QI announces the successful completion of its pivotal clinical study.
A total of 19 readers completed the study which measured the effectiveness of the QuantX™ system in the task of distinguishing between cancerous and non-cancerous breast lesions. In this study, the QuantX system achieved the primary endpoint of a statistically significant increase in radiologist’s performance. When QuantX was used in image interpretation, AUC increased from 0.7055 to 0.7575 (p=0.0408), compared to interpretation using tools available in existing commercial systems. Mean sensitivity increased from 90.4% to 94.2%.
Performance improved for all of the readers in the study, and very significantly, the study demonstrated a reduction in false negatives (i.e. missed cancers) by 39%.
Each radiologist read 111 retrospectively collected cases in a sequential read protocol. The first read was performed using tools that mimic the tools available on current breast MRI analysis workstations. The case dataset was enriched with cancer cases in order to more effectively measure the performance of the QuantX system.
Reader demographics were well distributed across the key characteristics as follows: eight from academic and eleven from private practice settings, nine with greater than 10 years of breast imaging experience and ten with 10 or fewer, and eight who successfully completed a breast imaging fellowship versus eleven who didn’t. As far as direct breast MRI experience: eight had 5 years or less, seven had 5-10 years’ experience and four had greater than 10 years.
This very successful clinical study result was included in a De Novo 510(k) package and submitted to the FDA for product clearance review.