Quantifying T2-FLAIR Mismatch Using Geographically Weighted Regression and Predicting Molecular Status in Lower-Grade Gliomas

Mohammed, Shariq; Ravikumar, Visweswaran; Warner, Elisa; Patel, Sohil H.; Bakas, Spyridon; Rao, Arvind; Jain, Rajan

doi:10.3174/ajnr.a7341

Cited by 13 publications

(6 citation statements)

References 27 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The two strongest predictors of IDH mutation are T2_O_texture_1-Imc2 and FLAIR_W-LLH_texture-4_DependenceVariance. From previously published work, combining visual inspection of T2-weighted and FLAIR images for developing glioma diagnostic models has performed well in the clinic [ 50 ], and has been validated with computational quantitative analysis [ 51 ]. Our results provide indirect support of this finding and lay the foundation for further explainability of these radiomic features and clinical translation.…”

Section: Resultsmentioning

confidence: 99%

A Pipeline for the Implementation and Visualization of Explainable Machine Learning for Medical Imaging Using Radiomics Features

Severn

Suresh

Görg

et al. 2022

Sensors

View full text Add to dashboard Cite

Machine learning (ML) models have been shown to predict the presence of clinical factors from medical imaging with remarkable accuracy. However, these complex models can be difficult to interpret and are often criticized as “black boxes”. Prediction models that provide no insight into how their predictions are obtained are difficult to trust for making important clinical decisions, such as medical diagnoses or treatment. Explainable machine learning (XML) methods, such as Shapley values, have made it possible to explain the behavior of ML algorithms and to identify which predictors contribute most to a prediction. Incorporating XML methods into medical software tools has the potential to increase trust in ML-powered predictions and aid physicians in making medical decisions. Specifically, in the field of medical imaging analysis the most used methods for explaining deep learning-based model predictions are saliency maps that highlight important areas of an image. However, they do not provide a straightforward interpretation of which qualities of an image area are important. Here, we describe a novel pipeline for XML imaging that uses radiomics data and Shapley values as tools to explain outcome predictions from complex prediction models built with medical imaging with well-defined predictors. We present a visualization of XML imaging results in a clinician-focused dashboard that can be generalized to various settings. We demonstrate the use of this workflow for developing and explaining a prediction model using MRI data from glioma patients to predict a genetic mutation.

show abstract

Section: Resultsmentioning

confidence: 99%

A Pipeline for the Implementation and Visualization of Explainable Machine Learning for Medical Imaging Using Radiomics Features

Severn

Suresh

Görg

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Future studies on the quantification and automated detection of partial T2-FLAIR mismatch may allow more objective identification of this sign. For example, geographically weighted regression has been shown to accurately identify T2-FLAIR mismatch in lower-grade gliomas 21 and may potentially be extendable to grade 4 gliomas.…”

Section: Discussionmentioning

confidence: 99%

Association of partial T2-FLAIR mismatch sign and isocitrate dehydrogenase mutation in WHO grade 4 gliomas: results from the ReSPOND consortium

Lee,

Patel,

Mohan

et al. 2023

Neuroradiology

Self Cite

View full text Add to dashboard Cite

show abstract

“…Typically, they do not show contrast enhancement and have low relative cerebral blood volume (rCBV) on dynamic T2* perfusion-weighted MRI sequences. Some show the T2-fluid-attenuated inversion recovery (FLAIR) mismatch sign ( 4 ) , which is characteristic of, although not exclusive to, this histological type. High-grade (grade 3 and 4) astrocytomas can show areas of contrast enhancement and high rCBV.…”

Section: Adult-type Diffuse Gliomasmentioning

confidence: 99%

The 2021 World Health Organization classification of gliomas: an imaging approach

Fernandes,

Teixeira,

Mamere

et al. 2023

Radiol Bras

View full text Add to dashboard Cite

The purpose of this pictorial essay is to describe the recommendations of the 2021 World Health Organization classification for adult-type and pediatric-type gliomas and to discuss the main modifications in relation to the previous (2016) classification, exemplified by imaging, histological, and molecular findings in nine patients followed at our institutions. In recent years, molecular biomarkers have gained importance in the diagnosis and classification of gliomas, mainly because they have been shown to correlate with the biological behavior and prognosis of such tumors. It is important for neuroradiologists to familiarize themselves with this new classification of central nervous system tumors, so that they can use this knowledge in evaluating and reporting the imaging examinations of patients with glioma.

show abstract

Quantifying T2-FLAIR Mismatch Using Geographically Weighted Regression and Predicting Molecular Status in Lower-Grade Gliomas

Cited by 13 publications

References 27 publications

A Pipeline for the Implementation and Visualization of Explainable Machine Learning for Medical Imaging Using Radiomics Features

A Pipeline for the Implementation and Visualization of Explainable Machine Learning for Medical Imaging Using Radiomics Features

Association of partial T2-FLAIR mismatch sign and isocitrate dehydrogenase mutation in WHO grade 4 gliomas: results from the ReSPOND consortium

The 2021 World Health Organization classification of gliomas: an imaging approach

Contact Info

Product

Resources

About