Deep learning auto-segmentation of cervical skeletal muscle for sarcopenia analysis in patients with head and neck cancer

Mohamed A Naser; Kareem A Wahid; Aaron J Grossberg; Brennan Olson; Rishab Jain; Dina El-Habashy; Cem Dede; Vivian Salama; Moamen Abobakr; Abdallah S R Mohamed; Renjie He; Joel Jaskari; Jaakko Sahlsten; Kimmo Kaski; Clifton D Fuller

doi:10.3389/fonc.2022.930432

Deep learning auto-segmentation of cervical skeletal muscle for sarcopenia analysis in patients with head and neck cancer

Front Oncol. 2022 Jul 28:12:930432. doi: 10.3389/fonc.2022.930432. eCollection 2022.

Authors

Affiliations

¹ Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, United States.
² Department of Radiation Medicine, Oregon Health & Science University, Portland, OR, United States.
³ Medical Scientist Training Program, Oregon Health & Science University, Portland, OR, United States.
⁴ Department of Clinical Oncology, Menoufia University Shibin El Kom, Shibin El Kom, Egypt.
⁵ Department of Computer Science, Aalto University School of Science, Espoo, Finland.

Abstract

Background/purpose: Sarcopenia is a prognostic factor in patients with head and neck cancer (HNC). Sarcopenia can be determined using the skeletal muscle index (SMI) calculated from cervical neck skeletal muscle (SM) segmentations. However, SM segmentation requires manual input, which is time-consuming and variable. Therefore, we developed a fully-automated approach to segment cervical vertebra SM.

Materials/methods: 390 HNC patients with contrast-enhanced CT scans were utilized (300-training, 90-testing). Ground-truth single-slice SM segmentations at the C3 vertebra were manually generated. A multi-stage deep learning pipeline was developed, where a 3D ResUNet auto-segmented the C3 section (33 mm window), the middle slice of the section was auto-selected, and a 2D ResUNet auto-segmented the auto-selected slice. Both the 3D and 2D approaches trained five sub-models (5-fold cross-validation) and combined sub-model predictions on the test set using majority vote ensembling. Model performance was primarily determined using the Dice similarity coefficient (DSC). Predicted SMI was calculated using the auto-segmented SM cross-sectional area. Finally, using established SMI cutoffs, we performed a Kaplan-Meier analysis to determine associations with overall survival.

Results: Mean test set DSC of the 3D and 2D models were 0.96 and 0.95, respectively. Predicted SMI had high correlation to the ground-truth SMI in males and females (r>0.96). Predicted SMI stratified patients for overall survival in males (log-rank p = 0.01) but not females (log-rank p = 0.07), consistent with ground-truth SMI.

Conclusion: We developed a high-performance, multi-stage, fully-automated approach to segment cervical vertebra SM. Our study is an essential step towards fully-automated sarcopenia-related decision-making in patients with HNC.

Keywords: auto-segmentation; deep learning; head and neck cancer; sarcopenia; skeletal muscle index.

Abstract

Grants and funding