Post-Acquisition Hyperpolarized 29Silicon Magnetic Resonance Image Processing for Visualization of Colorectal Lesions Using a User-Friendly Graphical Interface

Abstract

Medical imaging devices often use automated processing that creates and displays a self-normalized image. When improperly executed, normalization can misrepresent information or result in an inaccurate analysis. In the case of diagnostic imaging, a false positive in the absence of disease, or a negative finding when disease is present, can produce a detrimental experience for the patient and diminish their health prospects and prognosis. In many clinical settings, a medical technical specialist is trained to operate an imaging device without sufficient background information or understanding of the fundamental theory and processes involved in image creation and signal processing. Here, we describe a user-friendly image processing algorithm that mitigates user bias and allows for true signal to be distinguished from background. For proof-of-principle, we used antibody-targeted molecular imaging of colorectal cancer (CRC) in a mouse model, expressing human MUC1 at tumor sites. Lesion detection was performed using targeted magnetic resonance imaging (MRI) of hyperpolarized silicon particles. Resulting images containing high background and artifacts were then subjected to individualized image post-processing and comparative analysis. Post-acquisition image processing allowed for co-registration of the targeted silicon signal with the anatomical proton magnetic resonance (MR) image. This new methodology allows users to calibrate a set of images, acquired with MRI, and reliably locate CRC tumors in the lower gastrointestinal tract of living mice. The method is expected to be generally useful for distinguishing true signal from background for other cancer types, improving the reliability of diagnostic MRI.

Keywords: GUI; MRI; colorectal cancer; diagnostic imaging; hyperpolarization; image processing; silicon particles.

Figure 1

Spurious signal and artifacts distort apparent signal intensity. Three in vivo studies (a–c), with overlaid illustration of mouse anatomy, show examples of issues and artifacts that misrepresent relevant signal for targeted studies. (i) Spilled or leaked particles show concentrated signal outside of the mouse. (ii) Leftover particles in syringe distort the signal intensity of targeted particles. (iii) Background noise due to day-to-day variations with the MRI can distort expected signal-to-noise ratios. (iv) RF Overflow artifacts can cause distortion in visualization of signal strength.

Figure 2

Comparing signal intensity across studies. (a) Two separate studies where maximum signal intensity is displayed according to the highest value for each respective study. (b) The same studies after post-processing to display maximum signal intensity normalized using the highest signal measured across all studies after individual normalization of static background per study. (c) The adjusted signal intensity after co-registration with anatomical hydrogen MRI.

Figure 3

Processing artifacts skew displayed intensity. False signal in raw silicon images from two mice (a,d), indicated by arrows superimposed on anatomical image (b,e), is still present after using a standard threshold for noise reduction. Noise intensity may change between studies due to day-to-day variation in the magnetic field. After normalization for noise levels present for individual studies, the false signal is largely eliminated (c,f).

Figure 4

Quantitative analysis shows processed signal registers to tumor location. Analysis of SNR (a) and CNR (b) for four experimental groups using ROIs drawn for mouse body (c), tumor locations (d), and non-tumor locations (e) of the same area as those from (d). Tumor locations were determined with anatomical MRI and were confirmed following excision of the colonic tissue. n = 3 for each cohort. The star indicates the example study (c–e) data point after analysis [22].

Figure 5

Post-processing reveals true signal. Mice were anesthetized prior to being imaged in a 7 T small animal research MRI. Imaging was performed using coronal slices with the mouse in a supine position. (a) Pre-processed silicon signal. Spurious signal obfuscates ability to detect true tumor sites. (b) Post-processed image of the same study, co-registered with anatomical imaging. True signal becomes evident after processing to reduce artifact-associated signal. Color bar in arbitrary units of highest intensity.