International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue 10 | Oct 2024
p-ISSN: 2395-0072
www.irjet.net
SMALL MEDICAL IMAGE IDENTIFICATION USING YOLOV8 ALGORITHM Shilpa.Chippalakatti1* Dr. C.H.Renumadhavi2* Dr. Andhe Pallavi3* 1Department of Electronics and Communication, Sir M Visvesvaraya Institute of Technology, Affiliated to
Visvesvaraya Technological University, Bengaluru, Karnataka
2Department of Electronics and Instrumentation, RV College of Engineering, Affiliated to Visvesvaraya
Technological University, Bengaluru, Karnataka
3Department of Artificial Intelligence and Machine Learning, RNS Institute of Technology, Affiliated to
Visvesvaraya Technological University, Bengaluru, Karnataka ---------------------------------------------------------------------***--------------------------------------------------------------------1.1 Background Abstract - Brain tumor disease refers to any Brain irregularity causing its damage. There are several kinds of Brain ailments. Benign growths are rarely life threatening and can be removed by specialists. Brain malignant tumor is one of the world's leading causes of cancer death. Identifying malignant growth tissue is a troublesome and tedious task. This research aims to optimize the latest YOLOv8 model to improve its detection of small Images and compare it with another different version of YOLO models. To achieve this goal, we used the classical deep learning algorithm YOLOv8 as a benchmark and made several improvements and optimizations. There is significantly less information and statistical analysis presented related to cholangiocarcinoma and hepatoblastoma. This research focuses on the image analysis of these two types of cancer. The framework's performance is evaluated using 2871 images, and a dual hybrid model is used to accomplish superb exactness. The first is the SID (Small Image Detection) dataset [11], a collection of images specifically curated and annotated for small Image detection tasks. Small Image detection aims to identify and highlight an image’s most visually distinctive Images or regions. The second one is the bacterial colony dataset [12], which is a collection of images specifically focused on bacterial colonies grown in a laboratory setting. The aftereffects of both neural networks are sent into the result prioritize that decides the most ideal choice for image arrangement. The technique entails extracting characteristics from the start and combining them with residual and pre-trained weights. This deep learning YOLOv8 system demonstrates the concept of illuminating elements of a pre-trained deep neural network's decision-making process by an examination of inner layers and the description of attributes that contribute to predictions.
Image detection is an essential task in the computer vision field, which is widely used in real-time video analysis, automatic driving, and intelligent security. Before 2014, traditional target detection algorithms required extracting features manually, which was time-consuming and unstable. The SOTA algorithm DPM [1] detector at that time, although inference speed was faster than others and could adapt to slight deformation, could not adjust to large-scale rotation and showed low robust ability. The dynamic shape method and deep learning YOLOv8 calculations involve recognizing inappropriate sources of info and furnish definite results with an identical viable allowance, such as recognizing a wide range of Brain problems in solitary handling. The topology of medical imaging methodologies is presented in Fig. 1. Hepatic carcinoma (HCC) is the subsequent reason for malignant growth deaths around the world, and it is the most normal essential cause of hepatocellular disease in individuals. Unlike other diseases, the incidence of HCC is rising [2]. YOLOv8 are useful in detecting and analyzing quickly and dependably the HCC in individuals, bringing about better results [3]. As the quality and accessibility of cross-sectional imaging have expanded, the requirement for intrusive demonstrative biopsies has diminished, driving imaging to a more fundamental position with an evident status, especially in essential Brain malignancies [4]. The Brain is likely the most often examined organ, and metastases are processes that aid in the detection, diagnosis, and management of hepatic disorders.
Key Words: Brain Tumor Detection, YOLOv8 Algorithm Deep Learning, Discrete Wavelet Transform (DWT).
1.INTRODUCTION Computerized image handling uses processing to deal with advanced images by using an advanced PC. Many issues emerge in the current framework, for example, giving countless faulty rates, over-segmenting tumor regions, high time complexity, low exactness, and it is a troublesome to deal with constructions of high inconsistency with a lot of disturbances that happen during division. © 2024, IRJET
|
Impact Factor value: 8.315
|
Fig -1: The architecture of YOLO consists of a backbone, neck, and head [11].
ISO 9001:2008 Certified Journal
|
Page 425