X-ray-like images from ultrasound scan through AI helpful for scoliosis diagnosis: Study

China: In a recent study, published in Ultrasonics, a team of researchers from China described the use of artificial intelligence (AI) algorithm that creates x-ray-like images from ultrasound data to visualize adolescent idiopathic scoliosis. 

The results could be of significance for children at risk of scoliosis progression with their body growth, a condition known as adolescent idiopathic scoliosis (AIS) that affects 1% to 3% of the adolescent population. This technique could reduce radiation exposure and future cancer risk. 

The gold standard for scoliosis diagnosis is standing X-ray radiograph with Cobb's method. However, its application gets restricted by radiation hazards, especially for close follow-up of adolescent patients. Ultrasound imaging compared with X-ray has the advantage of being radiation-free and real-time. 

To combine the advantages of the above two imaging modalities described above, Cong Bai, College of Computer Science & Technology, Zhejiang University of Technology, Hangzhou, China, and colleagues proposed an ultrasound to X-ray synthesis generative attentional network (UXGAN) to synthesize ultrasound images into X-ray-like images. In this network, a cyclically consistent network was adopted and was trained end-to-end. The addition of an attention module was done and different residual blocks were designed. 

The study included 202 children with adolescent idiopathic scoliosis -- all of whom underwent both ultrasound and x-ray. The quantitative comparison results showed the superiority of the method to the state-of-the-art CycleGAN methods. The researchers further compared the Cobb angle values measured on synthesized images and the real X-ray images, respectively. A good linear correlation (r = 0.95) was demonstrated between the two methods.

The researchers conclude, "the above results proved that the proposed method is of great significance for providing both X-ray images and ultrasound images based on the radiation-free ultrasound scanning." 

According to the study authors, "the study findings show significant promise."

"[Our] experimental results show the reliability and accuracy of the method," they concluded. "This method can provide radiation-free ultrasound and x-ray images, which is of great significance for diagnosing and monitoring adolescent idiopathic scoliosis."

Reference:

The study titled, "Ultrasound to X-ray synthesis generative attentional network (UXGAN) for adolescent idiopathic scoliosis," was published in the journal Ultrasonics. 

DOI: https://doi.org/10.1016/j.ultras.2022.106819