Bone Fracture Detection Using Morphological and Comparing the Accuracy with Genetic Algorithm Y. Rakesh

DOI: 10.47750/pnr.2022.13.S03.030
INTRODUCTION 
 
The use of X-ray diagnostic imaging in the identification of bone fractures inside the human body is critical. 
With image segmentation, medical practitioners can make better decisions and manage injuries more effectively. 
Medical x-ray image processing is used to further examine the recorded digital images in terms of improving 
diagnosis outcomes. Image morphology can be defined as a theory for analysis of spatial structures in opening 
and closed images (Kevin Zhou, Greenspan, and Shen 2017). Using morphological technique reduces noise, 
improves image segmentation, and emphasizes the fracture area. Because of the combined effect of the 
morphological technique and the canny edge recognition algorithm, the fracture edges are even more vividly 
exposed (Kalita et al. 2018). Applications of morphological algorithms are Morphological smoothing One way 
to achieve smoothing is to perform a morphological opening followed by a closing. Smoothing of the 
morphology Smoothing may be achieved by performing a morphological opening followed by a morphological 
closure gradient. In the source image segmentation (Bhattacharjee et al. 2020), the morphological gradient 
reveals strong gray-level changes (Chandra Shekhar Rao and Sammulal 2021). 
The existing system has 20,300 conference papers in Google Scholar and 248 IEEE journal papers in the current 
system. Previously, the majority of pattern object identification was done by deep learning and x-ray image 
analysis (Bharodiya and Gonsai 2019). This research focuses on identifying whether the iliopectineal line is 
interrupted, a key condition for automating the acetabular fracture categorization process and providing 
orthopedic surgeons with computer-assisted diagnosis (Damien et al. 2019). X-ray image enhancement 
techniques, morphology algorithm to extract features, and scan line algorithm to determine the site of a bone 
fracture (Muchtar et al. 2018). Because recognising apparent fractures including such open and cracked 
fractures is straightforward, even for a non-medical observer, the proposed approach focuses on discovering 
demanding fractures that are easily neglected (Hržić et al. 2019). Several anterior fracture classification systems, 
along with the Hansen, Jensen, Boyd-Griffin, Kyle, and AO/OTA classifications, have been devised and utilized 
in clinical practice for decades (Li et al. 2019). The creation and testing of fully automated deep learning 
techniques that are taught to detect anomalies on non-contrast head CT and X-ray images that require immediate 
treatment (Chilamkurthy et al. 2018). Nodules recognition, identification, extraction, and noninterpretive tasks 
are the four areas of deep learning breakthroughs in musculoskeletal radiology (Chea and Mandell 2020)Our