Abstract

A proper temporal model is essential to analysis tasks involving sequential data. In computer-assisted surgical training, which is the focus of this study, obtaining accurate temporal models is a key step towards automated skill-rating. Conventional learning approaches can have only limited success in this domain due to insufficient amount of data with accurate labels. We propose a novel formulation termed Relative Hidden Markov Model and develop algorithms for obtaining a solution under this formulation. The method requires only relative ranking between input pairs, which are readily available from training sessions in the target application, hence alleviating the requirement on data labeling. The proposed algorithm learns a model from the training data so that the attribute under consideration is linked to the likelihood of the input, hence supporting comparing new sequences. For evaluation, synthetic data are first used to assess the performance of the approach, and then we experiment with real videos from a widely-adopted surgical training platform. Experimental results suggest that the proposed approach provides a promising solution to video-based motion skill evaluation. To further illustrate the potential of generalizing the method to other applications of temporal analysis, we also report experiments on using our model on speech-based emotion recognition.

Original languageEnglish (US)
Article number6915721
Pages (from-to)1206-1218
Number of pages13
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume37
Issue number6
DOIs
StatePublished - Jun 1 2015

Fingerprint

Hidden Markov models
Markov Model
Motion
Evaluation
Emotion Recognition
Formulation
Synthetic Data
Model
Labeling
Experiment
Labels
Likelihood
Ranking
Experiments
Attribute
Target
Training
Skills
Requirements
Experimental Results

Keywords

  • emotion recognition
  • Relative hidden markov model
  • relative learning
  • surgical skill
  • temporal model

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Software
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

Relative hidden Markov models for video-based evaluation of motion skills in surgical training. / Zhang, Qiang; Li, Baoxin.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 37, No. 6, 6915721, 01.06.2015, p. 1206-1218.

Research output: Contribution to journalArticle

@article{699a356ae37548428cb222ca8ee3ccaf,
title = "Relative hidden Markov models for video-based evaluation of motion skills in surgical training",
abstract = "A proper temporal model is essential to analysis tasks involving sequential data. In computer-assisted surgical training, which is the focus of this study, obtaining accurate temporal models is a key step towards automated skill-rating. Conventional learning approaches can have only limited success in this domain due to insufficient amount of data with accurate labels. We propose a novel formulation termed Relative Hidden Markov Model and develop algorithms for obtaining a solution under this formulation. The method requires only relative ranking between input pairs, which are readily available from training sessions in the target application, hence alleviating the requirement on data labeling. The proposed algorithm learns a model from the training data so that the attribute under consideration is linked to the likelihood of the input, hence supporting comparing new sequences. For evaluation, synthetic data are first used to assess the performance of the approach, and then we experiment with real videos from a widely-adopted surgical training platform. Experimental results suggest that the proposed approach provides a promising solution to video-based motion skill evaluation. To further illustrate the potential of generalizing the method to other applications of temporal analysis, we also report experiments on using our model on speech-based emotion recognition.",
keywords = "emotion recognition, Relative hidden markov model, relative learning, surgical skill, temporal model",
author = "Qiang Zhang and Baoxin Li",
year = "2015",
month = "6",
day = "1",
doi = "10.1109/TPAMI.2014.2361121",
language = "English (US)",
volume = "37",
pages = "1206--1218",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "6",

}

TY - JOUR

T1 - Relative hidden Markov models for video-based evaluation of motion skills in surgical training

AU - Zhang, Qiang

AU - Li, Baoxin

PY - 2015/6/1

Y1 - 2015/6/1

N2 - A proper temporal model is essential to analysis tasks involving sequential data. In computer-assisted surgical training, which is the focus of this study, obtaining accurate temporal models is a key step towards automated skill-rating. Conventional learning approaches can have only limited success in this domain due to insufficient amount of data with accurate labels. We propose a novel formulation termed Relative Hidden Markov Model and develop algorithms for obtaining a solution under this formulation. The method requires only relative ranking between input pairs, which are readily available from training sessions in the target application, hence alleviating the requirement on data labeling. The proposed algorithm learns a model from the training data so that the attribute under consideration is linked to the likelihood of the input, hence supporting comparing new sequences. For evaluation, synthetic data are first used to assess the performance of the approach, and then we experiment with real videos from a widely-adopted surgical training platform. Experimental results suggest that the proposed approach provides a promising solution to video-based motion skill evaluation. To further illustrate the potential of generalizing the method to other applications of temporal analysis, we also report experiments on using our model on speech-based emotion recognition.

AB - A proper temporal model is essential to analysis tasks involving sequential data. In computer-assisted surgical training, which is the focus of this study, obtaining accurate temporal models is a key step towards automated skill-rating. Conventional learning approaches can have only limited success in this domain due to insufficient amount of data with accurate labels. We propose a novel formulation termed Relative Hidden Markov Model and develop algorithms for obtaining a solution under this formulation. The method requires only relative ranking between input pairs, which are readily available from training sessions in the target application, hence alleviating the requirement on data labeling. The proposed algorithm learns a model from the training data so that the attribute under consideration is linked to the likelihood of the input, hence supporting comparing new sequences. For evaluation, synthetic data are first used to assess the performance of the approach, and then we experiment with real videos from a widely-adopted surgical training platform. Experimental results suggest that the proposed approach provides a promising solution to video-based motion skill evaluation. To further illustrate the potential of generalizing the method to other applications of temporal analysis, we also report experiments on using our model on speech-based emotion recognition.

KW - emotion recognition

KW - Relative hidden markov model

KW - relative learning

KW - surgical skill

KW - temporal model

UR - http://www.scopus.com/inward/record.url?scp=84929192781&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929192781&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2014.2361121

DO - 10.1109/TPAMI.2014.2361121

M3 - Article

C2 - 26357343

AN - SCOPUS:84929192781

VL - 37

SP - 1206

EP - 1218

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 6

M1 - 6915721

ER -