What does this research mean for the field?

A radiomics-based machine learning model using T2-FLAIR digital subtraction images outperforms conventional neuroradiologist visual assessment in differentiating IDH-mutant astrocytomas from other non-enhancing low-grade gliomas. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This study aims to improve the differentiation of IDH-mutant astrocytomas from other low-grade gliomas using a radiomics-based machine learning model.

May 17, 2026NeuroradiologyOpen Access

T2-FLAIR digital subtraction radiomics versus neuroradiologist visual assessment for differentiating IDH-mutant astrocytomas from other non-enhancing low-grade gliomas: An externally validated machine learning study

View Full Paper

Ask AI

Bookmark

Authors

EDEmin DemirelSağlık Bilimleri Üniversitesi ODOkan DılekAdana Numune Eğitim ve Araştırma Hastanesi

Discussion

Member takes

Implication

Randomized trial compares machine learning diagnostic performance with visual assessment in gliomas, suggesting improvements in differentiation accuracy.

Key Points

This study aims to improve the differentiation of IDH-mutant astrocytomas from other low-grade gliomas using a radiomics-based machine learning model.
Included 193 patients with non-enhancing low-grade gliomas from two independent cohorts.
Developed T2-FLAIR digital subtraction images and selected ten radiomics features using machine learning techniques.
Compared diagnostic performance with neuroradiologist visual assessments using DeLong test.
Radiomics model achieved an AUC of 0.879 with sensitivity of 71.2% and specificity of 85.4% in the training cohort.
Outperformed neuroradiologist assessment with training AUC of 0.768 (p = 0.003) and external test AUC of 0.741 (p = 0.038).
Inter-reader agreement for conventional assessment was excellent (intraclass correlation coefficient = 0.89).

Cite This Study

Demirel et al. (2026) studied this question.

synapsesocial.com/papers/6a095c147880e6d24efe2160 https://doi.org/10.1007/s00234-026-04025-5

View Full Paper

Ask AI

Bookmark

Implication

Randomized trial compares machine learning diagnostic performance with visual assessment in gliomas, suggesting improvements in differentiation accuracy.

T2-FLAIR digital subtraction radiomics versus neuroradiologist visual assessment for differentiating IDH-mutant astrocytomas from other non-enhancing low-grade gliomas: An externally validated machine learning study

Discussion

Member takes

Implication

Key Points

Cite This Study

T2-FLAIR digital subtraction radiomics versus neuroradiologist visual assessment for differentiating IDH-mutant astrocytomas from other non-enhancing low-grade gliomas: An externally validated machine learning study

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider