Analysis of Blood Smear Microscopic Images Using ML: DL

Emerging AI technology intersects with several techniques simulating human cognitive processes that rely on creating algorithms that run in a dynamic computing environment. It is essential to put in place adequate tools of human intelligence. In this context, Machine Learning (ML) is becoming a real revolution that has greatly contributed to the generalization of AI. In principle, ML refers to a set of many methods aimed at automatically creating models from data. It allows computers to learn without being explicitly programmed to do so. Currently, ML occupies a growing interest in various field, such as: autonomous cars, smart cities, automatic checkouts, personal assistants, algorithmic finance, industrial robots, video games, medical diagnostics, scanners, etc. However, Our work focuses on the development of various machine learning techniques in application control especially medical process applications. The main objective is to implement a method for analyzing microscopic images of blood cells to fulfill tasks and facilitate human activity. More fundamentally, we developed a learning model to accelerate the process of diagnosing and detecting diseases with an accuracy rate that could exceed that of humans. Such analysis can be time-consuming and generate incorrect results for large-scale diagnostics. The process contains two essential phases: identification and classification of blood cells for the diagnosis and the prediction of certain abnormalities. First, we use convolution neural network CNN to extract blood image features and ML methods such as KNN, SVM, Bayes, for classification. Therefore, ML can help improve the performance, predictability, and accuracy of diagnostic systems for many diseases.