Deep learning-based metastasis detection in patients with lung cancer to enhance reproducibility and reduce workload in brain metastasis screening with MRI: a multi-center study

• Published in: Cancer imaging Vol. 24; no. 1; pp. 32 - 10
• Main Authors: Park, Yae Won, Park, Ji Eun, Ahn, Sung Soo, Han, Kyunghwa, Kim, NakYoung, Oh, Joo Young, Lee, Da Hyun, Won, So Yeon, Shin, Ilah, Kim, Ho Sung, Lee, Seung-Koo
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 01.03.2024; BioMed Central; BMC
• Subjects: Artificial intelligence; Blood; Brain; Brain cancer; Brain metastases; Brain Neoplasms - diagnostic imaging; Brain Neoplasms - secondary; Brain tumors; Cancer patients; Cancer therapies; Clinical medicine; Confidence intervals; Deep Learning; Early Detection of Cancer; Humans; Lung cancer; Lung Neoplasms - diagnostic imaging; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Medical diagnosis; Medical imaging; Metastasis; Neural networks; Patients; Radiation therapy; Readers; Reading; Reproducibility; Reproducibility of Results; Retrospective Studies; Software; Tumors; Workflow; Workload; Workloads; Deep learning; Brain metastases; Magnetic resonance imaging; Brain tumors
• ISSN: 1470-7330; 1740-5025; 1470-7330
• DOI: 10.1186/s40644-024-00669-9

To assess whether a deep learning-based system (DLS) with black-blood imaging for brain metastasis (BM) improves the diagnostic workflow in a multi-center setting. In this retrospective study, a DLS was developed in 101 patients and validated on 264 consecutive patients (with lung cancer) having newly developed BM from two tertiary university hospitals, which performed black-blood imaging between January 2020 and April 2021. Four neuroradiologists independently evaluated BM either with segmented masks and BM counts provided (with DLS) or not provided (without DLS) on a clinical trial imaging management system (CTIMS). To assess reading reproducibility, BM count agreement between the readers and the reference standard were calculated using limits of agreement (LoA). Readers' workload was assessed with reading time, which was automatically measured on CTIMS, and were compared between with and without DLS using linear mixed models considering the imaging center. In the validation cohort, the detection sensitivity and positive predictive value of the DLS were 90.2% (95% confidence interval [CI]: 88.1-92.2) and 88.2% (95% CI: 85.7-90.4), respectively. The difference between the readers and the reference counts was larger without DLS (LoA: -0.281, 95% CI: -2.888, 2.325) than with DLS (LoA: -0.163, 95% CI: -2.692, 2.367). The reading time was reduced from mean 66.9 s (interquartile range: 43.2-90.6) to 57.3 s (interquartile range: 33.6-81.0) (P <.001) in the with DLS group, regardless of the imaging center. Deep learning-based BM detection and counting with black-blood imaging improved reproducibility and reduced reading time, on multi-center validation.