SanStart/MUHAS_Elective_study

This is repository for my elective study at Muhimbili University of Health and Allied Sciences in Tanzania titled "Application of the State-of-the Art Artificial Intelligence Algorithms in Detecting Cervical Pre-Cancerous Lesions by Using Public Available Pap Smear Cytological Images" # MUHAS Elective Study: "Application of the State-of-the-Art Artificial Intelligence Algorithms in Detecting Cervical Pre-Cancerous Lesions by Using Public Available Pap Smear Cytological Images" Welcome to the GitHub repository for my elective study conducted at the Muhimbili University of Health and Allied Sciences (MUHAS) in Tanzania. This study focuses on the "Application of the State-of-the-Art Artificial Intelligence Algorithms in Detecting Cervical Pre-Cancerous Lesions by Using Publicly Available Pap Smear Cytological Images." ## Overview Cervical cancer remains a significant public health issue, especially in under-resourced settings. Early detection through screening can significantly reduce the morbidity and mortality associated with this disease. This project explores the potential of advanced artificial intelligence (AI) algorithms to improve the accuracy and efficiency of detecting pre-cancerous lesions in cervical cells using cytological images from Pap smears. ## Objectives - To evaluate the performance of various state-of-the-art AI algorithms in the classification of Pap smear images. - To assess the potential of these AI models to assist pathologists and reduce the workload in resource-limited settings. - To contribute to the global efforts in cervical cancer screening and prevention by providing an open-source tool for researchers and clinicians. ## Dataset The study utilizes publicly available datasets of Pap smear cytological images. The specific dataset used includes: CRIC Searchable Image Database ## Methodology The study employs several advanced AI models, including but not limited to convolutional neural networks (CNNs) and deep learning frameworks, to analyze and classify cytological images. The methodology section details the preprocessing steps, model architectures, training processes, and evaluation metrics used in the study. ## Results This section presents the findings of the study, including model performance metrics such …