Since 1999 the company worked closely with UCSD Medical School (laboratory of Professor Michael Andre) and VA San Diego on development of computer-aided decision making support system that would enable radiologists and other medical imaging professional to enhance their productivity and reduce error margins in false positives and false negatives diagnostic decisions.
A well-defined rule-based system has been developed for scoring 0-5 the Level of Suspicion (LOS) based on qualitative lexicon describing the ultrasound appearance of breast lesion. The purposes of the research are to asses and select one of the automated LOS scoring quantitative methods developed during preliminary studies in benign biopsies reduction. The study has used Computer Aided Imaging System (CAIS) to improve the uniformity and accuracy of applying the LOS scheme by automatically detecting, analyzing and comparing breast masses. The overall goal is to reduce biopsies on the masses with lower levels of suspicion, rather that increasing the accuracy of diagnosis of cancers (will require biopsy anyway). On complex cysts and fibroadenoma cases experienced radiologists were up to 50% less certain in true negatives than CAIS. Full correlation analysis was applied to determine which of the proposed LOS quantification methods serves CAIS accuracy the best. The CAIS system (Image Companion®, Data Companion® software) is developed by Almen Laboratories and was used to achieve the results.
DATA SET USED FOR CLINICAL VALIDATION
IRB-approved clinical cases with known findings retrieved in
Biopsy-proven or two-year follow up for benign (PACS)
malignant and 78% benign cases
OVERALL PERFORMANCE OF THE VALIDATED BREAST ULTRASOUND CAD SYSTEM
Predictive Value of 90.3%
Predictive Value of 96.5%
Read more about the protocol design (400K PDF file)
Read more about the image processing tools used
Read more about the results of ROC analysis (1 MB download ZIP file)
Read more about statistical scoring method (500K PDF File)
about latest results of clinical validation (600K PDF File)