Color-coded visualization of magnetic resonance imaging multiparametric maps.

Area of Science:

• Medical Imaging
• Radiology
• Data Visualization

Background:

• Multiparametric MRI (mpMRI) is increasingly used for clinical diagnoses, such as prostate cancer.
• Conventional mpMRI visualization uses a 1D color scale, limiting information encoding.
• Human visual perception effectively uses 3D color spaces for information interpretation.

Purpose of the Study:

• To develop and evaluate a novel tri-variate color-coded visualization method for mpMRI data.
• To assess the potential of encoding more information from mpMRI datasets.
• To improve diagnostic performance by leveraging human visual perception of 3D color space.

Main Methods:

• Developed a new method for tri-variate color-coded visualization of mpMRI data.
• Applied the method to preclinical data (rat model of acute kidney injury).
• Evaluated diagnostic performance in a blinded clinical study of 13 prostate cancer mpMRI datasets with 5 observers.

Main Results:

• The novel visualization method produced characteristic patterns in preclinical kidney injury imaging.
• In prostate cancer mpMRI, the tri-variate color-coded visualization showed comparable positive and negative predictive values to conventional methods.
• Human observers successfully utilized the novel visualization technique.

Conclusions:

• The developed tri-variate color-coded visualization method is effective for mpMRI data.
• This technique allows for richer information encoding than conventional 1D scales.
• The method shows broad applicability for visualizing various multivariate MRI datasets and can aid in clinical diagnosis.