Three Material Decomposition using a Triple-Layer Flat-Panel Detector

Value Proposition:

• Method for direct, real-time visualization of bone, water, and iodine contrast agents.
• Allows for better visualization of moving structures; reduces imaging artifacts induced by motion (i.e. beating heart)

 Technology Description

• Researchers at Johns Hopkins have developed a method for fast processing of data acquired using a triple-layer flat-panel detector (TL-FPD) for enhanced visualization and interpretation of radiographic or fluoroscopy images.
• This technology includes a machine learning model for decomposition of imaging data into bone, water, or iodine.

Unmet Need

• Current methods for standard radiography and fluoroscopy presents overlapping anatomy, which obsures visualization of critical features. Additionally, contrast agent is not sufficient 
•  Spectral radiography acquisition via dual-energy imaging are unable to fully separate iodine contrast agents from calcium.
• Additionally, current contrast visualization methods, such as digital subtraction angiography (DSA), are sensitive to motion and can lead to artifacts that confound image processing.
• Triple layer detector is a promising acquistion strategy to address these issues but requires an accurate, real-time (for angiography applications) processing algorithm for material decomposition 

Stage of Development

• Machine-learning model for image processing was developed for fast, accurate material decomposition of data acquired with a triple layer flat panel detector
• Looking for partners to commercialize the technology for integration into radiographic or fluoroscopy-based diagnostic and interventional imaging.

 Data Availability

• Data available upon request.

Publication

N/A