Comparative Study of Principle and Independent Component Analysis of CNN for Embryo Stage and Fertility Classification

Comparative Study of Principle and Independent Component Analysis of CNN for Embryo Stage and Fertility Classification

Anurag Sinha, Tannisha Kundu, Kshitiz Sinha
Copyright: © 2022 |Pages: 28
DOI: 10.4018/IJFSA.296594
(Individual Articles)
No Current Special Offers


background: Applications of deep learning for the societal issues are one of the debatable concerns where the community medicine and implication of artificial intelligence for the societal issues are a big concern. This article, it is shown the applications of neural networks in clinical practice for reproduction procedure enhancement. And this is a well-known issue where image analysis has the exact applications. In Embryology, fetal abnormality early-stage detection and diagnosis is one of the challenging tasks and thus, needs automation in the process of tomography and ultrasonic imaging. Also, Interpretation and accuracy in the medical imaging process are very important for accurate results.
Article Preview

1. Introduction

With the rise in the scope of the artificial intelligence, it is being implemented in every film for model accuracy and estimation. In healthcare, artificial intelligence is being used as providing the many different solutions, deep learning and medical image processing is the part of and sub domain of artificial intelligence which is being implemented in health care sector for the disease diagnosis, medical imaging, disease prediction and clinical image analysis. Medical image processing is one of the significant and crucial tasks in terms of its accuracy and precision of its result.(Ahmad et al., 2018) In Embryology the early stage, abnormality detection is very complex as it has the many different kind of internal complexities and different kind of paradigms associated with the abnormalities. Which is associated based upon the different kind of physical and genetic parameters. Thus, medical imaging and the classification of the embryo based on its stage and cycles and identification of the abnormality becomes very essential to safeguard the Mothers and the child life. Perceiving the reasonability in the embryo and collecting and classification the images based on the different features is a very crucial task and vulnerable to mistake and different kind of unusual Nets. In terms of image classification, as the Image classification task based upon the image features and in depth feature extractions, this is totally dependent upon the deep learning algorithm.

In the conventional methods of the health care for the medical image analysis and classification is specifically when it comes to ultrasound and tomography imaging. The accuracy and interpretation of the system is a very challenging task because it lacks the automation system and algorithm for the accurate results in image classification of the. Clinical analysis and sometimes this result the symptoms become same and disease becomes different, because of the inaccurate report and process. Thus, to overcome this issue we are proposing the novel algorithm which is based upon the multilayer perceptron (MLP) for the image classification of the embryo for the early stage. (Altuntaş et al., 2019b)Abnormality detection and diagnosis based upon the input based upon the input data of the embryo and those data are the collected image set of the menstrual cycle of the Embryology which is nothing but all the fertility cycles which occurs during the 9th month of pregnancy. This image data set is collected from the different clinical sources and has been pre-processed based on the image pre-processing technique to remove the noise and blur in the data set. We have also used the encoding be on the Python environment for pre-processing and removing the filtration of the noise for the better productivity of the module.In this paper, we are proposing the novel algorithm based upon multilayer perceptron, which is one of the finite algorithm of the deep learning and which is known for its multi layer of the convolution which has a feed forward neural network kind of structure with which understands the features and forward them for the learning process in a. Very accurate and. Dense. Process which makes the model accurate in terms of its finiteness and feature extractions. To overcome the issue of the incorrectness in the result of and report of the medical imaging clinical analysis we have proposed this model, also overcomes the same issue by automating the process of tomography and ultrasound imaging. For the optimization and accuracy of the model and data set, we have optimized the data and the algorithm by using the independent and principal component analysis process of the statistics and machine learning for optimization and reducing the dimensionality of the data and reducing the unnecessary global functions of the algorithm. So, in this process we have analyzed the data and its feature levels of the data set based upon the univariate and multivariate component analysis which signifies the. Principal data labels which have the maximum participation in the model, accuracy and productiveness, and thus by reducing the dimensions of the data. It makes the data more segregated and it becomes easy to explore. Thus the data become optimized and reduced by the irrelevant noise and redundancies.

Complete Article List

Search this Journal:
Volume 13: 1 Issue (2024)
Volume 12: 1 Issue (2023)
Volume 11: 4 Issues (2022)
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing