| Start Date: | 8/25/2016 | Start Time: | 10:00 AM |
| End Date: | 8/25/2016 | End Time: | 12:00 PM |
|
Event Description
BIOMED PhD Thesis Defense
Title:
Breast and Skin Cancer Detection and Depth Profiling by Tissue Stiffness Contrasting Using Array Piezoelectric Fingers (PEFs)
Speaker:
Xin Xu, PhD Candidate, School of Biomedical Engineering, Science and Health Systems
Advisors:
Wan Y. Shih, PhD, Associate Professor, School of Biomedical Engineering, Science and Health Systems
Wei-Heng Shih, PhD, Professor, Department of Materials Science and Engineering
Abstract:
Most imaging techniques project a 3-dimensional (3D) tumor into a 2-dimensional (2D) image that lacks the depth information. The ability to provide not only the lateral dimensions of tumors but also the depth profile is important for accurately sizing the tumor and is crucial for preliminary staging of the tumor prior to surgery, improving biopsy accuracy, and minimizing incomplete surgical removal of the tumors. Although computer tomography (CT) and magnetic resonance imaging (MRI) can provide tumor 3D images CT scans exposes patients to additional radiation risks and MRI is expensive. In addition, these techniques may not be suitable for assessing certain tumors such as skin cancers.
Piezoelectric finger (PEF) is a tissue stiffness sensor developed at Shih and Shih’s lab that can measure the elastic modulus of tissues both in vitro and in vivo. Because breast tumors are stiffer than surrounding tissues, it is possible to detect and image breast tumors by contrasting the higher-elastic modulus regions with the surrounding tissues. In addition, a PEF with a larger contact area can assess the stiffness of tissues at a larger depth. It is thus possible to use PEFs with different contact areas to probe for depth profiles of tumors.
The goal of this study is to develop the methodology to use array PEFs not only to detect breast tumors and skin tumors but also image their locations and sizes in 3D for various applications. In Aim 1, a handheld probe containing an array of four PEFs of the same contact area (6.5 mm) is developed together with a custom-built circuit board to detect breast tumors in 40 patients. The results show that PEF detected 96% of breast tumors, including 100% of palpable and 67% non-palpable malignant tumors. Among the 28 patients with mammography records, PEF detected 92% malignant tumors while mammography only detected 80%. Furthermore, PEF detection was not affected by mammography density, indicating that PEF is promising for detecting breast tumors in young women and women with dense breasts for whom mammography is ineffective.
In Aim 2, tumor depth profiles was determined using the stiffness measurements by a set of PEFs with contact sizes 4.1 - 9.8 mm on model breast tumors of clays embedded in gelatin coupled with a spring model. The locations of the top of bottom-supported model breast tumors were determined within 1.1 mm of the actual values. For suspended model breast tumors both the top and the bottom margins were determined within 2.1 mm of the actual values, indicating that it is a promising methodology for tumor depth profiling. In addition to the depth accuracy, the current spring model-based methodology has the advantage of being instant as compared to the inversion simulations (IS) using finite element analysis (FEA) which gives similar accuracy but is tedious and time-consuming.
In Aim 3, a mechanical model of skin was established as a two-layer structure with the stiffer layer representing the epidermis and dermis (skin) on top of the softer subcutaneous layer. The elastic modulus and thickness of skin were then simultaneously determined using the stiffness measurements obtained with PEFs of different contact sizes of <3 mm coupled with an empirical formula for a two-layer structure derived from Green’s function calculations. Both the elastic modulus and the thickness of the skin layer were resolved within <10% of the actual values in skin phantoms, and porcine skins and validated by FEA.
In Aim 4, the lateral extent and the depth profile of model skin cancers of clay embedded in porcine skins were determined using the stiffness measurements with PEFs of various contact sizes of <3 mm coupled with a modified spring model taking into account of the two-layer nature of skin. The lateral sizes of model skin cancers determined by PEF were within an error of 1 mm and the estimated depth profiles showed good agreement with the actual thickness with <0.4 mm discrepancy.
In conclusion, PEF is capable detecting breast cancer with sensitivity better than mammography and independent of mammography density. In addition, using a set of PEFs of different contact areas coupled with simple spring-model calculations the depth profiles of both breast cancer and skin cancer can be accurately determined to facilitate 3D breast cancer/skin cancer imaging. |
|
Location:
Bossone Research Center, Room 709, located at 32nd and Market Streets. |
Audience:
Undergraduate StudentsGraduate StudentsFacultyStaff |
|
