An HMM framework for optimal sensor selection with applications to BSN sensor glove design

Laparoscopic surgical training is a challenging task due to the complexity of instrument control and demand on manual dexterity and hand-eye coordination. Currently, training and assessing surgeons for their laparoscopic skills rely mainly on subjective assessment. This paper presents a Body Sensor Network (BSN) sensor glove for laparoscopic gesture recognition and objective assessment of surgical skills. An HMM framework is proposed for the selection of sensors to maximize the sensitivity and specificity of gesture recognition for a given set of laparoscopic tasks. With the proposed framework, the optimal location as well as the number of the sensors can be determined. The sensors used in this study include accelerometers and fiber optic bend sensors. Experimental data is collected by participants wearing the glove while performing simple laparoscopic tasks. By using the proposed HMM framework, sensor correlation and relevance to task recognition can be determined, thus allowing a reduction in the number of sensors used. Results have shown that it is possible to establish the intrinsic correlation of the sensors and determine which sensors are most relevant to specific gestures based on the proposed method.

[1]  Warren D. Smith,et al.  Ergonomic problems associated with laparoscopic surgery , 1999, Surgical Endoscopy.

[2]  Keith Worden,et al.  Optimal sensor placement for fault detection , 2001 .

[3]  A. Darzi,et al.  Assessing operative skill , 1999, BMJ.

[4]  Parvati Dev,et al.  The fundamental manipulations of surgery: a structured vocabulary for designing surgical curricula and simulators. , 2004, The Journal of the American Association of Gynecologic Laparoscopists.

[5]  A. Gallagher,et al.  Objective Psychomotor Skills Assessment of Experienced, Junior, and Novice Laparoscopists with Virtual Reality , 2001, World Journal of Surgery.

[6]  C. Giebmeyer,et al.  Ergonomic aspects of five different types of laparoscopic instrument handles under dynamic conditions with respect to specific laparoscopic tasks: An electromyographic-based study , 2004, Surgical Endoscopy And Other Interventional Techniques.

[7]  A. E. Park,et al.  The effect of using laparoscopic instruments on muscle activation patterns during minimally invasive surgical training procedures , 2003, Surgical Endoscopy And Other Interventional Techniques.

[8]  Blake Hannaford,et al.  Markov modeling of minimally invasive surgery based on tool/tissue interaction and force/torque signatures for evaluating surgical skills , 2001, IEEE Transactions on Biomedical Engineering.

[9]  Bernhard Schölkopf,et al.  Support vector channel selection in BCI , 2004, IEEE Transactions on Biomedical Engineering.

[10]  H. Bubb,et al.  Laparoscopic surgery and ergonomics: It’s time to think of ourselves as well , 2003, Surgical Endoscopy And Other Interventional Techniques.

[11]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[12]  Mário A. T. Figueiredo,et al.  Similarity-based classification of sequences using hidden Markov models , 2004, Pattern Recognit..