The subject will be inside the MRI chamber to the waist with the wrist module normally located just inside the scanner bore, dependent on the size of the test subject. Photo courtesy of the National Rehabilitation Hospital, Washington DC.

Transformation of a Load Cell

Using load cell sensors in conjunction with fMRI (functional Magnetic Resonance Imaging) to study the muscle-brain connection of patients who have experienced stroke or head trauma is a new and promising breakthrough for studying rehabilitation techniques. However, to accomplish this, very precise and accurate measurements must be conducted to see how the different therapies are – or are not – working.

The main problem is that the patient is asked to apply a variety of loads that need to be replicated at different times during the course of treatment. Changes in the patient’s response over time are used to gage progress of the therapy.

In the past, the therapist would ask the patient to apply loads to the therapist’s hand or arm. Then the therapist would jot down notes describing the level of response. This is not a precise way to measure variable loading, and week to week variations are not easily recognized.

Instead, a multi-axis load cell is the proper measuring tool to use; however, all standard load cell sensors and fMRI systems fit together a little worse than a lighted match and gasoline – this is not a natural technology paring.

Load cells are typically made out of ferrous materials. This is a natural disaster near any type of MRI equipment. If a standard load cell was placed near an MRI, it would literally fly directly into the magnetic field. Even the slightest level of magnetic material near an MRI can be extremely dangerous.

Because of this, medical developers first experimented with silicon strain gages for this type of measurement. In theory, these should have worked well, but in actual application, the signals from this type of sensor configuration were too distorted. The clarity and accuracy of standard sensors was needed.

The challenge was then on to see if high accuracy load cells currently in use in other applications could be modified for use in an fMRI environment. That’s when science turned to JR3. We have specialized in many customized load cell designs for a wide range of unusual applications, including many for extreme environments for NASA projects–but this was a totally new challenge.

Aside from the ferrous materials, the second challenge was to overcome the serious noise generated by the strong MRI magnetic fields. We had developed a basic design for load cells that were used in high noise, high interference conditions in some industrial plants such as for special arc welding lines were the sensor is actually mounted on the welding head of a robot. This was a good starting point.

Standard 6 axis load cells from which the magnetic compatible versions were developed. Photo courtesy of JR3, Inc.

By the nature of the application, these sensor configurations are immune to significant noise issues, but even this noise immunity had to be enhanced if the sensor measurement was to be efficient in an MRI environment.

To do this we tightly twisted the internal wiring from each strain gage and minimized the loops in the harness within the sensor to avoid “antennas” injecting noise into the Wheatstone bridge circuits.

The next step was that all potentially magnetic material had to be removed from the sensor. Even materials that are normally considered non-magnetic, such as certain types of stainless steel, had to be removed. Steel fasteners, bolts, nuts and screws were replaced with brass and in a few instances titanium or a bronze type material. The type of material used depended on the specific location and the strength required, bronze or titanium used where extreme strength is needed.

From an actual specification standpoint, the high level capability of the load cells remained the same as their industrial cousins, just the packaging and materials changed. Making these changes, all standard model load cells can be made MRI compatible.

However, even the tiny connector at the end of a cable that interfaced the load cells within the MRI area had to be replaced. There had to be some searching to find totally non-magnetic replacements for all packaging and connection elements.

Subjects lightly grasp the plastic handle extending from the load cell and exert wrist flexion-extension or ulnar-radial deviation forces against it. Arrows denote the direction of the applied forces for wrist flexion or extension. The forearm is supported by four padded bumpers which helps minimize forces from propagating up the arm necessitating contraction of proximal arm muscles. The plastic wedge is mounted to a plastic base which the subject lays on to anchor the device. Photo courtesy of the National Rehabilitation Hospital, Washington DC

Also, normal production could not be used. Even in a relatively clean production environment, minute amounts of metal dust were too much of a risk. Every precaution had to be in place to prevent any bits or chips of steel dust to get onto or into the sensor, as they would be brought into the magnet. Even if they were too small to do any physical harm to personnel, patients or to the fMRI equipment, they would distort the image.

Therefore, special procedures normally used for clean-room strain gage mounting only had to be expanded to encompass the needs of a building a totally non-magnetic load cell.

A System Example

One system that is currently in use incorporates a set of specially designed filters that allow subjects to be directly attached to the load cell and transmit mechanical signals and brain data to a remote computer. This tracks the effectiveness of therapy by comparing unimpaired and impaired subjects before and after intervention therapies.

The fMRI system can register torque and movement of upper limbs by using a "joy stick" device or other peripheral device. These systems delve into the mysteries of how the brain controls muscle activity and subsequent movement. The goal is to better understand how the brain recovers from injuries such as stroke and traumatic brain injury, and to see how it responds to various therapeutic interventions.

There is not only a great magnetic field when generating images within an MRI, but there are also RF (radio frequency) pulses being generated as well as rapidly changing orthogonal magnetic fields. The radio frequency pulses caused problems with the signal picked up from the load cell. Some analog filters were already built into the load cells so in addition to those, a second stage of filters were added. The filter cut off frequency was reset to a much lower frequency. The second stage used elliptical filters to give faster roll off of the cut-off frequency.

A representative scan of the motor cortex during a wrist flexion behavioral task located at 20% maximum exertion level.

On these load cells the electronics to amplify the signal are usually built in. But, for this particular application, the electronics were moved to an external box kept in a separate room housing the computers that control the MRI, not within the MRI chamber with the load cell.

The signals from the load cell are generated in analog, filtered by the custom box and then translated for analog to digital conversion of the signal for display on the computer screen. Using this filtration segmenting, when a subject generated a shoulder or elbow torque against the load cell, the signal was nearly as clean as those in the experiments outside of the fMRI environment using standard sensors.

Researchers can look directly at brain activity while patients are generating isometric contractions at their shoulder or elbow.

Concluding Thoughts

The concept of incorporating load cells for measuring limb control was used first for basic comparison of muscle torque and movement in standard rehabilitation situations using EEG technology. The idea of joining the data from non-ferrous load cells to the real time data imaging of fMRI scanning for complete detail of how the brain was reacting offered a substantial breakthrough in this area.

The concept of using load cell sensors in combination with fMRI is gaining wide acceptance. Currently, similar systems are being replicated at major medical facilities worldwide for rehabilitation tracking. The availability of non-magnetic load cell sensors, combined with fMRI technology and linked via computer is opening doors for further development of systems for other medical areas.

For more information visit www.jr3.com.