Oxipit�s ChestGlass has been selected as one of the top projects for the 2018 Society for Imaging Informatics in Medicine (SIIM) Innovation Challenge. ChestGlass is the first-to-market chest X-ray search solution that allows to find radiologically similar images in a given database.
ChestGlass is a Deep Learning-based image search solution which takes an unannotated image straight from the X-ray machine and uses the database of the hospital to find the most radiologically similar cases along with their reports. The ability to find similar cases within large databases assists the radiologist by making it easier to investigate radiographs with rare or difficult findings, as well as to carry out research. It will be especially useful for radiology residents and radiology professionals making a shift to another modality.
�Several decades ago chest X-ray technology has moved from photographic film to digital. By now, healthcare institutions have archived millions of images in databases which are virtually unused. Today, with excellent tools long available for general information search, radiology is ready for another step forward,� says Gediminas Peksys, CEO of Oxipit. �The novel possibility to query radiology databases not only by keywords or study IDs but also by full images will permanently alter the radiological workstation functionality expectations.�
While superficially similar to image search products in other industries, ChestGlass is custom tailored for radiological image search. It takes into account radiological findings and their localizations present in the image (pixel data only). In this way, ChestGlass seamlessly overcomes barriers of language and report structure diversity.
"This innovative solution has been made possible by cooperation between our clinical partners and the team of excellent data scientists at Oxipit. It is a pleasure to see their work widely appreciated," adds Mr. Peksys.
In addition to the chest X-ray radiograph search, Oxipit offers a spectrum of fully automatic solutions based on Deep Learning: chest X-ray computer-aided diagnosis (CAD); MRI brain anatomic segmentation, ischemic area and lesion detection/quantification; spine X-ray vertebrae segmentation and osteochondrosis/listhesis CAD.
Oxipit is currently looking for individual as well as institutional partners for pilot deployment and investigational use in the clinical environment. For more information, go to www.oxipit.ai.
About Oxipit (www.oxipit.ai)
Oxipit is a computer vision software startup specialized in medical imaging. With a team of award-winning data scientists and medical doctors, the company aims to introduce recent Artificial Intelligence/Deep Learning breakthroughs to everyday clinical practice. Oxipit�s solutions bring value to radiologists (less routine tasks), healthcare institutions (more productive radiologists), and patients (more accurate diagnosis).
About SIIM (www.siim.org)
The Society for Imaging Informatics in Medicine (SIIM) represents more than 2000 healthcare professionals with interest in enterprise information and image management. SIIM is a not-for-profit organization focused on education, clinical practices, research, and innovation in the field of imaging informatics. The society is committed to actively contributing to the development and advancement of imaging informatics professionals and scientists worldwide and to improving the quality, safety, and efficiency of healthcare.
