SIGN Fracture Care International, located in the Tri-Cities Research District, manufactures stainless steel nails used in surgeries to treat long bone fractures in third-world countries. The firm estimates that SIGN-trained doctors have treated 250,000 patients over the past 20 years. During that time, the company has accumulated 500,000 X-rays of various procedure images and outcomes. The images are housed in a data base that serves as a learning hub for doctors worldwide. However, sorting through the extensive collection of images proved difficult, so SIGN turned to Pacific Northwest National Laboratory’s machine learning scientists to develop computer vision tools that identify surgical implants in the images, making it easier to navigate the vast database and improve surgical outcomes. Lacking good initial examples, the PNNL team taught the computer to focus on the implants and not mistake the fingers holding the X-ray image for one of the implant’s screws.

The PNNL research team is now focused on training the computer to recognize the image quality and automatically prompt a doctor to upload a usable image. Currently, SIGN’s founder, Dr. Zirkle, manually approves hundreds of images a day. Automating the image approval process would free up time for Dr. Zirkle to focus on teaching or other tasks.