TY - JOUR
T1 - An artificial intelligence method using FDG PET to predict treatment outcome in diffuse large B cell lymphoma patients
AU - Ferrández, Maria C.
AU - Golla, Sandeep S.V.
AU - Eertink, Jakoba J.
AU - de Vries, Bart M.
AU - Lugtenburg, Pieternella J.
AU - Wiegers, Sanne E.
AU - Zwezerijnen, Gerben J.C.
AU - Pieplenbosch, Simone
AU - Kurch, Lars
AU - Hüttmann, Andreas
AU - Hanoun, Christine
AU - Dührsen, Ulrich
AU - de Vet, Henrica C.W.
AU - Hoekstra, Otto S.
AU - Burggraaff, Coreline N.
AU - Bes, Annelies
AU - Heymans, Martijn W.
AU - Jauw, Yvonne W.S.
AU - Chamuleau, Martine E.D.
AU - Barrington, Sally F.
AU - Mikhaeel, George
AU - Zucca, Emanuele
AU - Ceriani, Luca
AU - Carr, Robert
AU - Györke, Tamás
AU - Czibor, Sándor
AU - Fanti, Stefano
AU - Kostakoglu, Lale
AU - Lee, Sze Ting
AU - Zijlstra, Josée M.
AU - Boellaard, Ronald
AU - PETRA
A2 - Hutchings, Martin
A2 - Loft, Annika
N1 - Publisher Copyright:
© 2023, Springer Nature Limited.
PY - 2023/12
Y1 - 2023/12
N2 - Convolutional neural networks (CNNs) may improve response prediction in diffuse large B-cell lymphoma (DLBCL). The aim of this study was to investigate the feasibility of a CNN using maximum intensity projection (MIP) images from 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) baseline scans to predict the probability of time-to-progression (TTP) within 2 years and compare it with the International Prognostic Index (IPI), i.e. a clinically used score. 296 DLBCL 18F-FDG PET/CT baseline scans collected from a prospective clinical trial (HOVON-84) were analysed. Cross-validation was performed using coronal and sagittal MIPs. An external dataset (340 DLBCL patients) was used to validate the model. Association between the probabilities, metabolic tumour volume and Dmaxbulk was assessed. Probabilities for PET scans with synthetically removed tumors were also assessed. The CNN provided a 2-year TTP prediction with an area under the curve (AUC) of 0.74, outperforming the IPI-based model (AUC = 0.68). Furthermore, high probabilities (> 0.6) of the original MIPs were considerably decreased after removing the tumours (< 0.4, generally). These findings suggest that MIP-based CNNs are able to predict treatment outcome in DLBCL.
AB - Convolutional neural networks (CNNs) may improve response prediction in diffuse large B-cell lymphoma (DLBCL). The aim of this study was to investigate the feasibility of a CNN using maximum intensity projection (MIP) images from 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) baseline scans to predict the probability of time-to-progression (TTP) within 2 years and compare it with the International Prognostic Index (IPI), i.e. a clinically used score. 296 DLBCL 18F-FDG PET/CT baseline scans collected from a prospective clinical trial (HOVON-84) were analysed. Cross-validation was performed using coronal and sagittal MIPs. An external dataset (340 DLBCL patients) was used to validate the model. Association between the probabilities, metabolic tumour volume and Dmaxbulk was assessed. Probabilities for PET scans with synthetically removed tumors were also assessed. The CNN provided a 2-year TTP prediction with an area under the curve (AUC) of 0.74, outperforming the IPI-based model (AUC = 0.68). Furthermore, high probabilities (> 0.6) of the original MIPs were considerably decreased after removing the tumours (< 0.4, generally). These findings suggest that MIP-based CNNs are able to predict treatment outcome in DLBCL.
UR - http://www.scopus.com/inward/record.url?scp=85167764089&partnerID=8YFLogxK
U2 - 10.1038/s41598-023-40218-1
DO - 10.1038/s41598-023-40218-1
M3 - Journal article
C2 - 37573446
AN - SCOPUS:85167764089
SN - 2045-2322
VL - 13
JO - Scientific Reports
JF - Scientific Reports
IS - 1
M1 - 13111
ER -