The Role of Posture, Magnification, and Grip Force on Microscopic Accuracy BASIL SAFWAT, 1 EILEEN L. M. SU, 1 ROGER GASSERT, 1 CHEE LEONG TEO, 2 and ETIENNE BURDET 1 1 Department of Bioengineering, Imperial College London, South Kensington Campus, SW7 2AZ London, UK; and 2 Department of Mechanical Engineering, National University of Singapore, 119260 Singapore, Singapore (Received 12 June 2008; accepted 26 February 2009; published online 11 March 2009) Abstract—While tremor has been studied extensively, the investigations thus far do not give detailed information on how the accuracy necessary for micromanipulations is affected while performing tasks in microsurgery and the life sciences. This paper systematically studies the effects of visual feedback, posture and grip force on the trial error and tremor intensity of subjects holding a forceps-like object to perform a pointing task. Results indicate that: (i) Arm support improves accuracy in tasks requiring fine manipu- lation and reduces tremor intensity in the 2–8 Hz region, but hand support does not provide the same effect; hence freedom of wrist movement can be retained without a significant increase in trial error. (ii) Magnification of up to 910 is critical to carry out accurate micromanipulations, but beyond that level, magnification is not the most important factor. (iii) While an appropriate grip force must be learned in order to grasp micro-objects, such as a needle, without damaging them, the level of grip force applied does not affect the endpoint accuracy. Keywords—Microscopy, Manipulation, Accuracy, Posture, Magnification, Grip force, Microsurgery. INTRODUCTION Microsurgery and Fine Motor Control The execution of goal-directed movements by humans involves the combined use of several factors including visual input, 5,29 proprioceptive feedback, 9,29 and motor control strategies. 12 Some factors specifi- cally affect the accuracy of fine motor tasks such as microsurgery or other manipulations carried out under the microscope (e.g., in the life sciences). Knowledge of how these limiting factors interact and to what extent they each affect overall performance will help suggest strategies on how to maximize performance in micro- manipulation. One factor, easily addressed, is the limit of human vision. Surgeons performing manipulations on small blood vessels or nerves will often use a microscope or other magnifying device in an attempt to increase accuracy. 10,22 Indeed, some of the objects operated on would be indistinguishable without magnification. The range of magnifications used begins at 92.5 to 98.0 (for surgical loupes) and increases to 940 (for oper- ating microscopes). Another simple method is to adopt a posture with appropriate support to increase control and accuracy. 1 Other factors limiting the accuracy are less easy to address. These include tremor, 10 delays in sensory- motor pathways, 27 and inaccurate interpretations of sensory information. 30 Physiological Tremor Physiological tremor is the name given to the appar- ently involuntary, approximately rhythmical oscilla- tions present in the motion of a limb. These oscillations result from both peripheral and central origins. 7 The frequency band of physiological tremor is often quoted as lying at 8–12 Hz, although this figure does not take into account the mechanical properties of the body part where the frequency is being measured, which may have an altering effect. Because of this, peaks are reported in the regions around 2–4, 8–12, 20–25 Hz, and occasion- ally 40 Hz. 13,28 Much previous work has centered on investigating the effect of peripheral loading on finger tremor; in these studies, the mechanics of a limb is changed (e.g., by the use of weights) and the effect of this change on the tremor and electromyographic (EMG) recordings is explored. 28 In the current study, the effect of posture, grip force and visual feedback on tremor recordings was investigated. Address correspondence to Etienne Burdet, Department of Bio- engineering, Imperial College London, South Kensington Campus, SW7 2AZ London, UK. Electronic mail: bsafwat@gmail.com, e.su06@imperial.ac.uk, gassertr@ethz.ch, e.burdet@imperial.ac.uk The experiments were carried out at National University of Singapore and at Imperial College London. Annals of Biomedical Engineering, Vol. 37, No. 5, May 2009 (Ó 2009) pp. 997–1006 DOI: 10.1007/s10439-009-9664-7 0090-6964/09/0500-0997/0 Ó 2009 Biomedical Engineering Society 997