| Autor: |
Abed Malti, Malik Benmansour |
| Rok vydání: |
2020 |
| Předmět: |
|
| Zdroj: |
Advances in Intelligent Systems and Computing ISBN: 9783030366636 |
| DOI: |
10.1007/978-3-030-36664-3_37 |
| Popis: |
In this paper, we are interested in surgical skill evaluation. We present three artificial recurrent neural networks with a Long Short Term Memory (LSTM) architecture with the purpose of providing an objective assessment of surgeons who performed three basic surgery tasks on a workbench: Knot-Tying, Needle-Passing and Suturing. The tasks are represented by sequences of kinematic data recorded from the DaVinci surgical system. We train the three LSTMs with kinematic data of subjects with three different expertise levels: expert, intermediate and novice to which we associate three scores: 1, 0.7 and 0.4 respectively. These kinematic data were taken from JIGSAWS which is a free-to-use public dataset. We designed three LSTMs with the same architecture but each one focuses on assessing performance of surgeons on one of the three surgery tasks. We compare this approach with another one that uses a classic Recurrent Neural Network (RNN) architecture and also with an approach that uses a simple Deep Neural Network architecture. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
