Volume estimates by imaging methods: model comparisons with visible woman as the reference

Area of Science:

• Medical Imaging
• Anatomical Modeling
• Body Composition Analysis

Background:

• Accurate estimation of tissue and organ volumes is essential for medical imaging and body composition analysis.
• Existing volume estimation models vary in their accuracy and applicability.

Purpose of the Study:

• To compare the accuracy of four distinct volume estimation models against actual tissue and organ volumes.
• To determine the most reliable model for calculating body compartment volumes from medical imaging data.

Main Methods:

• Utilized 1-mm thick Visible Woman images segmented for five major components, including adipose tissue.
• Calculated actual volumes from 1730 slices and compared them with four models (truncated cone/pyramid vs. two-column/parallel trapezium).
• Systematically varied parameters like between-slice interval and initial slice to assess model performance.

Main Results:

• The two-column model consistently yielded results identical to reference volumes across all compartments and intervals.
• The truncated cone model produced volumes smaller than reference volumes.
• The two-column model demonstrated a lower coefficient of variation compared to the truncated cone model.

Conclusions:

• The parallel trapezium and two-column models provide more accurate tissue volume estimations than truncated pyramid and cone models.
• These findings have significant implications for improving the accuracy of imaging-based body compartment volume calculations, particularly for adipose tissue.