A comparison of two different software packages for analysis of body composition using computed tomography images

Rollins, Katie E.; Awwad, Amir; Macdonald, Ian A.; Lobo, Dileep N.

doi:10.1016/j.nut.2018.06.003

Cited by 23 publications

(14 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…23,24 A commonly used software tool for semi-automatic image segmentation is sliceOmatic, which uses pixel thresholding with region growing and has been shown to allow reliable assessment of body composition in studies. 25 The purpose of this study is to externally evaluate a new AI-based workflow, which we trained to automatically detect a predefined CT slice at the third lumbar vertebra (L3) and automatically perform complete image segmentation for analysis of body composition, and to compare its performance with that of an established threshold-based semi-automatic segmentation method in terms of speed and accuracy of tissue area calculation. The new AI-based software tool for assessment of body composition is fully integrated into the interface of a widely used picture archiving and communications system (PACS).…”

Section: Introductionmentioning

confidence: 99%

First PACS‐integrated artificial intelligence‐based software tool for rapid and fully automatic analysis of body composition from CT in clinical routine

Beetz

Maier

Segger

et al. 2021

JCSM Clinical Reports

View full text Add to dashboard Cite

BackgroundTo externally evaluate the first picture archiving communications system (PACS)‐integrated artificial intelligence (AI)‐based workflow, trained to automatically detect a predefined computed tomography (CT) slice at the third lumbar vertebra (L3) and automatically perform complete image segmentation for analysis of CT body composition and to compare its performance with that of an established semi‐automatic segmentation tool regarding speed and accuracy of tissue area calculation.MethodsFor fully automatic analysis of body composition with L3 recognition, U‐Nets were trained (Visage) and compared with a conventional image segmentation software (TomoVision). Tissue was differentiated into psoas muscle, skeletal muscle, visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT). Mid‐L3 level images from randomly selected DICOM slice files of 20 CT scans acquired with various imaging protocols were segmented with both methods.ResultsSuccess rate of AI‐based L3 recognition was 100%. Compared with semi‐automatic, fully automatic AI‐based image segmentation yielded relative differences of 0.22% and 0.16% for skeletal muscle, 0.47% and 0.49% for psoas muscle, 0.42% and 0.42% for VAT and 0.18% and 0.18% for SAT. AI‐based fully automatic segmentation was significantly faster than semi‐automatic segmentation (3 ± 0 s vs. 170 ± 40 s, P < 0.001, for User 1 and 152 ± 40 s, P < 0.001, for User 2).ConclusionRapid fully automatic AI‐based, PACS‐integrated assessment of body composition yields identical results without transfer of critical patient data. Additional metabolic information can be inserted into the patient's image report and offered to the referring clinicians.

show abstract

Section: Introductionmentioning

confidence: 99%

First PACS‐integrated artificial intelligence‐based software tool for rapid and fully automatic analysis of body composition from CT in clinical routine

Beetz

Maier

Segger

et al. 2021

JCSM Clinical Reports

View full text Add to dashboard Cite

show abstract

“…Finally, variations in density depending on acquisition factors, scanner machines and software have been described. 29 These limitations could be overcome by calculating the thrombus/blood density ratio to smooth out acquisition variations.…”

Section: Discussionmentioning

confidence: 99%

Post-EVAR Thrombus Density on Late Non-Contrast CT Scans Predicts Successful Aneurysm Exclusion

Rouer

Monnot

Fuda

et al. 2020

Vasc Endovascular Surg

View full text Add to dashboard Cite

Purpose: The incidence of type II endoleaks (ELII) after endovascular aneurysm repair (EVAR) ranges from 10–44%. Aneurysm thrombus density after EVAR could predict successful aneurysm exclusion. Materials and Methods: Twenty-seven patients with an abdominal aortic aneurysm (AAA) who had a CT scan within the first 45 days (early group) post-surgery or after 7 months (late group) were included. Thrombus density was analyzed on non-contrast enhanced CT scans. Results: A total of 5/13 (38%) patients in the early group had an ELII and 9/14 (64.3%) in the late group had a persistent ELII since surgery. In the early group, thrombus density was similar in patients with or without an ELII (mean: 39.9 ± 4.8 vs. 41.9 ± 3.4, p = 0.7; median: 38.7 ± 4.8 vs. 39.7 ± 3.1, p = 0.8). In patients with an ELII, there was no difference in thrombus density at 45 days and after 7 months (mean: 39.9 ± 4.8 vs. 40.2 ± 2.1, p = 0.9; median: 38.7 ± 4.8 vs. 38 ± 2.6, p = 0.9). In patients without an ELII, thrombus density was significantly higher at 45 days than after 7 months (mean: 41.9 ± 3.44 vs. 25.7 ± 2.0, p = 0.005; median: 39.7 ± 3.11 vs. 24.4 ± 1.5, p = 0.004). In patients with an ELII, thrombus density was significantly higher after 7 months than in patients without an ELII (mean: 40.2 ± 2.1 vs. 25.7 ± 2.0. p = 0.001; median: 38 ± 2.6 vs. 24.4 ± 1.5, p = 0.003). Conclusion: Low thrombus density after EVAR on late unenhanced CT scans predicts aneurysm exclusion.

show abstract

“…However, care must be taken to ensure that the correct image is analyzed because of the high resolution of many scanners capturing multiple images within L3 vertebra. Each software can also provide differing results for the same image, so care must be taken to ensure standardization . Because of these drawbacks, at this point, additional large clinical trials are necessary prior to widespread adoption of this technology in critically ill patients, and we would recommend that most readers work with radiology colleagues at their institute prior to implementing in their practice.…”

Section: Different Modalities For Body Composition Assessmentmentioning

confidence: 99%

Body Composition Technology: Implications for the ICU

2018

View full text Add to dashboard Cite

Malnutrition continues to be highly prevalent in hospitalized and critically ill patients and is associated with significant morbidity and mortality. Additionally, survivors of critical illness have an increased risk for sarcopenia, which leads to weakness and physical debilitation that can persist for years. Nutrition risk assessment tools have been developed and validated in critically ill patients but have limitations. Variables such as body weight, body mass index, weight change, or percentage of food intake can be difficult to obtain in critically ill patients and may be misleading given changes in body composition, such as an increase in body water. Assessment of body composition through new techniques provides a unique opportunity to counter some of these limitations and develop improved methods of nutrition risk assessment based on objective data. The present manuscript provides a review of the most commonly available clinical technology for assessment of body composition (bioimpedance, computed tomography, and ultrasound), including data from trials in critically ill patients highlighting the benefits and weaknesses of each modality.

show abstract

A comparison of two different software packages for analysis of body composition using computed tomography images

Abstract: All four body composition measures were statistically significantly different by the software package used for analysis; however, the clinical significance of these differences is doubtful. Nevertheless, the same software package should be used if serial measurements are being performed.

Cited by 23 publications

References 10 publications

First PACS‐integrated artificial intelligence‐based software tool for rapid and fully automatic analysis of body composition from CT in clinical routine

First PACS‐integrated artificial intelligence‐based software tool for rapid and fully automatic analysis of body composition from CT in clinical routine

Post-EVAR Thrombus Density on Late Non-Contrast CT Scans Predicts Successful Aneurysm Exclusion

Body Composition Technology: Implications for the ICU

Contact Info

Product

Resources

About