Details

Autor(en) / Beteiligte

• Shuaib, Haris
• Barker, Gareth J
• Sasieni, Peter
• De Vita, Enrico
• Chelliah, Alysha
• Andrei, Roman
• Ashkan, Keyoumars
• Beaumont, Erica
• Brazil, Lucy
• Rowland-Hill, Chris
• Lau, Yue Hui
• Luis, Aysha
• Powell, James
• Swampillai, Angela
• Tenant, Sean
• Thust, Stefanie C
• Wastling, Stephen
• Young, Tom
• Booth, Thomas C

Titel

Overcoming challenges of translating deep-learning models for glioblastoma: the ZGBM consortium

Ist Teil von

• British journal of radiology, 2023-01, Vol.96 (1141), p.20220206-20220206

Ort / Verlag

England

Erscheinungsjahr

2023

Quelle

MEDLINE

Beschreibungen/Notizen

• To report imaging protocol and scheduling variance in routine care of glioblastoma patients in order to demonstrate challenges of integrating deep-learning models in glioblastoma care pathways. Additionally, to understand the most common imaging studies and image contrasts to inform the development of potentially robust deep-learning models. MR imaging data were analysed from a random sample of five patients from the prospective cohort across five participating sites of the ZGBM consortium. Reported clinical and treatment data alongside DICOM header information were analysed to understand treatment pathway imaging schedules. All sites perform all structural imaging at every stage in the pathway except for the presurgical study, where in some sites only contrast-enhanced -weighted imaging is performed. Diffusion MRI is the most common non-structural imaging type, performed at every site. The imaging protocol and scheduling varies across the UK, making it challenging to develop machine-learning models that could perform robustly at other centres. Structural imaging is performed most consistently across all centres. Successful translation of deep-learning models will likely be based on structural post-treatment imaging unless there is significant effort made to standardise non-structural or peri-operative imaging protocols and schedules.