuals and linear bottlenecks. CVPR(2018).
[21] Szegedy, C. et al. Going deeper with convolutions. CoRR abs/1409.4842 (2014).
[22] He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. CoRR abs/1512.03385 (2015).
[23] Xie, S., Girshick, R., Dollar, P., Tu, Z. & He, K. Aggregated residual transformations for deep neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1492–1500 (2017).
[24] Hu, J., Shen, L. & Sun, G. Squeeze-and-excitation networks. CoRR abs/1709.01507 (2017).
[25] Burgos-Artizzu, Xavier P., et al. "Evaluation of deep convolutional neural networks for automatic classification of common maternal fetal ultrasound planes." Scientific Reports 10.1 (2020): 1-12.