This article reports on the preliminary findings of a pilot study on evaluating fish fin damage using a convolutional neural network. Keeping track of fish health is a challenging task, with one critical aspect being the assessment of fish fin. To achieve this objective, video data of pikeperch fish in an aquarium was collected, followed by image extraction and annotation. The annotated dataset was then used to train a CNN for evaluations. Results from the modelling matrices revealed that improving the dataset and adjusting certain parameters can help achieve exceptionally significant precision, recall, and F1 scores. These findings underline the potential of deep learning methods in automating aquaculture health assessment.