Validation of an Artificial Intelligence–Based Prediction Model Using 5 External PET/CT Datasets of Diffuse Large B-Cell Lymphoma

The aim of this study was to validate a previously developed deep learning model in 5 independent clinical trials. The predictive performance of this model was compared with the international prognostic index (IPI) and 2 models incorporating radiomic PET/CT features (clinical PET and PET models). <b>Methods:</b> In total, 1,132 diffuse large B-cell lymphoma patients were included: 296 for training and 836 for external validation. The primary outcome was 2-y time to progression. The deep learning model was trained on maximum-intensity projections from PET/CT scans. The clinical PET model included metabolic tumor volume, maximum distance from the bulkiest lesion to another lesion, SUV<sub>peak</sub>, age, and performance status. The PET model included metabolic tumor volume, maximum distance from the bulkiest lesion to another lesion, and SUV<sub>peak</sub> Model performance was assessed using the area under the curve (AUC) and Kaplan-Meier curves. <b>Results:</b> The IPI yielded an AUC of 0.60 on all external data. The deep learning model yielded a significantly higher AUC of 0.66 (<i>P</i> < 0.01). For each individual clinical trial, the model was consistently better than IPI. Radiomic model AUCs remained higher for all clinical trials. The deep learning and clinical PET models showed equivalent performance (AUC, 0.69; <i>P</i> > 0.05). The PET model yielded the highest AUC of all models (AUC, 0.71; <i>P</i> < 0.05). <b>Conclusion:</b> The deep learning model predicted outcome in all trials with a higher performance than IPI and better survival curve separation. This model can predict treatment outcome in diffuse large B-cell lymphoma without tumor delineation but at the cost of a lower prognostic performance than with radiomics.

Authors

Institutions

Publication Details

Journal
Journal of Nuclear Medicine
Published
2024-10-03
DOI
https://doi.org/10.2967/jnumed.124.268191
Citations
12
Primary Topic
Medical Imaging Techniques and Applications
Type
article
Field-Weighted Citation Impact
4.85

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Validation of an Artificial Intelligence–Based Prediction Model Using 5 External PET/CT Datasets of Diffuse Large B-Cell Lymphoma

Christine Hanoun, Sándor Czibor, Martijn W. Heymans, Martine E.D. Chamuleau et al.
12 citations
Journal of Nuclear Medicine
Medical Imaging Techniques and Applications
4.85
article

Validation of an Artificial Intelligence–Based Prediction Model Using 5 External PET/CT Datasets of Diffuse Large B-Cell Lymphoma

Christine Hanoun, Sándor Czibor, Martijn W. Heymans, Martine E.D. Chamuleau, Tamás Györke, Luca Ceriani, Lars Kurch, Ronald Boellaard, Jakoba J. Eertink, Sally F. Barrington, N. George Mikhaeel, Josée M. Zijlstra, Pieternella J. Lugtenburg, Sandeep S.V. Golla, Gerben J.C. Zwezerijnen, Sanne E. Wiegers, Maria C. Ferrández, Andreas Hüttmann, Ulrich Dührsen, Emanuele Zucca
article en
12 citations

Abstract

The aim of this study was to validate a previously developed deep learning model in 5 independent clinical trials. The predictive performance of this model was compared with the international prognostic index (IPI) and 2 models incorporating radiomic PET/CT features (clinical PET and PET models). <b>Methods:</b> In total, 1,132 diffuse large B-cell lymphoma patients were included: 296 for training and 836 for external validation. The primary outcome was 2-y time to progression. The deep learning model was trained on maximum-intensity projections from PET/CT scans. The clinical PET model included metabolic tumor volume, maximum distance from the bulkiest lesion to another lesion, SUV<sub>peak</sub>, age, and performance status. The PET model included metabolic tumor volume, maximum distance from the bulkiest lesion to another lesion, and SUV<sub>peak</sub> Model performance was assessed using the area under the curve (AUC) and Kaplan-Meier curves. <b>Results:</b> The IPI yielded an AUC of 0.60 on all external data. The deep learning model yielded a significantly higher AUC of 0.66 (<i>P</i> < 0.01). For each individual clinical trial, the model was consistently better than IPI. Radiomic model AUCs remained higher for all clinical trials. The deep learning and clinical PET models showed equivalent performance (AUC, 0.69; <i>P</i> > 0.05). The PET model yielded the highest AUC of all models (AUC, 0.71; <i>P</i> < 0.05). <b>Conclusion:</b> The deep learning model predicted outcome in all trials with a higher performance than IPI and better survival curve separation. This model can predict treatment outcome in diffuse large B-cell lymphoma without tumor delineation but at the cost of a lower prognostic performance than with radiomics.

Journal of Nuclear MedicineVol. 65(11)
Semmelweis University (HU), King's College London (GB), Dutch Cancer Society (NL), Breast Cancer Now (GB), London Cancer (GB), Institute of Oncology Research (CH), Amsterdam Neuroscience (NL), Kings Health Partners (GB), King's College School (GB), University Hospital Leipzig (DE), Swiss Group For Clinical Cancer Research (CH), Erasmus MC Cancer Institute (NL), Amsterdam University Medical Centers (NL), Cancer Center Amsterdam (NL), Università della Svizzera italiana (CH), University of Duisburg-Essen (DE), Vrije Universiteit Amsterdam (NL), Leipzig University (DE)
Amgen, Pfizer, AstraZeneca, Gilead Sciences, Celgene
Openalex Percentile: Top 4%
Medical Imaging Techniques and Applications
4.85
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.