MONDAY, Aug. 14, 2023 (HealthDay News) — A fully automated deep learning pipeline can accurately measure sarcopenia in head and neck squamous cell carcinoma (HNSCC), which is associated with disease outcomes, according to a study published online Aug. 10 in JAMA Network Open.

Zezhong Ye, Ph.D., from Harvard Medical School in Boston, and colleagues developed and externally validated a fully automated image-based deep learning platform for cervical vertebral muscle segmentation and skeletal muscle index (SMI) calculation. Associations with survival and treatment toxicity outcomes were examined using data for 899 patients with HNSCC undergoing primary radiation with abdominal computed tomography scans and complete clinical information.

The researchers found that the dice similarity coefficients for the validation set and internal test set (96 and 48 participants, respectively) were both 0.90, with a mean 96.2 percent acceptable rate between two reviewers on external clinical testing (377 participants). Across datasets, estimated cross-sectional area and SMI values were associated with manually annotated values. Compared with no sarcopenia, SMI-derived sarcopenia was associated with worse overall survival and longer feeding tube duration (162 versus 134 days).

“If further validated, our end-to-end deep learning pipeline could be incorporated into standard clinical practice for directing future treatment approaches and clinical decision making, as well as for individualized supportive measures, including nutrition guidance and physical therapy,” the authors write.

Several authors disclosed ties to the pharmaceutical and biopharmaceutical industries.

Abstract/Full Text

Copyright © 2023 HealthDay. All rights reserved.