450 BME class project: Optical detector for medical ultrasound Proposed by: Biomedical Ultrasonics Lab, Biomedical Engineering, 3417 G.G. Brown Customer: Monica Spisar ([e-mail], 647-3201) Medical ultrasound is currently limited in spatial resolution by the maximum operating frequencies of the transducers used to generate and detect the ultrasound signal. Two dimensional arrays for high frequency (>30MHz) ultrasound imaging are difficult to construct using conventional piezoelectric technology. A promising alternative involves optical generation and detection of ultrasound, where the array element size is defined by the focal spot of a laser beam. Interferometers are used for the optical detection of ultrasound; a laser signal is intensity modulated in response to an ultrasound signal. The cavity length of a plane-parallel (FabryPerot) interferometer is modified by the ultrasound, and the intensity of the reflected laser light follows that modification (see figure 1).
laser
optical switch
photodetector
interferometric cavity
ultrasound signal
Figure 1 A technical challenge associated with this method is sustaining a constant DC tuning level as the laser beam scans the cavity to define the array elements. Small variations in the cavity thickness, as well as laser frequency drifts, will essentially modulate the optical signal in the same way as the ultrasound signal of interest, but on a much slower time scale. To avoid confounding these effects with the actual ultrasound signal, the optical path length through the cavity can be adjusted by tilting the cavity to adjust for these slow intensity modulations. This tilt-tuning must be automated to make imaging possible (typical arrays involve 100x100 elements). Automation of the cavity tilt-tuning is the proposed design project. This will involve building a system that uses a feedback circuit which monitors the DC output of the reflected laser intensity and uses that information to control pushers (perhaps piezoelectric) that adjust the tilt of the cavity to achieve a pre-selected DC level. The cavity itself will be provided (and templates for the feedback circuitry are available, if desired), but all other elements will be designed and constructed by the team members.
BME 450 class project: System measuring elastic properties within a slab of soft tissue Proposed by: Biomedical Ultrasonics Lab, Biomedical Engineering, 3423 G.G. Brown Customer: Ramon Erkamp ([e-mail], 764-4121) Characterizing the nonlinear elastic behavior of tissue provides useful information for designing ultrasound algorithms that measure these properties non-invasively. To do this, a system was developed that measures the distribution of elastic properties in a slab of tissue by pushing a piston onto it while measuring the resulting force.
A horizontal positioning system and a rotational table under stepper motor control are used to position a piston at a particular coordinate above the slab of tissue. A third stepper motor positioning system pushes the piston down into the tissue, while an electronic balance measures the resulting force. The whole slab is systematically scanned to obtain an image of the elastic distribution. The whole system is controlled by a PC, stepper motors through the parallel port and the electronic balance through the serial port.
A major problem of the current system is that it is too slow - a scan may take up to one full day and the tissue degrades during these measurements. If multiple locations could be measured at the same time, it would greatly speed up the procedure. Now the scale can no longer be used for force measurement, instead we get a system with multiple pistons and within each of the pistons a force sensor. Good quality commercial load cells start at $300, much too expensive if a system with many pistons is to be built. A set of 4 good quality semiconductor strain gages on the other hand only costs about $10, so a custom designed strain gage based sensor would be preferable.
Conventionally strain gages are used in a bridge configuration, and mechanically mounted in such a way that when force is applied one pair is stretched while the other is compressed. Two possible designs sensor that might do this are shown on the left. The bending type sensor will cause a fairly large force dependent piston displacement, the axial load sensor may need unconventionally soft materials for high enough strain in the gages.
Some of the issues involved in sensor design are: gage selection, structural design, material selection, temperature compensation, calibration, measurement error evaluation, and gage bonding issues. The second part of the project involves interfacing these sensors to the system PC. This will most likely involve selecting and purchasing an appropriate AD capture board, and developing some C-code to grab the data. The current system runs in DOS mode (to allow real time stepper motor control over the parallel port), software was written in C and compiled using the Borland C++ compiler. The sensor interface design has to be compatible with the already existing parts of the system. The desired force range is 0-1Newton, with an accuracy of around 0.001 Newton. Skills useful in completing project: basic analog circuit design, machining (access to metal machining tools would be convenient), basic mechanical physics, programming in C, and general knowledge about PC hardware.
BME 450 class project: Dynamic Brain Activation Phantom Proposed by: fMRI, Biomedical Engineering, 3440 G.G. Brown Customer: Scott Peltier ([e-mail], 764-4121) The goal is to design and construct a computer-controlled phantom to simulate small changes in image intensity that can mimic the changes in image intensity seen in functional brain imaging experiments. The purpose of the phantom is to evaluate the time resolution of a variety of image acquisition and processing methods in a controlled environment. Basic tasks: 1) determine the mechanism of changing image intensity (mechanical movement of materials, change of composition (dopants) of flowing materials, rapid but controlled temperature changes, others), 2) design phantom and select materials (plexiglass cylinders, doped water, siliconized gels, etc.) including design of system for changing image intensity (mechanical translation, etc), 3) construct phantom, 4) design user interface and controlling program for computer, 5) write program user interface/controller, 6) test in an MRI system.
BME 450 class project: Microscope Sample Chamber Fluid Exchanger Proposed by: Cellular and Molecular Biomechanics, Biomedical Engineering, 300 NIB Rm 952 Customer: Ajit Joglekar ([e-mail], 764-3676) During study and manipulatio...