Towards near real-time assessment of surgical skills: A comparison of feature extraction techniques

•Feature extraction techniques play an important role in an automated surgical skill assessment system.•We proposed a framework to rigorously compare the performance of different feature extraction techniques on automated surgical skill assessment in short time interval manner.•A comparative analysis was carried out on nine well-known feature extraction techniques.•The CNN deep learning technique outperforms all other techniques with an overall accuracy of 96.84, 92.75 and 95.36% for suturing, knot tying and needle passing, respectively.