Home

Breast Cancer Screening Device

Sponsored by Dr. Mohammad Eghtedari

Moores Cancer Center - Department of Radiology

Link to Executive Summary

Project Background

Breast cancer is the most common type of cancer in women and the second leading cause of death among women in the United States. Early detection of tumors is key to increasing the survival rate. However, current methods of mammography in a Clinical Breast Examination (CBE) is often expensive, bulky, and requires trained professionals to operate the machinery and interpret the results. Bulky machines also have difficulty imaging around the auxiliary tail and the upper outer quadrant of the breast.

Our sponsor, Dr. Mohammad Eghtedari, MD, PhD, of UCSD Medical School's Department of Radiology, would like to develop a personal, hand-held device that enables self-examination for breast tumors. This CBE device could detect possible tumors in its early stages and prompt people to seek more medical attention if an abnormality would be found. The CBE device would use transducers to send a sound wave through the breast tissue, which increases in speed if it passes through a tumor. By measuring the time of flight and distance traveled, the speed can be determined and categorized as either normal or abnormal.

After initial meetings with the sponsor, the engineering team was able to identify requirements for the device that would make the product both effective and open to the possibility of commercialization. Some of these requirements were technical for precise measurements while others were more general to make the device more ergonomic and user-friendly.

The project requires the fabrication of a device that can comfortably detect the presence of cancerous breast tumors for everyday users. It must:

• • Utilize contact transducers with a diameter of 6 mm and a frequency of 2.25 MHz

  • Have an accuracy within 1% of the published speed of sound in the medium

  • Indicate easily to the user if an abnormality is present or not

  • Be capable to operate with just one hand

• Have the ability to make multiple measurements quickly and safely

Deliverables

One handheld device with:

• • A caliper to measure the distance across the breast

  • Two transducers that will be used to find the time of flight of an emitted pulse

  • Two sleeves with which to fixate the transducers upon the caliper

  • An LED light or LCD screen that will indicate the presence or lack of an abnormality

  • A means to send the caliper distance to a microcontroller that integrates with the sponsor’s electronics

  • A spring and/or holder with which to increase the usability and efficiency of the device

Proper documentation that includes:

• • Verification data and analysis

Final Design

The final design of this device consists of a modified digital caliper with a separate electronics box containing a numeric display. The modified calipers have a pair of 3D printed sleeves on the caliper arms that house the required ultrasonic transducers. The back side of the calipers contain a set of 3D printed finger grips for comfortable holding and ease of use of the device. The separate electronics box houses the PCB and microcontroller that, in conjunction, perform the speed calculation which is then displayed on the attached LCD screen.

The caliper sleeves were designed using the modeling software SolidWorks and were prototyped with a Makerbot 3D printer before final production on the Objet Connex 3D printer. The Makerbot was chosen for the initial phase due to its inexpensive material and user-friendly software while the Connex has the ability to produce the varying materials and stiffnesses necessary for the final design. TangoBlack, the softest material available on the Connex, was used for the transducer housing to allow for user comfort when taking readings. It also had the added benefit of quick and easy insertion and removal of the transducers from the device. Vero Clear, the stiffest material available on the Connex, was used for attachment of the sleeve to the caliper itself.

The caliper sleeve design includes slots for 4-40 hex nuts, to ensure contact with the caliper itself and thus provides a firmer attachment. Additionally, a clipping mechanism was added to constrain any vertical movement as well as holes for set screws to prevent any undesired lateral movement. Lastly, cable clips were added underneath the caliper sleeve to keep the transducer cables secured. The final design of the sleeves with all the aforementioned components can be seen below in Figure 2 .

Figure 2: Caliper sleeve designed to house the transducers.

The finger grips were designed to add comfortability to the design during usage. They utilize the same materials as the transducer sleeves in order to achieve the same secure attachment to the caliper while also providing optimal comfort to the user. During operation, the user will use a chosen finger (typically the index or middle finger) as well as their thumb to open and close the caliper across the breast and obtain speed measurements. In order to secure the fingers to the grips, a velcro strap has been added to ensure proper control of the device.

Figure 3: Caliper grips used to open and close the calipers

The electronics housing was designed as a means to keep the electrical components of the design safe and avoid damage. The housing itself consists of an opening to which the LCD screen display is attached, inner shelves for the PCB and microcontroller, a fan duct for cooling of the PCB, and a removable door for easy access to the PCB and microcontroller. Additionally, cable holes in the housing allow for complete operation of the device while the electronics are kept safe.

Figure 4: Housing unit to host electronic components.

The PCB was designed as a means to obtain data from both the caliper and transducers independently. For the calipers, the circuit amplifies the 1.5 volt logic level of the calipers to a 5 volt logic level for compatibility with the microcontroller and data processing. For the transducers, the circuit pulses and receives a signal from the transducers to obtain a time of flight data that is also sent to the microcontroller.

A Parallax Propeller microcontroller, provided by the sponsor, then performs the speed calculation from the caliper’s distance and transducer’s speed data. Later, the controller outputs this sound speed to the numeric display on the housing to be interpreted as abnormal or normal.

Performance Results

Speed of Sound Data: Oil Medium

Speed of Sound Data: Toothpaste Medium

Speed of Sound Data: Moisturizer Medium

Comparison of Results to Initial Performance Requirements

From the results it was concluded that the calipers are indeed precise and accurate. When the two values were compared the modified calipers were within the 0.1mm tolerance required. The only instance this was not the case was in the water bottle measurements. However, the water bottle was flexible which proved to be difficult to measure with normal calipers. The modified calipers were actually better suited and gave a more precise measurement. This was due to the larger contact patch between the calipers and the bottle.

Future tests were done in which the caliper and transducers were able to measure the time of flight and distance data for a bottle of water, oil, toothpaste and moisturizer. Looking at the tables above, the speed of sound calculation with standard deviation of approximately .6% of the mean. This verified a consistent speed of sound for each medium, meaning that the caliper and transducers system accurately measured the speed of sound through a single medium for a various distances.

When the tests were conducted on two different mediums, the results were also similar to the expected theoretical values, as shown in the tables below. The graph above shows the measured speed plotted against the distance. The distribution of the data is linear with a slight deviation.

Speed of Sound Data: Water and toothpaste Medium

Speed of Sound Data: Water and moisturizer Medium

It is clear from the tables above that there is a change when another medium is present. The change in speed reflects the presence of a non-uniform medium or the presence of more than one medium. When the speed of sound was measured through water and the moisturizer it was 1550.5 m/s. The theoretical value calculated from the time of flight obtained through the measured speed through each medium separately was 1547 m/s. This shows that the theoretical data and the obtained data matched up really well and that detection can be made a difference in about 100 m/s in the speed of sound through the mediums.