In this paper, we introduce a full-reference quality assessment model for laparoscopic videos. The perceived quality of medical imaging in general is of upmost importance, as visual degradations may lead to severe negative impacts on diagnostic accuracy. Laparoscopy, also known as keyhole surgery, is a camera-assisted operation performed in the abdomen or the pelvis of the patient. Unlike the conventional utilization of telescopic rod lens systems, digital laparoscopy uses a miniature digital camera at the end of the laparoscope, and therefore, the surgeon fully relies on the quality of the medical video. In our scientific contribution, we utilize different image quality measures for each frame of the laparoscopic videos. We implement a regression neural network architecture on the frame-level features with the associated mean opinion score as labels. Finally, we calculate the average of the predicted frame-level scores to compute the overall quality score. The performance of the proposed model is evaluated on the well-known LVQ laparoscopic video dataset. The evaluation results confirm that our model is competitive with the state-of-the-art 2D full-reference and no-reference supervised algorithms. Furthermore, the model demonstrates robust performance across all distortion types of the dataset.