Detection of Downsyndrome in Fetus Images Using Deep Learning Algorithms

Down Syndrome (DS) detection process is carried out during the first trimester. The normal fetus and the DS-affected fetus did not differ significantly, according to the author's findings. The impact of nasal bone length in the second trimester stages for an effective detection of DS is the focus of the researchers' attention because of this limitation. Hence, this research work proposes and develops the automatic fetus image classification and detection system which uses deep learning algorithms. This proposed system contains pre-processing and deep learning algorithm modules for obtaining the fetus image classification results. The pre-processing module is used to improve the quality of the fetus image using histogram enhancement process. Then, the enhanced fetus image is now given to deep learning module which uses LeNet and Alexnet algorithm to perform the classification process. The final output is either healthy fetus or abnormal fetus.