Authenticating Biomedical Images Using Water Marking Technique In Machine Learning

Abstract

Pneumonia is one of the most prevalent illnesses that kills newlinechildren and the elderly around the world, particularly in developing newlinenations. Approximately 7% of the world s population suffers from newlinepneumonia, which kills over 700,000 young people annually. The air newlinemembranes may expand with fluid or pus as a result of pneumonia. A newlineradiologist will frequently use chest X-ray images to identify the newlinecondition. Physicians must be guided through the diagnosis procedure newlineusing computer-aided diagnostic tools. A concerning number of digital newlinepictures are being sent over computer networks at once. A watermark can newlinebe added to a medical image to help authenticate it and guarantee its newlineintegrity. The accuracy of diagnosis needs to be increased immediately. newlineThe preservation and safeguarding of the patient s life makes security the newlinemedical industry s top priority. The primary goal of this study is to newlinedevelop a new digital image processing method that incorporates security newlinefeatures to accurately identify pneumonia disease in its early stages. The newlineproposed approach, called the enhanced Dynamic Learning Neural newlineNetwork (eDLNN), combines the traditional DLNN algorithm with the newlineSupport Vector Classification (SVC) algorithm. This method has been newlineshown to be effective in detecting pneumonia disease at an early stage, newlinebut it is crucial to perform security checks during the testing phase to newlineverify the accuracy of the results. Additionally, the hospital logo is newlineappropriately watermarked on the relevant testing images. The image has newlinebeen altered. If not, the suggested method forgoes processing the image. newlinevi newlineThese security measures highlight the medical sector and raise the bar newlineeven higher. With such technology, patients can receive proper, error-free newlinecare as well. The system processes an appropriate Kaggle dataset newlineconsisting of 5,856 Chest X-Ray images categorized into two groups: newlinePneumonia and Normal. They have a positive impact on the medical field newlineprotection system in terms of processing these photographs, accurately newlinediagnosing pneumonia