Area of Focus/Speciality:
Lunit, abbreviated from “learning unit,” is a venture company devoted to developing advanced medical image analytics and novel imaging biomarkers via cutting-edge deep learning technology. Lunit has proven high-end technology, recognized at international competitions such as ImageNet (5th place, 2015), TUPAC 2016 (1st place), and Camelyon 2017 (1st place), surpassing top companies like Google, IBM Research Zurich, and Microsoft Research Asia.
We envision a near-future when our systems would greatly help physicians make accurate, consistent, and efficient clinical decisions. We have so far focused on chest and breast imaging as well as digital pathology. Our algorithms for chest x-ray and mammography have been clinically validated to perform better than experts and significantly improve the diagnostic performance of its users.
At this year’s RSNA, Lunit is launching Lunit INSIGHT, a cloud-based real-time imaging AI platform that is currently available to the public at https://insight.lunit.io/. The platform delivers Lunit’s state-of-the-art AI models; the first one to be unveiled is the chest x-ray solution that accurately detects lung cancer, pneumothorax, tuberculosis, and pneumonia.
Press and Publications:
Lunit INSIGHT for Chest Radiography accurately detects
lung nodules/mass (e.g. lung cancer), consolidation (e.g. pneumonia, tuberculosis), and pneumothorax in chest x-
ray images, presented via a heatmap and corresponding confidence score. Its accuracy levels reach 97-99% in AUC according to yet to be published clinical studies.
Lunit INSIGHT for Mammography accurately detects
breast cancer lesions in mammograms, presented via
a heatmap and corresponding confidence score. Its
accuracy levels reach +94% in AUC according to yet to
be published clinical studies.