International Research Journal of Engineering and Technology (IRJET) Volume: 11 Issue: 04 | Apr 2024
e-ISSN: 2395-0056 p-ISSN: 2395-0072
www.irjet.net
Blood Group Detection Using Image Processing and Deep Learning Jashwanth Sai Ganta
Dr. Mohana Roopa Y
Mary Rishitha
Jaya Surya Pulivarthi
Department of Computer Department of Computer Department of Computer science and Engineering science and Engineering science and Engineering Institute of Aeronautical Institute of Aeronautical Institute of Aeronautical Engineering Hyderabad, Engineering Hyderabad, Engineering Hyderabad, Telangana Telangana Telangana ----------------------------------------------------------------------***-----------------------------------------------------------------------
Department Of Computer science and Engineering Institute of Aeronautical Engineering Hyderabad, Telangana
ABSTRACT— Blood group identification is an essential part of medical diagnostics and transfusion treatment procedures. Our novel approach to blood type detection is presented in this work using deep learning and image processing techniques. Orientated FAST, rotated BRIEF (ORB), and scale-invariant feature transform (SIFT) algorithms are used in our method for feature extraction, while convolutional neural networks (CNNs) are used for blood group photo classification. The image processing pipeline optimizes contrast and minimizes noise in blood group photos using SIFT and ORB before extracting discriminative features. Then, utilizing these traits, a CNN model is trained to precisely categorize blood types. As the CNN model is designed to recognize and comprehend the unique patterns associated with different blood types, it might offer dependable and strong prediction. Using extensive trials on a variety of blood group imaging datasets, we evaluate our method. The results demonstrate the effectiveness of our method for identifying blood types, with outstanding classification accuracy and resistance to variations in image quality. Our recommended approach could make blood group identification in medical settings easier and allow for rapid, automated blood sample analysis. Additionally, merging deep learning with feature extraction techniques enhances the effectiveness and precision of blood type prediction, hence improving transfusion management and patient care. Keywords—Transfusion therapy, medical diagnostics, Scale-invariant feature transform (SIFT), Orientated FAST and Rotated BRIEF (ORB) algorithms, Feature extraction, Convolutional neural networks (CNN), Preprocessing, Contrast, Discriminative features, Resilience to changes, Automated blood sample analysis, Patient care, Transfusion management.
I. INTRODUCTION Deep learning and image processing are two examples of cutting-edge technologies that have revolutionized the area of blood group identification and offer a practical means of enhancing and automating blood type detection. In conventional blood group identification © 2024, IRJET
|
Impact Factor value: 8.226
|
procedures, manually evaluating and interpreting blood samples is a time-consuming and prone to human error procedure [1]. By combining deep learning and image processing techniques, a novel approach has been implemented to address these challenges, improving transfusion management and patient care while automating blood type diagnosis [6]. The recommended method uses convolutional neural networks (CNNs) for blood group photo categorization in conjunction with feature extraction utilizing orientated FAST, rotated BRIEF (ORB), and scaleinvariant feature transform (SIFT) methods [7]. This approach uses an image processing pipeline to preprocess blood group images to improve contrast and minimize noise before using the SIFT and ORB algorithms to extract discriminative features. Blood type classification that is accurate and reliable is made possible by the CNN model that is trained using these variables [11]. The goal of this research is to enhance transfusion management and patient care by speeding up and accurately analyzing blood sample analysis and improving automated blood type identification [11]. The proposed method combines deep learning with feature extraction techniques in an effort to increase the precision and efficacy of blood type prediction in medical settings. The investigations, which employ a wide range of blood group imaging datasets, show how adaptable and accurate the system is at classifying images, even when picture quality varies [13]. Deep learning methods combined with image processing technology for blood type determination have many potential applications in the medical industry. Through automation and increased accuracy in blood type detection, this technology can accelerate blood transfusion processes and perhaps improve patient outcomes [6]. In order to overcome the shortcomings of manual blood group identification processes, CNNs are coupled with feature extraction algorithms to provide a more reliable and effective solution.[9] The proposed technique may preprocess blood group images, extract discriminative features, and train a CNN model, which makes blood group detection comprehensive and creative. The focus on improving blood type prediction efficiency and accuracy through extensive testing on many blood group imaging datasets highlights the method's ISO 9001:2008 Certified Journal
|
Page 97