Poster Poster Program Diagnostic and Interventional Radiology Physics

Full-Range Dose Automatic Phantom-Less QCT Bone Mineral Density Calibration Algorithm: Development and Validation

Abstract
Purpose

To propose a full-range dose automatic Phantom-less QCT bone mineral density (BMD) calibration algorithm, aiming to address the increased BMD measurement error caused by non-linear changes in CT values under different CT scan doses.

Methods

1.) Segmentation model training: 200 spinal CT selected from CTSpine1K, TotalSegmentator was used to segment the thoracolumbar T11-L2 segments and locate the vertebral centroid. CT slices of paravertebral muscles and subcutaneous fat at the central level of the vertebrae were extracted, reviewed and annotated by radiologists, and then used to train a UNet muscle-fat segmentation model. 2.) Classification of scanning parameters: Thoracolumbar CT data covering the full dose range from 3 CT devices were collected, totally 10,189 CTs. The Qwen2.5-VL model was utilized to extract dose reports, which were automatically clustered by K-means according to CTDI intervals (0-4, 4-8, 8-12, 12-16 mGy) combined with tube voltage and tube current, resulting in 11 common scan parameter combinations. 3.) Dataset division: For each parameter combination, patient and ESP scan data were divided into training set and test set at a ratio of 8:2. 4.) Training and validation: a.) Training: UNet segments muscles and fat and calculates their CT values. Combined with ESP data, the calibrated curve with phantom was fitted to obtain the reference densities of muscles and fat under each parameter. b.) Validation: BMD errors were calculated on ESP test dataset with patient-specific phantom-less model, compared with traditional 120kV QCT.

Results

The DSC of the 2D UNet for muscle and fat segmentation were 0.94 and 0.92. BMD validation results showed MAE under all scanning parameters were 8.42mg/cm3(ours) v.s. 23.07mg/cm3(traditional QCT).

Conclusion

This method can achieve accurate BMD calibration under various common CT scan parameter doses without additional phantom scans, and can utilize low-dose chest CT for physical examination to screen for osteoporosis.

People

Related

Similar sessions

Poster Poster Program
Jul 19 · 07:00
B-Trac – Breast Tissue Rotation and Compression Apparatus for Calibration

Mammography (compressed 2D) and MRI (uncompressed 3D) capture breast tissue under different conditions, complicating tumor localization across modalities. To bridge this gap, we developed a customizable physical platform to simul...

Dayadna Hernandez Perez
Diagnostic and Interventional Radiology Physics 0 people interested
Poster Poster Program
Jul 19 · 07:00
Comprehensive Medical Physics Assessment of Digital Mammography Equipment: A Three-Year Multi-Site Evaluation of Technical Performance and Radiation Safety at 24 Saudi Arabian Healthcare Institutions (2022–2024)

To conduct a comprehensive multi-center audit evaluating the technical performance, image quality, and radiation safety of digital mammography systems across 24 unique healthcare facilities in Saudi Arabia. This study aims to est...

Sami Alshaikh, PhD
Diagnostic and Interventional Radiology Physics 0 people interested
Poster Poster Program
Jul 19 · 07:00
Starting Small: Implementing a CT Protocol Optimization Program

This talk describes our organization’s CT optimization program, and how we implemented it to make efficient use of limited physicist time.

Robert J. Cropp, PhD
Diagnostic and Interventional Radiology Physics 0 people interested