Grant: $413,808 - National Institutes of Health - Aug. 4, 2009
No votes have been cast for this award yet
Award Description: Fluorescent in situ hybridization (FISH) imaging has been proven as a powerful molecular imaging tool to detect cancers, predict cancer prognosis, and monitor therapy efficacy. However, manual FISH analytic method is a tedious procedure, and introducing inter-reader variability. It is difficult or unpractical to use manual FISH analytic method to detect a small number of abnormal chromosome cells from the overwhelming number of normal cells depicted on one specimen (i.e., screening for early cervical cancer using Pap smear). To solve this difficulty, we proposed to develop and test an innovative automated FISH imaging system that includes an novel optical image scanner and a computer-aided detection (CAD) scheme. Comparing to the currently available techniques, our system provides unique advantages. First, it substantially improves scanning efficiency and avoids the requirement of multiple image registration (that is a difficult task and often generates errors). Second, it allows real-time (on-line) CAD processing to detect analyzable or abnormal cells. Our hypothesis is that if all cells and FISH signals depicted on one specimen can be automatically scanned and correctly counted, then abnormal or carcinoma cells, the sign of cancer, can be detected, and the cervical cancer can be diagnosed with higher sensitivity and specificity in early stage. To test this hypothesis, we will develop a prototype system, assemble a large and diverse image database that includes both cytology and FISH images for each Pap smear examination. We will conduct a three-mode observer performance study in which each observer will read and diagnose each FISH specimen three times using three image display modes, which are (1) the conventional microscopic image viewing, (2) the automated FISH image viewing, and (3) the CAD-guided FISH image viewing. We will apply ROC methodologies to analyze and compare the screening and diagnostic performance between these three FISH image reading modes as well as the conventional cytology method based on the individual observer and the group of the observers. Through the follow-up of these cases and by updating the 'truth file' during this project, we can also investigate whether using this automated FISH imaging scanning and analysis system can help clinicians detect more cancers in the early stage. If the feasibility and clinical significance of this system is successfully tested and validated in this study for Pap smear samples, it can easily be applied to screening and diagnosis of many other types of cancers. Hence, this project can generate more follow-up clinical studies that were not feasible in the past. PUBLIC HEALTH RELEVANCE Fluorescent in situ hybridization (FISH) imaging has been proven as a powerful molecular imaging tool to detect cancers, predict cancer prognosis, and monitor therapy efficacy. This project aims to develop and test an automated FISH imaging and analysis system. The system includes a fluorescence microscopic imaging module and a computer-aided detection (CAD) scheme for analysis of FISH images acquired from Pap smear specimens. We will also conduct an observer performance study to assess whether and how this automated FISH image scanning and analysis method can help clinicians to screen and detect cervical cancers at an early stage, with improved accuracy and efficiency.
Project Description: As defined in the Award Description field
Jobs Summary: Post Doctoral Research Associate and Graduate Research Assistant. Assisting faculty members in a research or creative activity, or assuming responsibility for a designated research area (Total jobs reported: 1)
Project Status: Less Than 50% Completed
This award's data was last updated on Aug. 4, 2009. Help expand these official descriptions using the wiki below.