What is the clinical evidence from this study?

Study design: Observational. Population: Overweight and obesity (n=920). Intervention: ACD 3D U-Net and 3D nnU-Net vs. Manual reference segmentation and 3D U-Net. Primary outcome: Median DICE-SAT and DICE-VAT scores.

February 1, 2024Open Access

Automated abdominal adipose tissue segmentation and volume quantification on longitudinal MRI using 3D convolutional neural networks with multi-contrast inputs

Q: What does this research mean for the field?

ACD 3D U-Net and 3D nnU-Net provide rapid and highly accurate automated segmentation of abdominal subcutaneous and visceral adipose tissue on longitudinal MRI. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Key Result

ACD 3D U-Net and 3D nnU-Net provided rapid and accurate automated segmentation of abdominal subcutaneous and visceral adipose tissue on longitudinal MRI, achieving median Dice scores ≥0.976 and ICC >0.997.

Study Design

Type

Observational (n=920)

Multicenter

Yes

Structured PICO

Does automated abdominal SAT/VAT segmentation using 3D convolutional neural networks accurately quantify adipose tissue volume compared to reference measurements in adults with overweight/obesity?

Population

920 adults with overweight/obesity scanned twice at multiple 3 T MRI scanners and institutions

Intervention

Automated abdominal SAT/VAT segmentation using attention-based competitive dense (ACD) 3D U-Net and 3D nnU-Net with full field-of-view volumetric multi-contrast inputs

Comparator

Reference measurements

Outcome

Segmentation performance (Dice scores, false negatives, false positives) and volume agreement (intraclass correlation coefficient)surrogate

ACD 3D U-Net and 3D nnU-Net provide rapid and accurate automated quantification of abdominal SAT and VAT volumes on longitudinal MRI in adults with overweight/obesity.

Limitations

High memory requirement due to full-FOV volumetric inputs
Retrospective use of data with a specific MRI sequence and set of imaging parameters
Longitudinal data only had two time points separated by 6 months

Abstract

OBJECTIVE: Increased subcutaneous and visceral adipose tissue (SAT/VAT) volume is associated with risk for cardiometabolic diseases. This work aimed to develop and evaluate automated abdominal SAT/VAT segmentation on longitudinal MRI in adults with overweight/obesity using attention-based competitive dense (ACD) 3D U-Net and 3D nnU-Net with full field-of-view volumetric multi-contrast inputs. MATERIALS AND METHODS: 920 adults with overweight/obesity were scanned twice at multiple 3 T MRI scanners and institutions. The first scan was divided into training/validation/testing sets (n = 646/92/182). The second scan from the subjects in the testing set was used to evaluate the generalizability for longitudinal analysis. Segmentation performance was assessed by measuring Dice scores (DICE-SAT, DICE-VAT), false negatives (FN), and false positives (FP). Volume agreement was assessed using the intraclass correlation coefficient (ICC). RESULTS: ACD 3D U-Net achieved rapid ( 0.997) in longitudinal analysis. DISCUSSION: ACD 3D U-Net and 3D nnU-Net can be automated tools to quantify abdominal SAT/VAT volume rapidly, accurately, and longitudinally in adults with overweight/obesity.

AI से पूछें

Bookmark

View Full Paper

Cite This Study

Kafalı et al. (Thu,) conducted a observational in Overweight and obesity (n=920). ACD 3D U-Net and 3D nnU-Net vs. Manual reference segmentation and 3D U-Net was evaluated on Median DICE-SAT and DICE-VAT scores. ACD 3D U-Net and 3D nnU-Net provided rapid and accurate automated segmentation of abdominal subcutaneous and visceral adipose tissue on longitudinal MRI, achieving median Dice scores ≥0.976 and ICC >0.997.

synapsesocial.com/papers/6a14e40e253bd9cd3ce6281f https://doi.org/https://doi.org/10.1007/s10334-023-01146-3

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

AI से पूछें

Bookmark

View Full Paper