• Shubham Patil
• Sourish Vaghulade

Radiologists have the troublesome job of diagnosing various injuries through MRI and CT scans. This job is subject to errors which can prove costly to a patient's health as well as the radiologist's reputation. This is gravated by the fact that often times these MRIs are noisy and distorted. An aid in the form of a machine for this job will definitely be beneficial. In this paper, we deal with denoising such images and detecting the detecting the injury grade related to Anterior Cruciate Ligament (ACL). We try to examine the possibility of predicting diagnosis based on the detection of the ACL after clearing the initial scan. We begin this process by filtering out the noise from our input images using a DnCnn followed by various detection models such as HOG + SVM, GIST + SVM, HOG + RBF K-Means, GIST + RBF K-Means, HOG + SOM, HOG + Random Forest, GIST + Random Forest, and Autoencoder. We compare these approaches and pick out the one which is most efficient in terms of accuracy. Clever machines will make workers more productive more often than they will replace them.

Magnetic resonance imaging (MRI),Ligament, Computer-aided detection and diagnosis, Machine learning, Pattern recognition and classification, Feature extraction, Denosing Image

"Efficient Image Denoising and Semi-Automated Detection Of ACL Injury From MRI Scans", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 1, page no.347-357, January-2019, Available :http://www.jetir.org/papers/JETIR1901143.pdf

"Efficient Image Denoising and Semi-Automated Detection Of ACL Injury From MRI Scans", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 1, page no. pp347-357, January-2019, Available at : http://www.jetir.org/papers/JETIR1901143.pdf