Pathologic changes to brain tumors occurring over time provide a wealth of information to physicians treating the affliction. Researchers at the Mayo Clinic, Rochester and the University of Southern California have developed automated techniques to better document such changes.
Pathologic changes to brain tumors occurring over time provide a wealth of information to physicians treating the affliction. Researchers at the Mayo Clinic, Rochester and the University of Southern California have developed automated techniques to better document such changes.
Monitoring changes in brain tumor pathology requires periodic imaging examinations, providing the radiologist with enormous amounts of MR- or CT-generated data. Much of it is ambiguous in terms of which regions have changed, in what way, and to what degree.
A paper by researchers at the Mayo Clinic details the motivating factors for the production of an automated process to address this problem (J Digital Imaging, epub 29 June 2004).
Techniques that separate acquisition-related change from disease-related change, reduce the quantity of data presented to the radiologist, and produce objective, reproducible, and accurate metrics of disease course have great value in brain scan interpretation, said Julia Patriarche, Ph.D., of the radiology department at Mayo.
Patriarche and colleagues presented the most common approaches to change detection, ranging from manual inspection, measurement sampling, and volumetrics to warping and temporal analysis.
The paper suggests a number of technical approaches for separating acquisition-related changes from pathology-related changes:
Another approach to brain tumor change detection surfaced at the University of Southern California, where the second generation of a system that synthesizes brain imaging data gathered from electroencephalograms, magnetoencephalograms, and functional MR scans of brain activity patterns has completed alpha testing.
The software, Brainstorm 2.0, is available for free at http://neuroimage.usc.edu/brainstorm/. The program presents images as maps, slides, or even movies showing changes at a fine-grained (millisecond) time scale. It also offers detailed 3D renderings of brain surfaces.
The new version improves on version 1.0 through the addition of more algorithms, including ones to create virtual electrodes at designated listening points inside the brain. It also has useful data management features that enable users to easily index, search, and classify EEG, MEG, and MRI records.
Brainstorm, written in the widely used MatLab scientific programming language, allows users to quickly import data recorded in various widely used commercial imaging programs.
Brainstorm 2.0 appears to be popular in academia. In recent months, more than 300 copies of the alpha version have been downloaded.
MRI-Based AI Radiomics Model Offers 'Robust' Prediction of Perineural Invasion in Prostate Cancer
July 26th 2024A model that combines MRI-based deep learning radiomics and clinical factors demonstrated an 84.8 percent ROC AUC and a 92.6 percent precision-recall AUC for predicting perineural invasion in prostate cancer cases.
Breast MRI Study Examines Common Factors with False Negatives and False Positives
July 24th 2024The absence of ipsilateral breast hypervascularity is three times more likely to be associated with false-negative findings on breast MRI and non-mass enhancement lesions have a 4.5-fold likelihood of being linked to false-positive results, according to new research.
Can Polyenergetic Reconstruction Help Resolve Streak Artifacts in Photon Counting CT?
July 22nd 2024New research looking at photon-counting computed tomography (PCCT) demonstrated significantly reduced variation and tracheal air density attenuation with polyenergetic reconstruction in contrast to monoenergetic reconstruction on chest CT.