Vanderbilt University



Vanderbilt University

School of Engineering

Department of Biomedical Engineering

Multisensory Environment

for

Neurophysiological Monitoring

Team Members:

Gregory Apker

Vern Huang

Nazriq Lamien

Andrew Lin

Emma Sirajudin

Advisor:

Daniel Polley, Ph.D.

Mark Wallace, Ph.D.

Date of Submission:

April 24, 2007

Abstract:

Dr. Polley and Dr. Wallace are trying to explore the behavioral significance of physiological plasticity during awake recordings. However, they require a refinement of the sensory stimuli to improve its accuracy in location and time to more closely simulate a naturalistic environment, eliciting more realistic responses. Thus, the goal of our project is to create the hardware and software necessary to make these improved multisensory environments. This will include improving or updating rat backpack hardware, a wireless auditory transmitter-receiver system, and force plate calibration while also involving ground up development of a somatosensory stimulator. The software consists of three parts, which are visual, auditory and somatosensory; each of these parts must be able to collect and sort incoming data. Software requirements also included developing Pre-Pulse Inhibition software to be used in conjunction with the force plate. The development of the wireless auditory transmitter-receiver produced an experiment-ready model using Bluetooth technologies. The somatosensory stimulator system developed used a piezoelectric bending actuator with an 800μm deflection and accounted for and incorporated all of the necessary system parameters required for the design of a somatosensory tactile stimulator. In order to determine frequency ranges of hearing, an analysis of Forward Masking data was required. Tuning curves, the visual representation of frequency thresholds, are complex enough to have to be delineated subjectively by a person. The sensory software developed was also made into a flowchart diagram and serves as a guide to the researchers in the lab so that they can understand how the software works a little better. Another purpose of this analysis is to guide other programmers if they are interested to create software similar to this.

1. Introduction:

Many studies have shown the existence of large-scale plasticity in the visual, somatosensory, and auditory cortices of the brain. In addition, other research has focused on achieving a better grasp of multisensory interactions.

However, these areas of neurophysiological monitoring have a great deal of room for improvement. It is important to conduct these studies because they drive our understanding of pathologies and perception.

The labs of Dr. Polley and Dr. Wallace are trying to address the limitations of these past studies by adding new functionality to their multisensory environments. Awake recording capabilities will allow the exploration of the behavioral significance of physiological plasticity, compared to its simple characterization through anaesthetized recordings. Furthermore, the refinement of sensory stimuli through its accuracy in location and time will more closely simulate a naturalistic environment, eliciting more realistic responses.

The goal of our project is to create the tools, involving both hardware and software, necessary to make these improved multisensory environments. The development of multisensory environment hardware for use in neuronal response characterization in rats and cats involves several requirements. The hardware needed to produce visual, auditory, and somatosensory stimulation, which can be modulated in location and intensity and can be integrated to receive as well as deliver information. The development of environment-specific software for closed loop control of the environment also entails a few requirements. The software needed to allow user-defined stimulation parameters, initiate coordinated sensory stimulation, and record, associate, and organize the system output.

Our goals were outlined as such:

• Update and repair of existing hardware components

• Increase usability/durability of system components

• Install environment hardware and architecture

• Implementation of wireless transmitter-receiver system

• Manage frequency range, spectral fidelity, and channel crosstalk impact

• Ground-up development of somatosensory stimulator system

• Design stimuli delivery and mounting apparatus

• Offer precision and versatility in application without confounding artifacts

• Integrate designed components

• Modification of existing in-lab software

• Increase functionality of Pre-Pulse Inhibition protocol

• Convert software from Visual Basic to Matlab

• Development of visual and auditory software

• Control delivery of sensory stimuli in time

• Collect, sort, and analyze data

2. Methods:

2.1 Wireless transmitter-receiver

Two wireless transmitter-receiver systems were under consideration for use in the multisensory environment experiments. One system was a custom PCB design, which was quite expensive and still in prototype form, while the other system was a Bluetooth solution. An analysis of the wireless Bluetooth transmitter-receiver system involved a focus on frequency range, spectral fidelity, and the impact of crosstalk between the right and left channels of the earphones.

2.2 Ear-speaker interface

Working with live rats in the multisensory environment, ear molds needed to be produced in order to make well-fitted earbuds for the semi-permanent attachment of Etymotic 6i Isolator™ earphones to the ears of rats. Two materials, Cerebond Skull Fixture Adhesive and Jet Denture Repair mold, were analyzed with respect to their functionality as ear molds using the following characteristics: their density and weight, the strength of their bond to metal, the weakness of their bond to tissue, their malleability, and the ability to create a canal in them.

In the analysis of their density and weight, molds of identical volume were created by filling tubing that was 0.5 inches in length with the Cerebond and Jet materials. The density and weight of the materials were important for the future experiments on the rats because it would be preferable that the ear buds do not hinder the rats or make them uncomfortable to a significant degree.

To test the other characteristics, actual ear molds using rat ears were made using the Cerebond and Jet materials. This process involved burning two holes in the ears, placing two small screws in the holes, and letting the molding material set to the shape of the ear. Combined with washers, the screws were to be used to hold the earbud in place during live experiments.

2.3 Rat backpack hardware

Investigation into the successful combination of the component parts involved in the rat backpack hardware system was conducted through the perfection of the design of certain parts to prevent potential problems and to ensure its overall proper function in future experimental applications. First of all, research into braided electrical shielding material was conducted in order to obtain material that would protect the earphones from electromagnetic interference (EMI) produced by the surrounding environment. After the appropriate electrical shielding product was found, ordered, and delivered, designs for the electrical shielding of the rat backpack hardware system were illustrated.

The electrical shielding comprised three sections: (1) the rat backpack, (2) the combined earphone wires, and (3) the separated earphone wires. The shielding for sections (2) and (3) was accomplished with a combination of the braided expandable shielding material and cable ties. The electrical shielding for section (3) did not adequately cover the transducer in the earphone. Therefore, efforts to cover the transducer portion of the earphones with a thin sheet of copper were pursued. A copper sheet in the form of a cross with a hole in the middle was designed to wrap the earphone, effectively covering the transducer. The dimensions of the cross involved a 10 mm strip, an 8 mm strip, and a 3 mm hole. Once the copper shielding was folded over the earphone, super glue was applied to the edges to hold it in place. Since the braided electrical shielding rested over the copper shielding, a lightweight surgical tape was wrapped around the earphone to keep all of the stray ends together and to ensure contact between the two types of electrical shielding.

2.4 Simulation of hearing attenuation

An alternative solution to physically stitching shut the ear canal in rats in order to limit their hearing ability during early development was needed. Tight-fitting ear plugs were a potential solution that was explored. The length, the top diameter, and the bottom diameter were adjustable variables in the process of creating model ear plugs that fitted easily and snugly into the rat’s ear. Alterations to E-A-R ear plugs provided decent models. An assessment of the various ear plug designs was necessary to determine the effectiveness of “deafening” by judging the seal provided in the ear and finding the acceptable amount of material fitted in the ear canal among other things. The final design had a cone-like shape with a length of 12 mm, a top diameter of 8 mm, and a bottom diameter of 2 mm.

2.5 Force plate for PPI

In order to determine a relationship between a change in force and a change in voltage for the force plate transducer, a pulley system needed to be devised. The pulley system needed to meet the following characteristics: manual operation, minimization of friction, and uniform lifting of the mass. Materials were purchased in order to create a model of the pulley system: two different sizes of eye screws, a block of wood, and fishing line. Using a power drill, a pilot hole was made in the block of wood to insert the eye screw. After testing with masses up to 200 g attached to the fishing line, it was determined that the smaller eye screw is capable of supporting the required weights and creates more uniform lifting of the mass from a surface compared to the larger eye screw.

This model was approved and needed to be implemented into the actual multisensory environment. The force plate transducer inside the sound chamber was attached to a square piece of metal, which was secured with four screws at the corners. However, with the existing hardware in place, the pulley system was not able to be employed over the middle of the force plate. In order to achieve consistent calibration for future experiments, the location of the force plate needed to be standardized. Rotation and translation of the metal square were possible solutions that were explored. The location of the force plate was standardized to a position where the square piece of metal extended 1.5 cm past the edge of the opening of the sound chamber. Once the location of the force plate was standardized, an eye screw was affixed to the ceiling of the sound chamber, which already contained some small holes. One of the holes needed to be widened in order to adequately secure the eye screw.

2.6 Tactile stimulator

Development of a tactile stimulator for somatosensory activation was dictated by a number of performance requirements. In all current experimental protocols, the feline subject is placed in a fixed, consistent position to prevent the confounding input of movement. However, the ability to activate the somatosensory system from a variety of positions requires that the system be placed and functional in a number of orientations. That is, the system must be as reliable if it rotated 90 or 180 degrees in any plane as it is un-rotated. Not unrelated to this need for the system to be versatile in position is the need for it to also be small and portable. Portable for the reason stated above, easily placed and moved to a number of locations around the test animal. It must be small for a number of reasons, the most important of which being that lab space is limited and space is a commodity which is to be preserved if at all possible. Also, if the system is too large, it limits where it can be placed and used because of the potential of producing confounding artifacts in the recorded data. If the system enters the animal’s visual field, or if it produces too broad of a tactile stimulation, then it will result in corrupted/non-specific neuron recordings. Also, the system must be silent. Any perceivable noise will generate auditory neuronal stimulation in the animal, contaminating the data. A stimulation that is too forceful could also produce misleading data from the unintended firing of different nerve endings with the skin. Therefore, the touching force generated by the system must contain all of the following attributes: Gentle, controllable, and accurate/consistent.

Another important consideration in designing the system is its integration into the protocol software. Because of the precise sequencing required during multi-sensory experimentation, the activation of the tactile stimulus must be able to be driven by an electric signal initiated by the software. Therefore, the system is constrained to something which can be automated, accepting electrical signal as an impetus for activation. Furthermore, because timing is such an important parameter of the stimulus activation, the final system must be adequately responsive to rapidly changing input signal.

From the automating software, the somatosensory stimulator system will receive a +/- 5 volts input voltage. It is expected that an in-line amplifier will result in a +/- 60 volt RMS signal reaching the device. A trapezoidal wave is used for the input signal to produce a tap-and-hold tactile stimulation. While a square-wave would also produce this effect, the gentler acceleration of a trapezoidal input will protect the equipment from excessive wear while also allowing for a more controlled motion of the stimulator system. To further increase these benefits of the trapezoidal wave, the edges of the wave form will be synthetically rounded before reaching the system.

2.7 Multisensory protocol software

The software is an important feature of the experiment as it is the only way for the user to communicate with the hardware. The main purpose of the software is to record the activity of the neurons when experiments are being held and have them analyzed. Data recording, time intervals, and input commands are just some of the functions of the software. The software is also where the user inputs parameters of the desired stimulus and have the results recorded.

The software consists of three parts, which are visual, auditory and somatosensory. All of these three stimuli have different ways of recording data and are created to perform the experiments while recording data efficiently. The software is also designed to decrease minor errors or information during experiments therefore every single function in the software is created so that no single detail is left out.

The visual part of the software has been tested and was proven to be working perfectly well. Firstly, the user will input parameters such as the location and the type of object to be projected onto the screen. When the different locations are projected under the user's time intervals, the brain activity of the animal being tested are recorded and are analyzed using the options provided by the software.

Auditory works in a similar manner as visual does, but uses a set of speakers instead of visual objects. The user will input the parameters such as the type of noise and the location of where the noise will be projected under a time interval and the brain activity of the animal will be recorded. The analysis of the brain activity of the animal will then be produced.

The somatosensory part of the software works in a different way as visual and auditory because the software communicates with a probe that is being controlled by the parameters put in by the user. The user puts in the amount of pressure needed and also the location of where the probe touches the animal and the software records the brain activity as the experiment is in progress.

2.8 Pre-Pulse Inhibition protocol software

The big idea in implementing awake recording for plasticity study is the ability to use animal behavior as a metric for response to stimuli. One of the more interesting modalities for this is Pre-Pulse Inhibition, which in simple terms means that a startling effect will be made less startling if a subject learns to recognize a regularly-timed cue preceding it.

In such a study with animals, the amount of ‘startle’ is gauged by a force plate beneath the subject. Upon being startled, there is a reflex to exert a sudden force downward, as muscles contract in response. The magnitude of this force is a reliable indicator of how startled the subject is.

2.9 PPI stimuli hardware

Before starting tests with PPI, a few glaring deficiencies were noted in the stimuli hardware. For one, the LED TTL driver was the only component on the rack powered by battery, and for no particular good reason. This was successfully treated by creating a battery-shaped adaptor to a 9-volt AC adaptor that plugged into the wall; the AC adaptor also had to have its wires spliced to attach the right male plug.

The second part needing help was the pneumatic system. Its connection to the driver was via small, fragile, easily disconnected leads, and there was an audible leak. This was fixed by using BNC adaptors for a direct BNC connection to the driver, and by checking the hose network for the leak problem.

3. Results:

3.1 Wireless transmitter-receiver

From the analysis transmitter-receiver systems in frequency range, spectral fidelity, and the impact of crosstalk between the right and left channels of the earphones, it was determined that the Bluetooth system had the appropriate capabilities to carry out future experiments in the rats at a fraction of the cost of the alternative system.

3.2 Ear-speaker interface

After weighing the molds of identical volume, it was determined that the average mold weight of five Jet material molds was 92.2 mg, while the average mold weight of five Cerebond material molds was 108.4 mg. This difference was not significant, so other characteristics of the materials were investigated.

After the ear molds were made, it was concluded that the Jet material was more malleable for the creation of a canal, conformed to the shape of the ear better, and gripped the screws more tightly than the Cerebond material. However, the analysis of these two materials was discarded when high-fidelity Etymotic ETY Plugs™ were discovered to provide the desired function of holding the earphones in place when sewn into the rats’ ears. Furthermore, the Etymotic earplugs were compatible with the Etymotic earphones, forming a tight seal for a high fidelity transmission of sound.

3.4 Simulation of hearing attenuation

In order to obtain a quantitative measure of the range of non-attenuated frequencies and the drop-off in high frequencies through a surgically enclosed rat ear canal to reach the ultimate goal of finding the connection between the presentation of certain sound frequencies and the corresponding responses of neurons in the brain, a method to transfer data from the PULSE LabShop software to Matlab for analysis needed to be discovered. The real-time FFT data could be captured at a point in time and exported from PULSE LabShop to Matlab as a .txt file. From the “Function Group.txt” file, data could be retrieved, converted to dB for the y-axis data, and plotted on a logarithmic scale on the x-axis. Since the basilar membrane in the cochlea functions on a log base 2 scale, the x-axis representing frequency was changed accordingly. With a standard curve created by noting the input speaker voltage necessary to achieve an 80 dB sound intensity in a normal rat ear across a range of frequencies from 500 Hz to 64 kHz, it was determined that there was an average intensity decrease of approximately 40 dB in a ligated rat ear and that the dropoff in high frequencies occurred between 11 kHz and 16 kHz.

3.6 Tactile stimulator

Initially, the solution to the task of designing the silent somatosensory stimulator appeared in the form of a voice coil because it offered silent operation. However, this solution ultimately failed in positional versatility and represented too high an initial cost.

The system that was actually developed incorporated a piezoelectric bending actuator to provide the tapping motion. The Physik Instrumente PL 127.11 bending actuator matched all our performance criteria except for providing too short of a maximum deflection, only attaining approximately an 800um throw in practice when a 1.5mm through was desired. Deflection length not being a non-starter, this model of piezo-actuator was selected and ordered. The manufacturer’s product specifications are provided in the table and are an accurate representation of the systems capacities. The size of the actuator it self measured as being only 31.0 x 9.6 x 0.65 mm but 3mm of the piezo length were lost to mounting, which itself added about two inches to the length but only a little additional width. The only motion was a gentle wagging (bending) of the actuator and produced no noise artifact. Depending on the input signal, the actuator is able to attain a range of speeds between 3mm/sec and 200mm/sec. While no measure of tactile force was taken, it is assumed that the actuator could be controlled adequately via the input signal so as to not produce a confounding artifact.

In order to achieve the maximum displacement, the actuator requires +/-60 volts be delivered. To achieve this signal amplification from the lab’s 5 volt system, the DSM VF-500 Linear Amplifier/Voltage Follower was chosen on the basis its ability to provide the necessary amplification and offer the option accepting our trapezoidal input or generating its own signal wave.

| |

A Flexbar® articulating arm was chosen for its versatility and range of positions and orientations it offered the system. To further increase the actuator’s range of motion, a predesigned piezo-holder developed by the Vanderbilt Neuroscience Workshop was order providing a complete freedom of orientation. The final system is shown in figure 1.

3.7 Multisensory protocol software

We have not developed a fully functional multisensory software yet as for now only the visual part of the software is functional where else the auditory and somatosensory is still under development. The auditory software has actually been created, but when it is being tested, it has some difficulties communicating with the hardware. Therefore, this "bug" still needs to be taken care of and is still in progress. The somatosensory software has yet to be started.

3.8 Pre-Pulse Inhibition protocol software

This protocol was realized as an RPVDS™ programmed visual circuit on the DSP, with an associated Matlab™ GUI to initialize trial parameters. The program on the

|[pic] |

|Figure 2: Data flow of the PC, DSP, and sound interface hardware. |

DSP triggers stimuli presentation to the animal chamber. Transducers in the animal chamber transmit data back to the DSP. The DSP then buffers and sends data to the PC at periods allotted by the programmed circuit. This operation is illustrated in Figure 2 at right.

A big concern in programming a rack-mount DSP is the upper limit on processor usage. It turns out that the biggest use of processor cycles in such a protocol is not live calculations or analysis, since there was none, but instead the cycles needed to buffer and send data to the PC. Thus, data sampled from the transducers had to be decimated to a reasonable rate which would still satisfy the Nyquist criteria for characteristic data. The data stores for force plate readings and electrophysiology were taken at 2400 Hz, in intermittent intervals, while other only-display data were sampled at 240 Hz. The GUI polls at only 10 Hz. This lets the four processors run at about 55% on average, good with respect to the usual 69% guideline.

3.10 Forward Masking analysis software

In order to determine frequency ranges of hearing, an analysis of Forward Masking data was required. Tuning curves, the visual representation of frequency thresholds, are complex enough to have to be delineated subjectively by a person. This analysis program was made in Matlab. Blurring and averaging options were programmed in as a visual aid to the operator. Based on the operator input, more statistics were formed to better delineate masking effects.

Economics/Market Potential

The only potential customers for the specialized hardware we develop are researchers within the same field, which are not particularly numerous.  Also, our priorities on the project are much more directed toward quick and effective functionality.  Aesthetics and cost-optimization concerns which produce more commercially marketable turnkey hardware will not be given much concern, comparatively.  We do not anticipate much market potential for the hardware, despite its uniqueness. 

On the other hand, a very flexible and scalable software suite might be useable for a wider audience of researchers.  If well done, it can save researchers time and reduce the amount or errors in data analysis.  There are, however, similar software products which are available open-source and free.   

In general, development was not very focused toward an appealing sellable product, because there isn't sufficient market potential to justify the time.  Specific concerns of the labs of Dr. Polley and Dr. Wallace drove most development decisions.\

Safety Concerns

Users of the end product of our project are not particularly prone to many safety concerns. Because this is electrical equipment, there is the risk of electric shock, but most of the components operate at low voltage and are shielded for relatively foolproof consumer use. The pneumatic somatosensory equipment operates at high pressure and thus is a risk to the user, especially when initiating flow from a standstill, but this is mitigated by the location of the hose splitters and solenoid valve network near the back of the equipment. The LED flash used for startle in PPI is considerably bright, and thus should not be misused and viewed at full power for any extended period of time, but the brightness is safe enough as an occasional flash. And of course, in any animal testing, care should be taken to prevent harm to oneself as well as the animal. Multiple members underwent brief computer training in animal subject treatment and safety. A summary of a designSafe™ analysis of our project can be viewed at .

Social Impact

The experiments which could be done with the completed multisensory environment could provide the first systems-level analysis of developmental plasticity and its behavioral response in the mammalian auditory system and may improve our appreciation for the formative role of experience in shaping brain development and behavior.

The efforts made to optimize a closed-loop training program to improve thresholds could result in improved clinical methodologies for those afflicted with partial hearing loss.  Of course, this assumes that auditory pathologies have a significant neural component.  The pertinence of the developed training programs to specific hearing pathologies will be dependent on accurately simulating these pathologies on the rat model.  The lab of Dr. Polley has previously made some progress on this endeavor.  Thus such investigations into hearing pathologies could be made quite soon after the rig is completed.

As more researchers enter the multisensory processing field, there is a growing appreciation of the ubiquity of sensory processing disorders in a vast array of human clinical conditions (e.g., Alzheimer’s disease, autism, dyslexia, etc.) and a greater understanding of the non-linearities associated with multisensory processing. Consequently, the creation of a more naturalistic multisensory laboratory environment represents a necessary step in the development of practical tools that can be used in the early diagnosis and treatment of these disorders.

4. Conclusions:

4.6 Tactile stimulator

The proposed piezoelectric system plan accounted for and incorporated all the necessary system parameters required for the design of a somatosensory tactile stimulator. While the developed system could not provide the originally desired deflection length, it was decided that this was a secondary concern of the system and could be corrected later if the offered 800nm was insufficient. They designed and developed system matched all our other design criteria very well (small, silent, gentle, etc…). Ultimately the somatosensory system, once fully integrated in the auditory and visual environment, should provide the desired effect without any adverse artifacts.

The next step in the development of the tactile stimulator would be to attach a custom tip to the end of the bending actuator to increase the specificity of the tapping point. A custom tip would allow the user to pin-point to location of stimulation to ensure the researcher that knows exactly which nerve endings perceive the stimulation. Also, while it may not present any limitations currently, it is expected that the actuator deflection would need to be increased into the 1-2mm range. This may be done simply by adding an additionally piece to the end of the currently used actuator; however this may result in dropping the actuator’s resonance frequency into the experimental regime, compromising the systems fidelity.

4.7 Multisensory protocol software

We have created an analysis of the software in a form of a flowchart diagram which is used as a guide to the researchers in the lab so that they can understand how the software works a little better. Another purpose of this analysis is to guide other programmers if they are interested to create software similar to this.

4.8 Pre-Pulse Inhibition protocol software

The PPI protocol works reliably to a high level, and is to the point of being testable with manual weight calibration or a real live animal. Preliminary testing has already been done with using a free-field speaker to trigger the force plate, and no problems or issues seemed to occur. This bodes well for the future goal of implementing more complex protocols such as Mismatched Negativity and some forms of operant conditioning to specifically isolate hearing thresholds on varying levels of mental association. This will allow the experimenting necessary to hopefully lead to optimization of plasticity retraining protocols for clinical use.

5. References:

A. Dirks. Reduced startle reactivity and plasticity in transgenic mice overexpressing corticotropin-releasing hormone. Biological Psychiatry, Volume 51, Issue 7, Pages 583–590.

Lazar R and Metherate R. (2003) Spectral interactions, but no mismatch negativity, in auditory cortex of anesthetized rat. Hear Res. 2003 Jul;181(1-2):51-6.

Meredith MA and Stein BE. (1985) Descending efferents from the superior colliculus relay integrated multisensory information. Science 227: 657–659, 1985.

Meredith MA and Stein BE. (1983) Interactions among converging sensory inputs in the superior colliculus. Science 221:389–391.

M.S. Silverman and B.M. Clopton. (1977) Plasticity of binaural interaction. I. Effect of early auditory deprivation. J Neurophysiol, Nov 1977; 40: 1266–1274.

Appendix A: Innovation Workbench

Innovation Situation Questionnaire (ISQ)

Brief description of the situation

Modify flow energy

Modify pressure, resistance: pneumatic principles should be considered for effective somatosensory stimulus via valve-controlled air puffs

Produce impacts

A silent adjustable somatosensory stimuli is desired.

Reduce the effects of an undesired action

Localize a harmful action: attaching wires to the animal introduces the possibility of stresses on these wires transmitting to stresses to their connections, resulting in bad contacts over time and ruined electrical signals. A commutator should be used so that torques on the wire are localized to the freely spinning commutator head, isolating the connections from this stress.

Reduce overall dimensions

Make an object dismountable: the containment structure for the rat should include easily swappable components for stimulus presentation and animal interaction, for flexible configurations in the variety of experiments

Improve productivity

Concentrate energy: sound delivery to the rat's ear should be from very close proximity to the ear canal, and insulated to reduce external interference

Improve convenience

module principle: levels of hierarchy in the implementation of programs for stimulus presentation and analysis; encapsulation

Adapt the tool to the user: explicit visualization tools and extensive options for greatest ease for analysis software user

Reduce cost

Disposable Objects: Use disposable materials to buffer animal wear from valuable equipment.

Restoration: Use materials which can be cleaned after exposure to animals

Reduce time wasted

Preliminary placement: Utilize a standard/fixed location for each stimuli to reduce time lost to determine placement.

Eliminate Idling: As little time as appropriate should be lost between stimulation/data collections.

Use Post-Process Time: Some calculations on the data are essential enough that they should be done right after data is collected, but they are to slow to perform live during collections. Thus these calculations are done as immediate post-processing following an experiment.

Reduce noise

Isolation: sound stimuli are delivered wirelessly to avoid interference to wired electrophysiology leads

Detailed description of the situation

Supersystem - System - Subsystems

System name

Sensory Environments for Awake Neurophysiological Recording

System structure

The core of our system are the two sensory environments being built, one for rats and the other for cats.

Controller rack-mounted electronics control the stimuli presentation for training or experiments, as well as mediate the data collected by the environment.

Analysis software takes this data to format it and facilitate making scientific conclusions.

Supersystems and environment

120 Hz electrical interference from electricity present at electrical outlets

Currently the chambers in which stimuli are presented to the animal are within soundproof chambers, which should greatly reduce acoustic interference from the outside.

Systems with similar problems

The specifications for stimulus presentation are fulfilled by current commercial turnkey systems. We can use several features from these examples for use in our own devices, with exceptions made for cost reduction and greater integration between them (such as all being triggered by standard TTL pulses).

Input - Process - Output

Functioning of the system

The system, in the end, will effectively make the animal subject a sort of 'black box' module to which stimuli are input and electrophysiological signals are output.

System inputs

The system is built within a multisensory environment where the inputs to the system include auditory signal, visual stimulus, and somatosensory stimulus.

System outputs

The outputs are neurophysiological, musculoskeletal and autonomic responses which are quantified from the electrophysiological signals that are obtained from the 'black box'.

Cause - Problem - Effect

Problem to be resolved

How do we develop a sensory environment capable of stimulating multiple sensory responses in a controlled way?

Mechanism causing the problem

A lack of pre-existing hardware and software packages designed to meet sensitive physiological requirements.

Undesirable consequences if the problem is not resolved

The data will be inconsistent due to signal contamination from outside sources or uncontrolled/inappropriate stimulation of sensory neurons.

Other problems to be solved

How can we specifically excite sensory neurons without the need for mechanical stimuli?

Past - Present - Future

History of the problem

Pre-process time

Predetermining placement of stimulators and setting them up prior to the experimental procedure. Also, developing a consistent system for the attachment of hardware to the animal when needed.

Post-process time

Data analysis can be used to develop a calibration or adjustment factor to help ameliorate signal contamination/distortion.

Resources, constraints and limitations

Available resources

Available to us are the pooled knowledge of our two advisors, Dr. Polley and Dr. Wallace, as well the materials and space within their lab in MRB- 3. The Vanderbilt library system and online periodical subscriptions offer a wide base of informational resources. There is also some pre-existing programming available to build upon and learn from.

Allowable changes to the system

In the development of the physical environment, there is a small degree of freedom in changing the system due to the pre-existing hardware. The software being developed offer a greater allowable change but is also somewhat limited by the work done previously.

Constraints and limitations

There has already been considerable investment into the current environment and software package both in terms of time and money. Further, physical size and cost limit our ability to stray far from the current design.

Criteria for selecting solution concepts

Solution concepts should not impart an excessive economic burden. Solutions should present a greater degree of control of each stimuli in the physical environment and a greater efficiency in data analysis from the software. Another solution criteria would regard how quickly the design/program can be implemented. The potential for use by outside labs performing similar research should be considered when selecting a solution.

Problem Formulation and Prioritize Directions

Problem Situation

[pic]

Find a way to eliminate, reduce, or prevent poor stimulator design

Isolate the system from the harmful effect

Ear plugs can be used to firmly place auditory stimulators in the ear and isolate ears from outside noise. Conducting the study in a dark room or under a shade may help to reduce unwanted visual stimulation.

Counteract the harmful effect

Encapsulate and insolate any device which may produce an unwanted stimulus, such as muffling the sound made by the somatosensory probe.

Eliminate the cause of the undesired action

Design sensory probes which do not produce an auditory artifact. Also, use visually stimulus which does not produce any transient noise associated with turning the bulb on/off.

Reduce the harmful results

Isolating and insulating the devices producing the unwanted stimulation such that they are not perceived by the animal.

Find a way to eliminate, reduce, or prevent poor software design in order to avoid misleading data analysis

Enhance useful parameters

Reliability

Inputs that are used in the software should be meaningful and quantifiable in every function, and read at the right time so that the inputs used are the corresponding response from a respective stimulus and not from another.

Action speed

Optimizing the program by making sure that the codes are written in the shortest possible length. Consistent use of parameters between functions so that multiple conversions can be avoided.

Controllability

By having a proper general user interface that enables the user to control the software properly from a control menu.

Reduce the harmful results

By having the software to run in synchronization with the input coming in.

Find a way to eliminate, reduce, or prevent misleading data analysis under the conditions of poor software design

Lower harmful parameters

Object complexity

By making sure that the inputs are organized, labeled correctly, and backed up to prevent any confusions while running the software.

Time wasted (utilize time resources)

Consistent use of parameters between functions.

Reduce the harmful results

Reliability in every single function despite the poor design of the software. Identify harmful results so that maintenance of the software could be done periodically.

-----------------------

[pic]

Figure1: Somatosensory tactile stimulator system.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download