Figure 4 | Scientific Reports

Figure 4

From: Performance of the deep convolutional neural network based magnetic resonance image scoring algorithm for differentiating between tuberculous and pyogenic spondylitis

Figure 4

Two examples of probability measurements within each image. (a) A 72-year-old woman with pyogenic spondylitis. Sky-blue boxes indicate pyogenic lesions detected by the deep convolutional neural network classifier. The probability of pyogenic spondylitis for this image is 1.00, which is the highest value among the detected pyogenic lesions. Because no tuberculous lesion is observed, the probability of tuberculous spondylitis is 0. (b) A 31-year-old woman with tuberculous spondylitis. Two red boxes indicate tuberculous lesions and a sky-blue box indicates a pyogenic lesion. The probability of tuberculous and pyogenic spondylitis is 0.99 and 1.00, respectively. Abbreviations: Pyo, pyogenic spondylitis; Tb, tuberculous spondylitis.

Back to article page