Hemodynamic property incorporated brain tumor segmentation by deep learning and density-based analysis of dynamic susceptibility contrast-enhanced magnetic resonance imaging (MRI).

Background: Magnetic resonance imaging (MRI) is a primary non-invasive imaging modality for tumor segmentation, leveraging its exceptional soft tissue contrast and high resolution. Current segmentation methods typically focus on structural MRI, such as T1-weighted post-contrast-enhanced or fluid-attenuated inversion recovery (FLAIR) sequences. However, these methods overlook the blood perfusion and hemodynamic properties of tumors, readily derived from dynamic susceptibility contrast (DSC) enhanced MRI. This study introduces a novel hybrid method combining density-based analysis of hemodynamic properties in time-dependent perfusion imaging with deep learning spatial segmentation techniques to enhance tumor segmentation. Methods: First, a U-Net convolutional neural network (CNN) is employed on structural images to delineate a region of interest (ROI). Subsequently, Hierarchical Density-Based Scans (HDBScan) are employed within the ROI to augment segmentation by exploring intratumoral hemodynamic heterogeneity through the investigation of tumor time course profiles unveiled in DSC MRI. Results: The approach was tested and evaluated using a cohort of 513 patients from the open-source University of Pennsylvania glioblastoma database (UPENN-GBM) dataset, achieving a 74.83% Intersection over Union (IoU) score when compared to structural-only segmentation. The algorithm also exhibited increased precision and localized predictions of heightened segmentation boundary complexity, resulting in a 146.92% increase in contour complexity (ICC) compared to the reference standard provided by the UPENN-GBM dataset. Importantly, segmenting tumors with the developed new approach uncovered a negative correlation of the tumor volume with the scores in the Karnofsky Performance Scale (KPS) clinically used for assessing the functional status of patients (-0.309), which is not observed with the prevailing segmentation standard. Conclusions: This work demonstrated that including hemodynamic properties of tissues from DSC MRI can improve existing structural or morphological feature-based tumor segmentation techniques with additional information on tumor biology and physiology. This approach can also be applied to other clinical indications that use perfusion MRI for diagnosis or treatment monitoring.

Label-Free Chemically and Molecularly Selective Magnetic Resonance Imaging.

Biomedical imaging, especially molecular imaging, has been a driving force in scientific discovery, technological innovation, and precision medicine in the past two decades. While substantial advances and discoveries in chemical biology have been made to develop molecular imaging probes and tracers, translating these exogenous agents to clinical application in precision medicine is a major challenge. Among the clinically accepted imaging modalities, magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) exemplify the most effective and robust biomedical imaging tools. Both MRI and MRS enable a broad range of chemical, biological and clinical applications from determining molecular structures in biochemical analysis to imaging diagnosis and characterization of many diseases and image-guided interventions. Using chemical, biological, and nuclear magnetic resonance properties of specific endogenous metabolites and native MRI contrast-enhancing biomolecules, label-free molecular and cellular imaging with MRI can be achieved in biomedical research and clinical management of patients with various diseases. This review article outlines the chemical and biological bases of several label-free chemically and molecularly selective MRI and MRS methods that have been applied in imaging biomarker discovery, preclinical investigation, and image-guided clinical management. Examples are provided to demonstrate strategies for using endogenous probes to report the molecular, metabolic, physiological, and functional events and processes in living systems, including patients. Future perspectives on label-free molecular MRI and its challenges as well as potential solutions, including the use of rational design and engineered approaches to develop chemical and biological imaging probes to facilitate or combine with label-free molecular MRI, are discussed.