P2-09-03: Breast Cancer Detection Using a Novel Functional Imaging Device Equipped with Multiparametric Computer Analysis

• Published in: Cancer research (Chicago, Ill.) Vol. 71; no. 24_Supplement; pp. P2 - P2-09-03
• Main Authors: Sella, T, Sklair-Levy, M, Maya, C, Rozin, M, Tanir, A, Maira, S, Libson, E, Izhaky, D
• Format: Journal Article
• Language: English
• Published: 15.12.2011
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/0008-5472.SABCS11-P2-09-03

Introduction: A novel prototype system performing functional three-dimensional (3D) infrared imaging (RUTH), coupled with multi-parametric computer analysis (MIRA), was evaluated for non-invasive breast cancer detection. The technique utilizes non-ionizing radiation and automatically provides risk assessment based on parameters derived from a clinically known training set. Materials and methods: Following IRB approval, prospective multi-center, two-armed blinded evaluation, was performed on 102 women, which included 43 patients with biopsy proven breast cancer and 59 women with a normal (BI-RADS 1) mammogram. RUTH imaging (Real Imaging Ltd, Israel) was acquired on a continuous temporal timeline for 5 minutes during which a cold-stress test was applied for one minute. Acquired data were analyzed using multiple computed algorithms (MIRA) which were combined to develop a master algorithm for risk assessment of breast cancer. Analysis was blinded to the clinical status. Results were displayed as risk assessment scale with 100 being suspicious for breast cancer and −100 inferring normal healthy breasts. Results Thirty-nine of 43 cancers (91.7%) were identified using this novel technology. The sensitivity and specificity of RUTH coupled with MIRA for detecting breast cancer were 90.7% and 67.8% respectively. The AUC defined by the ROC curve was 0.832. Twenty-seven of the 39 (70%) detected cancers were smaller than 20mm in size. Conclusions: This preliminary study shows adequate performance of the novel technology examined for detecting breast cancer. Performance is comparable to the reported performance of new adjunctive breast imaging technologies such as tomosynthesis and contrast-enhanced mammography, and superior to the published sensitivity of mammography. Our results warrant further evaluation of the RUTH and MIRA technologies in detecting and characterizing breast cancer. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P2-09-03.