Muscle segmentation in axial computed tomography (CT) images at the lumbar (L3) and thoracic (T4) levels for body composition analysis

Dabiri, Setareh; Popuri, Karteek; Feliciano, Elizabeth M. Cespedes; Caan, Bette J.; Baracos, Vickie E.; Beg, Mirza Faisal

doi:10.1016/j.compmedimag.2019.04.007

Cited by 75 publications

(65 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The validated technique uses CT-based analysis at the L3 level, as this was the level that the initial validation calculations were performed in order to extrapolate to the whole body. Recently, several studies have looked at body composition analysis at the fourth thoracic vertebra as an alternative in patients who are undergoing a thoracic rather than abdominal procedure [34].…”

Section: Sarcopeniamentioning

confidence: 99%

Perioperative nutrition: Recommendations from the ESPEN expert group

et al. 2020

View full text Add to dashboard Cite

Background & aims: Malnutrition has been recognized as a major risk factor for adverse postoperative outcomes. The ESPEN Symposium on perioperative nutrition was held in Nottingham, UK, on 14e15 October 2018 and the aims of this document were to highlight the scientific basis for the nutritional and metabolic management of surgical patients. Methods: This paper represents the opinion of experts in this multidisciplinary field and those of a patient and caregiver, based on current evidence. It highlights the current state of the art. Results: Surgical patients may present with varying degrees of malnutrition, sarcopenia, cachexia, obesity and myosteatosis. Preoperative optimization can help improve outcomes. Perioperative fluid therapy should aim at keeping the patient in as near zero fluid and electrolyte balance as possible. Similarly, glycemic control is especially important in those patients with poorly controlled diabetes, with a stepwise increase in the risk of infectious complications and mortality per increasing HbA1c. Immobilization can induce a decline in basal energy expenditure, reduced insulin sensitivity, anabolic resistance to protein nutrition and muscle strength, all of which impair clinical outcomes. There is a role for pharmaconutrition, pre-, pro-and syn-biotics, with the evidence being stronger in those undergoing surgery for gastrointestinal cancer.

show abstract

Section: Sarcopeniamentioning

confidence: 99%

Perioperative nutrition: Recommendations from the ESPEN expert group

et al. 2020

View full text Add to dashboard Cite

show abstract

“…2 Therefore, various studies have attempted to solve this dilemma by conducting CT scans at different vertebra levels (T4, L1) for SMM estimates in NSCLC patients. 5,6,21,22 Although good results have been achieved, the applications were not so extensive. Herein, an accurate and reproducible classifier was constructed through the integration of a large panel of skeletal muscle CT radiomic features and high efficiency lightGBM model, to differentiate NSCLC patients with sarcopenia from those without sarcopenia, achieving high accuracy and AUC were recorded as 0.900 and 0.889 in optimal lightGBM mode.…”

Section: Discussionmentioning

confidence: 99%

Identifying sarcopenia in advanced non‐small cell lung cancer patients using skeletal muscle CT radiomics and machine learning

Dong

et al. 2020

Thoracic Cancer

View full text Add to dashboard Cite

Background Sarcopenia has been confirmed as a poor prognostic indicator of lung cancer. However, the lack of abdominal computed tomography (CT) hindered the application to assess the status of sarcopenia. The purpose of this study was to assess the ability of chest CT radiomics combined with machine learning classifiers to identify sarcopenia in advanced non‐small cell lung cancer (NSCLC) patients. Methods This study retrospectively analyzed CT images of 99 patients with NSCLC. Skeletal muscle radiomics were extracted from a single axial slice of the chest CT scan at the 12th thoracic vertebrae level. In total, 854 radiomic and clinical features were obtained from each patient. Feature selection was conducted with FeatureSelector module, optimal key features were fed into the lightGBM classifier for model construction, and Bayesian optimization was adopted to tune hyperparameters. The model's performance was evaluated by specificity, sensitivity, accuracy, precision, F1‐score, Matthew's correlation coefficient (MCC), Cohen's kappa coefficient (Kappa), and AUC. Results A total of 40 patients were found to have sarcopenia. Five optimal features were selected. In the base lightGBM model, the specificity, sensitivity, accuracy, precision, F1‐score, AUC, MCC, Kappa of validation set were 0.889, 0.750, 0.833, 0.818, 0.783, 0.819, 0.649, 0.648, respectively. After Bayesian hyperparameter tuning, the optimized lightGBM model achieved better prediction performance, and the corresponding values were 0.944, 0.833, 0.900, 0.909, 0.870, 0.889, 0.791, 0.789, respectively. Conclusions Chest CT‐based radiomics has the potential to identify sarcopenia in NSCLC patients with the lightGBM classifier, and the optimal lightGBM model via Bayesian hyperparameter tuning demonstrated better performance. Key points Significant findings of the study Our study demonstrates that chest CT‐based radiomics combined with lightGBM classifier has the ability to identify sarcopenia in NSCLC patients. What this study adds Skeletal muscle radiomics would be a potential biomarker for sarcopenia identity in NSCLC patients.

show abstract

“…The prospect of muscle segmentation on volumetric CT imaging using deep learning algorithms provides exciting opportunity for further work in this area and may overcome many of the challenges in sarcopenia measurement, improving precision and validity [17,18]. The trauma population is unique in the challenges it imposes given the heterogeneity of injuries in icted-in terms of severity, quantity and distribution of affected body areas.…”

Section: Discussionmentioning

confidence: 99%

Sarcopenia in Elderly Major Trauma; A Single Centre Retrospective Analysis of Psoas and Masseter Muscle Groups

Varma

Wilson

Naik

et al. 2020

Preprint

View full text Add to dashboard Cite

Background: There is an emerging role for the radiological evaluation of the psoas muscle as a marker of sarcopenia, and as a prognostic discriminant in elderly patients with traumatic injuries. Older trauma patients are more likely to undergo cranial than abdomino-pelvic imaging. Identifying sarcopenia using masseter cross sectional area (M-CSA) has shown correlation with mortality. We sought to determine the correlation between psoas: lumbar vertebral index (PLVI) and the M-CSA, and their association with health outcomes. Methods: Patients aged 65 or above, who presented as a trauma call over a 1 year period were included if they underwent cranial or abdominal CT imaging. Images were retrospectively analyzed to obtain PLVI and mean M-CSA measurements. Electronic records were abstracted for demographics and outcomes. Logistic regression methods, log scale analyses, Cox regression model and Kaplan-Meier plots were used to determine association of sarcopenia with outcomes. Results: There were 155 eligible patients in the M-CSA group and 204 patients in the PLVI group. Sarcopenia was defined as the lowest quartile in each group. Both PLVI and M-CSA measurements were available in 142 patients. Pearson’s correlation indicated a weakly positive linear relationship (r=0.35, p<0.001) between these. There was no statistical association between M-CSA sarcopenia status and any measured outcomes. Those with PLVI sarcopenia were more likely to die in hospital (adjusted OR 3.38, 95% CI 1.47-9.73, p=0.006) and at 2-years (adjusted HR 1.90, 95% CI 1.11-3.25, p=0.02). Only 29% patients with PLVI sarcopenia were discharged home, compared with 58% without sarcopenia (p=0.001).Conclusion: Sarcopenia, defined by PLVI, is predictive of increased in-patient and 2- year mortality. Our study did not support prognostic relevance of M-CSA. Further research should be directed at improving the validity of masseter measurements or identifying alternative radiological determinants of sarcopenia on cranial imaging.

show abstract

Muscle segmentation in axial computed tomography (CT) images at the lumbar (L3) and thoracic (T4) levels for body composition analysis

Cited by 75 publications

References 25 publications

Perioperative nutrition: Recommendations from the ESPEN expert group

Perioperative nutrition: Recommendations from the ESPEN expert group

Identifying sarcopenia in advanced non‐small cell lung cancer patients using skeletal muscle CT radiomics and machine learning

Sarcopenia in Elderly Major Trauma; A Single Centre Retrospective Analysis of Psoas and Masseter Muscle Groups

Contact Info

Product

Resources

About