IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. BME-29, NO. 7, JULY 1982 Measurement of Eye Movement with a Ferromagnetic Contact Ring YEHOSHUA Y. ZEEVI, MEMBER, IEEE, AND JEHUDA ISH-SHALOM Abstract-A new magnetic method for measurement of eye move- ment is presented. It is based on measurement of differential induc- tance variations of two C-shaped coils due to the movement of a conical ferromagnetic ring which fits on the sclera. High sensitivity over a wide dynamic range is obtained by using a tuned bridge and synchronous de- tector. Separation of the detector coils from the magnetic field source improves the resolution and stability, and enables a two-dimensional measurement with one pair of C-coils. INTRODUCTION IN THE fields of vision research and clinical ophthalmology, improvements of the current methods of measuring eye movements are needed in order to make more precise and con- venient measurements. For the whole spectrum of movements to be monitored,. a resolution of a few seconds of arc over a range of about 800 of the visual field is required [11 , i.e., the system has to span a measurement range of some five decades [2]. This is a rather difficult task in any mechanical or elec- trical measurement and is particularly formidable given the physiological constraints. Of the wide variety of methods proposed in the past and im- plemented in most reported studies, each has its advantages and limitations [2], [3], but none covers the entire dynamic range. Some of them, such as Robinson's electromagnetic search coil technique [4] -[6] and Cornsweet's double Purkinje technique [7], [8], come close to the required specifications, but are still cumbersome, expensive, and unsuitable for scaling down to portable size. We have therefore developed a new magnetic method which is based on measurement of magnetic field variations due to movement of a conical ferromagnetic ring fitted on the sclera. The principle of the transducer operation is presented first, stressing the physical constraints and their implications. Then, Manuscript received July 22, 1981; revised November 30, 1981. This work was supported in part by a grant from the United States-Israel Binational Science Foundation (BSF), Jerusalem, Israel, and in part by a fund for the encouragement of research at the Technion-Israel In- stitute of Technology, Haifa, Israel (050-351). Preparation of the manuscript was supported by the Man-Vehicle Laboratory, Massachu- setts Institute of Technology, Cambridge, MA 02139. This paper was presented in part at the 10th Annual IEEE Convention, Tel Aviv, Israel, October 1977. Y. Y. Zeevi is with the Man-Vehicle Laboratory, Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA 02139, and the Division of Applied Sciences, Harvard University, Cambridge, MA 02138, on leave from the Department of Electrical Engineering, Technion-Israel Institute of Technology, Haifa, Israel. J. Ish-Shalom is with the Man-Vehicle Laboratory, Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA 02139. the characteristics obtained thus far and a theoretical analysis of the transducer including comparison with experiments are given. Finally, a discussion of possible modifications, which may yet further improve the transducer characteristics, is presented. THE TRANSDUCER Principle of Operation The idea is to fit a conical ferromagnetic ring (thickness 0.1- 0.25 mm, approximate weight 0.1 g) on the sclera and to track the eye's position by measuring the ring's position. The ferro- magnetic ring is designed to be embedded in a silicon rubber suction ring (adopted from Collewijn [6]) to fit on the sclera around the iris as shown in Fig. 1. The ring position can be detected by measurement of varia- tions in a magnetic field, using a coil situated near the eye. In our configuration, variations of magnetic field are measured through changes in the inductance of a coil. Furthermore, we use two symmetrical coils wound on two ferromagnetic C-shaped cores (Fig. 2) which form a magnetic circuit with the ring. Changes in inductance may be thought of as being caused by variations in the magnetic resistances of the air gaps; inductance increases as movement of the ring shortens the magnetic pathway in the air, and vice versa. Since the C-coils are situated next to the eye so that the eyelid separates the coil ends from the ring, there is a wide unavoidable air gap in the magnetic pathway (approximately 4 mm). Because of the size of this air gap, the maximum rela- tive change in inductance is as small as 2 percent. Thus, in order to achieve a resolution of 0.010, relative variations of inductance of the order of 10-6 have to be measured. This rather difficult task is achieved by means of a tuned bridge and a synchronous detector (Figs. 2 and 3). Measurement Setup In principle, one C-coil would suffice, but to obtain the required sensitivity, one needs a differential measurement employing two C-coils. As a differential circuit, it improves linearity and reduces vertical-to-horizontal cross-axial coupling. It also reduces the effect of ambient temperature changes, of extraneous magnetic fields, and of moving ferromagnetic mate- rial in the vicinity of the transducer. As the nng moves in one direction, there is an increase in the inductance of one of the coils-say of LA by ALA -and a corresponding decrease in LB. For small movements, we can approximate ALA - -ALB AL. 0018-9294/82/0700-0511$00.75 1982 IEEE (1) 51 1