Search

Background and objectives Semen analysis is a primary and mandatory procedure to evaluate the infertility during clinical examination. This procedure includes the analysis and classification of normal and abnormal Sperm, selection and efficient tracking of healthy sperm in the sample. Many methods were proposed earlier for the analysis of semen. The fast sperm movement and high dense cluster of sperm is a challenging task for researchers. Methods The paper proposes a novel Faster Region Convolutional Neural Network (FRCNN) with Elliptic Scanning Algorithm (ESA) for classifying human sperm and a Novel Tail to Head movement algorithm (THMA) for the motility analysis and tracking. This proposed method improves the accuracy of computer assisted semen analysis (CASA). Results The proposed method outperforms and provides better results than existing methods. Method provides better accuracy of 97.37%. Sperm detection and identifying the sperm motility in the group is performed with minimum execution time of 1.12 s. Conclusions A novel FRCNN with ESA detection algorithm is proposed for the analysis of human sperm classification. This method provides an accuracy of 97.37%. A Tail head movement-based (THMA) algorithm is explained for the motility analysis.

Karyotyping of Banded Metaphase Chromosomes is one of the preliminary steps used in cytogenetics to analyze the chromosomes for diagnostic purposes. Deep learning is a subfield of machine learning concerned with structure and function of brain. It exploits a way to automate predictive analysis. The key aspect of deep learning is that the layers of features are not designed by human engineers. They are learned from data using a general purpose learning procedure. This paper proposes a convolution based deep learning to classify the chromosomes for automated karyotyping. The developed architecture allows us to train and test images that helps in predicting the chromosome abnormality. The performance analysis is based on loss and accuracy curves and the graphical representation clearly exhibits better classification results for this architecture.