Nondestructive Ultrasound Molecular Imaging With Higher Order Singular Value Decomposition

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Ultrasound Technology

Background:

• Ultrasound molecular imaging (UMI) relies on targeted microbubbles (MBs) to detect biomarkers.
• Distinguishing bound MBs from free MBs and tissue is crucial for accurate UMI.
• Current methods like differential Targeted Enhancement (DTE) are time-intensive and destructive.

Purpose of the Study:

• Introduce a novel, rapid, and non-destructive UMI technique.
• Utilize higher-order singular value decomposition (HOSVD) for signal separation.
• Develop a method to accurately identify bound MBs for improved biomarker detection.

Main Methods:

• HOSVD decomposes acoustic contrast sequences based on nonlinear content and temporal coherence.
• Nonlinear separation distinguishes tissue from MBs; temporal separation differentiates free from bound MBs.
• A bound MB indicator (χ) was defined from HOSVD output.

Main Results:

• HOSVD effectively separated signals, enabling distinction between free, bound MBs, and tissue.
• The bound MB indicator (χ) showed significantly higher values for bound MBs compared to free MBs and tissue.
• The HOSVD method demonstrated superior molecular signal enhancement (12 dB) and improved true/false positive detection rates compared to other non-destructive techniques.

Conclusions:

• HOSVD offers a rapid, non-destructive approach for UMI.
• This technique enhances the accuracy and efficiency of biomarker detection using targeted MBs.
• HOSVD has the potential to advance the clinical application of UMI.