Description:Background of the Invention
In India, congenital heart defects (CHD) are one of the leading causes of new-born mortality and morbidity. According to the report of cardiological Society of India's, CHD is responsible for about 11% of infant mortality. CHD prevalence at birth has been found to range from 7 to 12 per 1100 live births (Kapoor & Gupta 2008). In India, Color Doppler, 2-Dimensional Ultrasonography and Echocardiography are being used as reliable tool for diagnosing prenatal congenital heart defects. During the fetus conception, the birth defect occurs in the form of functional and structural malfunction. Early diagnose and characterization of the CHD types has a significant impact on the life expectancy of newborn babies.
Congenital heart disease (CHD) is the most frequent of all congenital anomalies by organ system. However, the prenatal diagnosis of congenital heart anomalies is difficult and poor due to the complex structure of the organ and its small size. Yet, the prenatal diagnosis of CHD improves the likelihood of survival and reduces morbidity. Ultrasound examination of the fetus is the only method of screening for CHD prenatally. Pregnant women and their families expect that their unborn child will be evaluated to ensure that it is normal, and prenatal diagnosis of congenital anomalies has become an integral part of prenatal care.
In order to distinguish anatomical features such as fetal heart chambers and blood arteries, the use of US imaging in CHD screening necessitate exclusive exposure. The prenatal CHD scanning is a complicated task due to continuous movements of the fetus and thin wall chambers of the fetus heart. As a result, cardiac screening has emerged as a frontier topic in prenatal screening. The main disadvantage of using Ultrasound Screening is the presence of speckle noise which is inherent in the US images. This becomes a critical task for the sonographers to interpret the fine details of the features of the heart. Due to the above factors, the US modality fails to capture the boundary regions associated with the fetus. According to Rychik, the new gynecologists and sonographers find it hard to infer the several planes of the US modality and it is extremely difficult to infer the accurate details by an image plane.
Objects of the invention
This invention is designed with a robust despeckling method to remove the speckle noise inherent in the ultrasound images using the Enhanced Perona Malik Filter and Gaussian Filtering for image smoothening (EPMGF). In order to delineate the 2D ultrasound images a novel segmentation method using K-means Clustering algorithm (KmCA) is proposed. To achieve immaculate segmented results few morphological operations such as Erosion, closing, Compliment, Thickening, Thinning and Clean are added with KmcA, Finally, a computationally intelligent classifier method based on Efficient Multilayer Deep Detection Perceptron (EMDDP) classifier is used to accurately classify the normal and abnormal US images. A final Graphical User Interface module is designed congenital heart diseases contrived with robust EPMFAGF, KmCA and MLDDP. This system is designed with less cost and doctors from remote place can access the details with less time.

Summary of the Invention
From the above reports it is stated that, there is a great need for the development of an automated tool to help care professionals in the identification of prenatal CHD’s. Obviously, this circumstance has a greater influence on developing a novel Computational Intelligence Technique’s with different modules of pre-processing, segmentation and classification to detect the prenatal CHD’s from 2D ultrasound images in order to enhance clinical diagnostic outcomes. Thus this design will definitely act as a helping tool for the new sonographers and the cases where they are in need of fine-tuning of images.
Existing Prior Art/Events

PATENT NUMER ASSIGNEE TITLE PUBLICATION DATE
EP2249707B1 Amir A. Sepehri
Arash Gharehbaghi
Device for automated screening of congenital heart diseases 13.3.2014
US20160262633A1 HarborUcla Device and method for screening congenital heart disease 15.9.2016
US11172890B2
Masimo Corp Automated condition screening and detection 16.11.2021
US10588518B2
Masimo Corp Congenital heart disease monitor 17.03.2020
Brief Description of the Drawings
FIG 1 shows an integrated modules of different image processing and deep learning techniques for the automatic detection of CHD’s.
Fig 2 is an example of GUI model which can be useful for diagnosis of disease precisely.
The block diagram shown in figure 1 explains the various steps involved in diagnosing the CHD’s. 101 describes the loading of fetal heart ultrasound images. Extracting the four chamber view (4CV) ultrasound images 102, then crop the region of interest 103. As it is a medical image more amount of speckle noises will be there and hence a pre-processing method is needed. Enhanced Perona malik filter is identified as it preserves very fine details like edges, contour and other important details of an image 104. Gaussian filtering is also added for better smoothening of an image. In order to segment the sonographic biomarkers of the fetal heart, K-means clustering technique is used 105. Diagnosing the segmentation result 106. Different computational intelligent technique unit 107. Feature extraction using sammon mapping technique 108. To classify the abnormal image from the normal image Multi layered deep detection perceptron 109. Outcome unit 110.

Figure 2 shows the graphical user interface model. 201 outer layer of the GUI module. 202 moniter for visualization. 203 preprocessing unit. 203 segmentaion unit, followed by segmentation we have feature extraction unit 204 and 205 edge detection block. To classify the given image as normal or abnormal using classification unit 206. Unit to save the result 207. To load the image 208. To start the function 210 and to logout from the model 209.

Detailed Description of the Invention
The fetal heart has thin wall chambers, which is one of the key reasons for developing a novel automatic detection to identify the CHD from ultrasound images. Thus, the process of designing and implementing an automatic detection for ultrasound image analysis is still a difficult task. Thus, there exist a demand to develop a computerized image analysis approach for detecting fetal cardiac structures from 2D 4CV ultrasound images, which would help the gynecologists to enhance the rate of prenatal CHD diagnosis. This proposed automatic detection model is the first system implemented to predict any kind of CHD from the ultrasound images. This proposed system is the first attempt designed to help the radiologist, sonographologists and gynecologists to interpret the diagnostic details from the ultrasound images and to take fine decision about the absence or presence of CHD’s.
Preprocessing the image with EPMGF based filtering is the first step in the diagnosis procedure. When compared with the existing filters, the robust EPMGF despeckling approach outperforms well. Image segmentation module combines the robust pre-processing methodology and segmentation technique based on K-means clustering approach to highlight the sonographic biomarker of the PVSD CHD. The simulated results proved that the proposed segmentation algorithm is predominantly suitable in delineating the ultrasound fetal RVLV chambers and to interpret the pathological markers of the US images. Classifying the US images as normal or abnormal is an important step in this research study. The segmented output is taken as input for the classifier. The proposed MLDDP classifier worked extensively in classifying the disease that is PVSD. This is the first pioneering method to quantify the features of the image automatically and to diagnose the fetal heart asymmetric appearance. This diagnostic classification accuracy level is quite high for predicting the images which contains anomaly
, C , Claims:I/We Claim,
1. an automatic prenatal congenital heart disease detection method comprising of anisotropic diffusion module, K-means clustering module and MLDDP module for diagnosing whether the US image is detected with congenital Heart Disease; this method will act as a secondary tool for the sonographers, radiologist and gynecologist for detecting either structural or functional disease of the heart prenatally.
2. The method claimed in 1 make use of advance deep learning classifiers to automatically detect the heart features to increase the accuracy rate.