3-D modeling-based decision support for optimizing quality of life following breast reconstruction

Breast reconstruction is a central component of the breast cancer treatment process for many women. The purpose of breast reconstruction is to recreate a breast form that is satisfying to the patient, facilitating her psychosocial adjustment o living as a breast cancer survivor. Decisions about breast reconstruction can be difficult and overwhelming for many women due to the number of options available and conflicting values and preferences. The long-term goal of our research is to enhance the consultation process for women undergoing breast reconstruction surgery. Our vision is a decision support system that will enable breast cancer patients, in consultation with their healthcare providers, to choose a reconstruction strategy with maximal potential to optimize psychosocial adjustment. This system for shared decision-making will use quantitative models to tailor the presentation of patient-specific information about breast reconstruction outcomes. Three common features of decision support systems are the knowledge base; the methodology for combining that knowledge with patient-specific information; and the communication mechanisms. With recent NIH support, we established a unique knowledge base through a longitudinal study of 500 women who underwent breast reconstruction. We also developed preliminary computational models for combining that knowledge base with patient-specific data. Our models are parameterized by experimental, clinical, and psychosocial data and employ techniques from 3D image analysis, biomechanical modeling and simulation, machine learning, and decision analysis. In the proposed project, we will develop predictive algorithms for combining the knowledge base and patient- specific information, and we will develop communication mechanisms both for gathering necessary patient- specific information and for delivering information about the decision basis to the patient and her multidisciplinary care team.