There have been significant breakthroughs in developing highly functional myoelectric prostheses, yet individuals who have experienced upper-limb loss consistently report low levels of satisfaction when performing daily tasks while using myoelectric prostheses. This research aims to evaluate the change in the user’s experience after completing a training program in which tasks are designed to facilitate adaptation to the myoelectric arm in performing the activities of daily living. One participant with a left transradial limb difference was recruited for this project. The user’s experience was evaluated by comparing task completion time, trunk and shoulder angles, object control, movement smoothness, vertical ground reaction forces, the center of pressure location, and selected muscle activation of her affected arm between baseline and post-training. The data collection was performed through a motion capture system, a pressure mat, and wireless EMG modules. While indications of potential improvements in balance, muscle efficiency, and functionality were present, the data were inconclusive as to the effectiveness of the training procedure. One outcome measure that showed improvement across most tasks was the task completion time, which, on average, for the targeted Box and Block Test (tBBT) task was reduced by 24.0 s, box lifting by 1.6 s, bottle pouring by 9.0 s, and the pulley task by 8.4 s. This project serves as part of a larger multi-visit study evaluating the effects of home-based functional training on facilitating users’ adaptation to the myoelectric arm.

In this work, a pre-trained convolutional neural network is employed to detect gunfire from audio excerpts of urban sounds. The pretrained convolutional neural network is fined-tuned with transfer learned features to a new task using a smaller number of training signals. Two CNN methods are applied to the time-frequency representation of the audio signals. The first CNN method is based on classifying specific events in audio signals. The second CNN method is an image-based analysis method. The accuracy of the two CNN results will be compared and analyzed based on gunfire type and retrieved urban multipath conditions. A k-means clustering algorithm is employed to identify gunfire types and parametric modeling to retrieve the urban multipath conditions.

In this work, a pre-trained convolutional neural network is employed to detect gunfire from audio excerpts of urban sounds. The pretrained convolutional neural