There is room for improvement in the accuracy of AI models that measure x-rays. Other studies have reported a mean absolute error (MAE) ranging from 4.3° to 8.1° when AI is used to assess lumbar lordosis 16, 17. Because they require relatively less preprocessing than other algorithms, and because they automatically learn to optimize filters, whereas traditional algorithms do so manually 13, 14, they may reduce the labor involved in measurement.Ī recent study 15 of CNNs showed that the standard errors for determining lumbar lordosis in scoliosis ranged from 2.7° to 11.5°. Artificial intelligence (AI) models using convolutional neural networks (CNNs) have excellent capabilities for image recognition 10, 11, 12. Human measurement error has been generally reported to be approximately 3° to 7°, and this value is believed to be similar for the cervical spine 7, 8, 9. In scoliosis, the accuracy of measurement has been verified so far. Obtaining the necessary measurements for many parameters before and after surgery for a large number of patients requires a great deal of labor 6. In recent years, digital viewer measurements became more common 5, but surgeons generally still had to obtain measurements manually. Historically, such measurements have been obtained by using a protractor on radiographs. Measuring cervical alignment in multiple positions is important in evaluating pathology and planning surgery 4. Similar content being viewed by othersĬervical alignment, an important clinical parameter in spine disorders, is associated with deformity, myelopathy, adjacent-segment disease, horizontal gaze, and health-related quality of life 1, 2, 3. However, because of the large errors in rare cases such as highly deformed ones, AI may, in principle, be limited to assisting humans. AI can assist in routine medical care and can be helpful in research that measures large numbers of images. The AI model measured cervical spine alignment with better accuracy than surgeons. The AI model had a significantly smaller error than Surgeon 1 and Surgeon 2 (P = 0.002 and 0.036). For comparison of other surgeons, the mean absolute error for measurement of 168 patients was 3.1° ± 3.4° for the AI model, 3.9° ± 3.4° for Surgeon 1, and 3.8° ± 4.7° for Surgeon 2. In fivefold cross-validation, the absolute error of the AI model was 3.3° in the average and 2.2° in the median. Additionally, the absolute error of AI measurements was compared with the error of other 2 surgeons’ measurements on 415 radiographs of 168 randomly selected patients. The absolute error of the AI measurements relative to the ground truth for 4546 x-rays was determined by fivefold cross-validation. This ground truth was split into training data and test data, and the AI model learned the training data. For all x-rays, the caudal endplates of C2 and C7 were labeled based on consensus among well-experienced spine surgeons, the data for which were used as ground truth. We included 4546 cervical x-rays from 1674 patients. This study aimed to develop AI for automated measurement of lordosis on lateral cervical x-rays. Artificial intelligence (AI) in the form of convolutional neural networks has begun to be used to measure x-rays. Manual measurement is time-consuming and burdensome to measurers. Cervical sagittal alignment is an essential parameter for the evaluation of spine disorders.

Lateral x ray of cervical spine