Opportunistic bone density screening for the abdominal radiologist using colored CT images: a pilot retrospective...

Area of Science:

• Radiology
• Medical Imaging
• Bone Densitometry

Background:

• Osteoporosis is a significant public health concern, often diagnosed using dual-energy X-ray absorptiometry (DXA).
• Abdominal CT scans are widely available and can potentially be utilized for bone density assessment.
• Developing non-invasive methods to detect low bone density from existing CT scans is crucial for early diagnosis and intervention.

Purpose of the Study:

• To develop and validate a precise and reproducible method for detecting reduced spinal bone density from abdominal CT images.
• To establish an optimal threshold for differentiating normal from low bone density using CT attenuation values.

Main Methods:

• A retrospective study analyzed nonenhanced abdominal CT images from 631 participants.
• Mean L3/L4 vertebral body attenuation was correlated with quantitative CT bone density (QCT) in a training cohort.
• Custom software generated grayscale and color-coded CT images, and five radiologists assessed bone density in a validation cohort using QCT, grayscale, attenuation, and color methods.

Main Results:

• An optimal mean attenuation threshold of 145 Hounsfield Units was identified for differentiating normal from low bone density.
• The color-coded image method achieved 91% accuracy, significantly outperforming grayscale (58%) and attenuation measurement (87%) methods (p < 0.001).
• Inter-observer agreement was excellent for the attenuation (ICC: 0.85) and color methods (ICC: 0.87), but poor for grayscale (ICC: 0.20).

Conclusions:

• Colored abdominal CT imaging provides a highly accurate and reproducible means for detecting low spinal bone density.
• This technique offers a promising, non-invasive approach to identify individuals at risk for osteoporosis using readily available CT scans.