Carpal Tunnel Syndrome (CTS) is a kind of compressive neuropathy that may develop from excessive use of a computer's keyboard and mouse. Early identification and prevention are essential in reducing the need for surgery and potentially saving lives. This study proposes a computer management system that utilizes deep learning to detect CTS through hand gestures during virtual keyboard and mouse operations. The system detects CTS risk by monitoring keyboard and mouse poses in real-time, using a Convolutional Neural Network (CNN) machine learning (ML) model. The model's accuracy in predicting posture is high, and its performance improves with increasing epochs. Accuracy on the training data is 98.72% and on the validation data, it is 98.83% for the keyboard's posture classification model at epoch 100. Likewise, at epoch 100, the accuracy of the mouse posture classification model is 98.97% on the training data and 99.08% on the validation data.