Examination of the sensor placement in the swallowing mensuration using Magneto-Impedance sensor

K. Hamaguchi¹ , M. Akutagawa², Y. Tegawa³, T. Ichikawa³, Y. Kinouchi²

¹ Graduate School of Advanced Technology and Science, The Univ. of Tokushima

² Institute of Technology and Science ,The Univ. of Tokushima

³ Institute of Health Biosciences ,The Univ. of Tokushima

Introduction

The objective of this study is to develop a screening device for the swallowing difficulty (dysphagia) using a magneto-impedance (MI) sensor. The video fluorography (VF)  or the video endoscopy are used to diagnose dysphagia. In the other hand, RSST (repetitive saliva swallowing test)¹⁾ is widely used as a simple test in some facilities where they have no VF apparatus.

The authors have developed a device to automate the RSST using a MI sensor. Availability of proposed method is examined in previous study²⁾. As results of experiments, movement of a laryngeal prominence could be measured using a small magnet and the MI sensor. But it was difficult to measure the swallowing movement under particular situations. Movement of body is one of such situations because the sensor measures variation of geomagnetism simultaneously with movement of the laryngeal prominence. In previous study, we confirmed that use of three-dimensional magnetic sensors was effective to reduce influence of body movement. In this study, we examine arrangement of the sensor and the magnet using simple simulation.

Method

The proposed device consists of a neodymium magnet, a MI sensor (Aichi Micro Intelligent, AMI302) and a personal computer for data acquisition. The MI sensor is capable to measure three -dimensional magnetic field. The magnet is attached on the laryngeal prominence, and the MI sensor is attached on breastbone. Relative position between the magnet and the sensor changes in the subject's swallowing. The magnetic field at the sensor reflects the swallowing.

 

 

In RSST, the subject is instructed to swallow saliva as fast as possible in 30 seconds. If the subject can swallow less than 3 times, he/she is suggested to need further investigation. In proposed system, basic procedure is same as original RSST except 10 seconds preceding measurement to obtain offset of the sensor output (Fig. 2). The offset is subtracted from measured values.

 

The change of  sensor output is described as

 

       

where V_x, V_y, V_z are sensor output of each axis and V_ox, V_oy, V_oz are offset of them.  V_ox, V_oy, V_oz  are calculated as average during offset period. They consist of geomagnetism and sensor offset.

 

Results

It is possible to count the swallowing movements from both Fig. 3 (a) and (b). Especially, Fig. 3 (b) summarizes the variance of 3 magnetic sensors. It is convenient to count using threshold.

 

Fig.4 shows a result of another subject of healthy adult (23 years old, male). In this subject, waveforms during the swallowing is blurred. It is difficult to apply the threshold to count the swallowing.

 

Difference between two subjects is size of the laryngeal prominence. The movement of the first subject was visually distinguishable. In other hand, that of the second subject was in small range. It  was also unclear by looking. This result means that the movement range of the laryngeal prominence has large individuality, and  it is necessary to place the magnet and the sensor so that the variance of magnetic field at sensor is as large as possible.

 

Discussion

As a result of experiment, it is important to chose proper placement of the magnet and the sensor. Further limitation of the placement is distance between them. The magnetic field at the sensor inversely proportional to the cube of distance. This means that the range of magnetic field near the magnet is too large to have sufficient precision of A/D converter. In the other hand, it is too small to measure the magnetic field when the distance is so large. For example, the magnetic field is smaller than the geomagnetism in the experimental system described above when the distance is further than about 5cm to 8cm. The appropriate distance is thought as that range. Another factor is orientation of the magnet. Strength of magnetic field is twice larger along the magnetic axis in comparison with border of the magnetic poles (Fig. 5). Figure 5 represents the change of magnetic field around the constant distance from a magnet (saturation magnetic field: 1T, height 6mm, diameter 7mm). The magnet is approximated as a magnetic dipole for simply calculation. The magnet at the  laryngeal prominence shows not only parallel movement but also rotating. The most appropriate angle to detect the rotation of magnet is 45 and 135 degrees because the slope of the curve is maximum.

 

 

Conclusion

An experimental device for automated RSST using a magnet and magnetic sensor is described. We confirm that the proposed device can measure the movement of the laryngeal prominence. However,  it is difficult to count the number of swallowing for some subjects with small laryngeal prominence. According to calculation of magnetic field distribution, optimal position of the sensor is 45 degrees away from the magnetic axis of the magnet. The next work is to measure the swallowing movement using derived arrangement.

 

References

1.  K.Oguchi, E.Saitoh, M.Mizuno, M.Baba, M.Okui, M.Suzuki, The Repetitive Saliva Swallowing Test (RSST) as a Screening Test of Functional Dysphagia (1) Normal Values of RSST'', The Japanese Journal of Rehabilitation Medicine Vol.37, 375-382, 2000

2.  K. Tanida, M. Akutagawa, Y. Kinouchi, T. Ichikawa and S. Hongama, Evaluation of Swallowing Movement by Posture Change Using a Magneto-impedance Sensor, JJ Mag. Dent., Vol.18, No.2, 15-19, 2009