Research Supplements to Promote Diversity in Health-Related Research

Duke University, Durham NC

Investigators

Linked publications, trials & patents

Abstract

ABSTRACT: Surgical tumor resection depends on the ability to distinguish tumor from normal tissue parenchymal. This determination is frequently made in real-time by the surgeon using direct visualization, mechanical and tissue handling properties at the time of resection, intraoperative pathology analysis and other intraoperative imaging methods. Extensive training and assistive intraoperative imaging devices allow the surgeon to remove the unhealthy tissue, and subsequent confirmation, classification and grading is often performed over the subsequent days to weeks through standard histopathologic, immunohistologic, and microscopic methods. This delay in the confirmation process (often taking up to a week or more to complete the following resection and in most instances not complete by the conclusion of the operation) can impact the timing of adjuvant treatments pending these critical data which guides subsequent clinical decisions for the patient. It is precisely this gap in treatment that we hope to explore. Both the Brain Tool Laboratory and Duke University Hospital have and support an infrastructure for researchers to tackle complicated and critical problems in healthcare. First, and more basically, we aim to evaluate the ability of a noncontact endogenous fluorescence excitation and emission device to classify tissue pre- and post-ablation. These findings would assuredly lend themselves to research conducted for laser interventions in surgery. But more fundamental is our aim to investigate whether the use of such device can improve the surgeon’s and pathologist’s ability to successfully classify tumor. Our system is designed to use endogenous fluorescence to interpret metabolic biomarkers of tumors and parenchyma on surgical explanted tissues to assist surgeons and pathologists in making an accurate diagnosis. Upon successful classification based on excitation and emission spectra, we further aim to create a technology of machine learning algorithms that distinguishes first between normal versus tumorous tissue and second among tissue subtypes. Contact-dependent methods have been pursued and studied in the past, especially successful in small animal models. However, our proposed noncontact method has not yet been investigated in the operating room. A surgeon’s decision to pursue a margin or lesion more aggressively depends on having the most accurate and accessible information available. By offering just that, we hope to tackle this need for an intraoperative detection device that determines not only the tumor margin but real-time pathological identification and classification. To this end, we will conduct the following Specific Aims: Aim 1: Investigate whether the use of a non-contact endogenous fluorescence excitation and emission device improves the surgeon’s and pathologist’s ability to successfully classify tumor. Aim 2: Evaluate the device’s ability to classify tissue pre- and post-ablation to expand its applicability in the TumorCNC.