論文種別	原著(症例報告除く)
言語種別	英語
査読の有無	その他(不明)
表題	Comparing artificial intelligence and physician performance in predicting IDH mutation status in glioma.
掲載誌名	正式名：NPJ digital medicine 略　 称：NPJ Digit Med ISSNコード：23986352/23986352
掲載区分	国外
巻・号・頁	pp.Online ahead of print
著者・共著者	Satoshi Takahashi, Masamichi Takahashi, Manabu Kinoshita, Mototaka Miyake, Risa Kawaguchi, Naoki Shinojima, Akitake Mukasa, Kuniaki Saito, Motoo Nagane, Ryohei Otani, Fumi Higuchi, Shota Tanaka, Nobuhiro Hata, Kaoru Tamura, Kensuke Tateishi, Ryo Nishikawa, Hideyuki Arita, Masahiro Nonaka, Takehiro Uda, Junya Fukai, Yoshiko Okita, Naohiro Tsuyuguchi, Yonehiro Kanemura, Fumiyasu Tsushima, Shingo Kakeda, Toshiaki Akashi, Toshiaki Taoka, Yoshiyuki Watanabe, Kei Yamada, Toshinori Hirai, Minako Azuma, Takashi Yoshiura, Jun Sese, Koichi Ichimura, Yoshitaka Narita, Ryuji Hamamoto
発行年月	2026/05
概要	Predicting isocitrate dehydrogenase (IDH) mutations in gliomas using magnetic resonance imaging (MRI) is clinically important for treatment planning. This study compared two artificial intelligence (AI) models, GliomaDepth-IDH (ResNet34-based) and GliomaVista-IDH (Vision Transformer-based), with 18 physicians (eight neuroradiologists, five neurosurgeons, and five neurosurgery residents) in predicting IDH mutation status. On the Brain Tumor Segmentation Challenge dataset, the GliomaVista-IDH AI model achieved an area under the curve (AUC) value of 0.97, significantly outperforming all physician groups. However, external validation on a Japanese cohort revealed performance degradation: GliomaDepth-IDH declined to an AUC of 0.75 and GliomaVista-IDH to 0.82, with GliomaVista-IDH showing significant calibration issues (Brier score = 0.32). High-performing physicians achieved comparable results (AUC = 0.88) with superior calibration (Brier score = 0.19). Inter-rater reliability analysis revealed substantial variability across physician groups. These findings suggest that AI models can assist many physicians, while experienced practitioners remain competitive with better-calibrated predictions in challenging domains.
DOI	10.1038/s41746-026-02695-2
PMID	42086700