Prediction of Eye Injuries in Sports by Artificial Intelligence
Oral Presentation
Paper ID : 2104-SSRC
Authors
چشم یار
Abstract
Eye injuries are a significant concern in various sports, leading to potential vision loss and long-term complications. Early detection and prevention are crucial, yet current methods often rely on subjective evaluations and lack objectivity. This paper explores the potential of artificial intelligence (AI) in predicting eye injuries within the sporting context.
We propose a novel machine learning framework leveraging real-time data acquired during game play. The framework incorporates: Based on the collection of images and videos taken with a smartphone, the degree of sports injuries can be recognized and described online by Image Processing and Deep learning based on hybrid models. On the other hand, based on the information entered in advance, it is possible to arrange the effects over time in the form of time series for sports and athletes. The model utilizes advanced machine learning algorithms, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), to process these multimodal data streams and predict the likelihood of eye injury in real-time. Therefore, a set of labeled images for sports events such as Retinal detachment, Keratitis and Corneal ulcer can be trained in hybrid models. Next, based on the new images, the condition of the athlete's eye can be classified according to the mentioned category of the mentioned eye diseases with high accuracy.
We evaluate the proposed framework on a curated dataset of sporting events with labeled eye injuries. We demonstrate the effectiveness of the approach in achieving high prediction accuracy compared to traditional methods. This research holds significant promise for enhancing athlete safety and reducing the incidence of eye injuries in sports.
We propose a novel machine learning framework leveraging real-time data acquired during game play. The framework incorporates: Based on the collection of images and videos taken with a smartphone, the degree of sports injuries can be recognized and described online by Image Processing and Deep learning based on hybrid models. On the other hand, based on the information entered in advance, it is possible to arrange the effects over time in the form of time series for sports and athletes. The model utilizes advanced machine learning algorithms, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), to process these multimodal data streams and predict the likelihood of eye injury in real-time. Therefore, a set of labeled images for sports events such as Retinal detachment, Keratitis and Corneal ulcer can be trained in hybrid models. Next, based on the new images, the condition of the athlete's eye can be classified according to the mentioned category of the mentioned eye diseases with high accuracy.
We evaluate the proposed framework on a curated dataset of sporting events with labeled eye injuries. We demonstrate the effectiveness of the approach in achieving high prediction accuracy compared to traditional methods. This research holds significant promise for enhancing athlete safety and reducing the incidence of eye injuries in sports.
Keywords