Shoulder Implant Manufacturer Detection by Using Deep Learning: Proposed Channel Selection Layer

Total Shoulder Arthroplasty (TSA) is the process of replacing damaged ball and socket joint in shoulder with a prosthesis made polyethylene metal components. After this procedure, intervention may be required as result damage to prosthesis, except for need an examination regarding at certain periods. If patient does not have information about model manufacturer treatment delayed. Artificial intelligence-assisted systems can speed up by classifying prosthesis. In study, artificial intelligence methods were applied classify implants using X-ray images. The detected proposed deep learning method. Besides, most commonly used machine classifiers same problem compare results show effectiveness addition, accuracy precision analysis measurements processing times performed reveal performance, accuracy, efficiency study. order measure performance method, it was compared studies on literature. As comparison, found that rate 97.2%, better than other studies. implant are classified carry out surgery best way model. With success system, applicability similar classifications has been shown. Differently from literature, channel selection formula presented method recommended selecting distinctive feature filters.