Dataset Information

An artificial intelligence method using FDG PET to predict treatment outcome in diffuse large B cell lymphoma patients.

ABSTRACT: 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 Dmax_bulk 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.

SUBMITTER: Ferrandez MC

PROVIDER: S-EPMC10423266 | biostudies-literature | 2023 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

An artificial intelligence method using FDG PET to predict treatment outcome in diffuse large B cell lymphoma patients.

Ferrández Maria C MC Golla Sandeep S V SSV Eertink Jakoba J JJ de Vries Bart M BM Lugtenburg Pieternella J PJ Wiegers Sanne E SE Zwezerijnen Gerben J C GJC Pieplenbosch Simone S Kurch Lars L Hüttmann Andreas A Hanoun Christine C Dührsen Ulrich U de Vet Henrica C W HCW Zijlstra Josée M JM Boellaard Ronald R

Scientific reports 20230812 1

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 <sup>18</sup>F-fluorodeoxyglucose (<sup>18</sup>F-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. 29 ...[more]

PMID: 37573446

Similar Datasets

Project description:We assessed the predictive value of new radiomic features characterizing lesion dissemination in baseline 18F-FDG PET and tested whether combining them with baseline metabolic tumor volume (MTV) could improve prediction of progression-free survival (PFS) and overall survival (OS) in diffuse large B-cell lymphoma (DLBCL) patients. Methods: From the LNH073B trial (NCT00498043), patients with advanced-stage DLCBL and 18F-FDG PET/CT images available for review were selected. MTV and several radiomic features, including the distance between the 2 lesions that were farthest apart (Dmaxpatient), were calculated. Receiver-operating-characteristic analysis was used to determine the optimal cutoff for quantitative variables, and Kaplan-Meier survival analyses were performed. Results: With a median age of 46 y, 95 patients were enrolled, half of them treated with R-CHOP biweekly (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) and the other half with R-ACVBP (rituximab, doxorubicin, cyclophosphamide, vindesine, bleomycin, and prednisone), with no significant impact on outcome. Median MTV and Dmaxpatient were 375 cm3 and 45 cm, respectively. The median follow-up was 44 mo. High MTV and Dmaxpatient were adverse factors for PFS (P = 0.027 and P = 0.0003, respectively) and for OS (P = 0.0007 and P = 0.0095, respectively). In multivariate analysis, only Dmaxpatient was significantly associated with PFS (P = 0.0014) whereas both factors remained significant for OS (P = 0.037 and P = 0.0029, respectively). Combining MTV (>384 cm3) and Dmaxpatient (>58 cm) yielded 3 risk groups for PFS (P = 0.0003) and OS (P = 0.0011): high with 2 adverse factors (4-y PFS and OS of 50% and 53%, respectively, n = 18), low with no adverse factor (94% and 97%, n = 36), and an intermediate category with 1 adverse factor (73% and 88%, n = 41). Conclusion: Combining MTV with a parameter reflecting the tumor burden dissemination further improves DLBCL patient risk stratification at staging.

Project description:BackgroundConvolutional neural networks (CNNs), applied to baseline [18F]-FDG PET/CT maximum intensity projections (MIPs), show potential for treatment outcome prediction in diffuse large B-cell lymphoma (DLBCL). The aim of this study is to investigate the robustness of CNN predictions to different image reconstruction protocols. Baseline [18F]FDG PET/CT scans were collected from 20 DLBCL patients. EARL1, EARL2 and high-resolution (HR) protocols were applied per scan, generating three images with different image qualities. Image-based transformation was applied by blurring EARL2 and HR images to generate EARL1 compliant images using a Gaussian filter of 5 and 7 mm, respectively. MIPs were generated for each of the reconstructions, before and after image transformation. An in-house developed CNN predicted the probability of tumor progression within 2 years for each MIP. The difference in probabilities per patient was then calculated between both EARL2 and HR with respect to EARL1 (delta probabilities or ΔP). We compared these to the probabilities obtained after aligning the data with ComBat using the difference in median and interquartile range (IQR).ResultsCNN probabilities were found to be sensitive to different reconstruction protocols (EARL2 ΔP: median = 0.09, interquartile range (IQR) = [0.06, 0.10] and HR ΔP: median = 0.1, IQR = [0.08, 0.16]). Moreover, higher resolution images (EARL2 and HR) led to higher probability values. After image-based and ComBat transformation, an improved agreement of CNN probabilities among reconstructions was found for all patients. This agreement was slightly better after image-based transformation (transformed EARL2 ΔP: median = 0.022, IQR = [0.01, 0.02] and transformed HR ΔP: median = 0.029, IQR = [0.01, 0.03]).ConclusionOur CNN-based outcome predictions are affected by the applied reconstruction protocols, yet in a predictable manner. Image-based harmonization is a suitable approach to harmonize CNN predictions across image reconstruction protocols.

Project description:PurposeThis study aimed to develop a novel analytic approach based on 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography/computed tomography ([18F]FDG PET/CT) radiomic signature (RS) and International Prognostic Index (IPI) to predict the progression-free survival (PFS) and overall survival (OS) of patients with diffuse large B-cell lymphoma (DLBCL).MethodsWe retrospectively enrolled 152 DLBCL patients and divided them into a training cohort (n = 100) and a validation cohort (n = 52). A total of 1245 radiomic features were extracted from the total metabolic tumor volume (TMTV) and the metabolic bulk volume (MBV) of pre-treatment PET/CT images. The least absolute shrinkage and selection operator (LASSO) algorithm was applied to develop the RS. Cox regression analysis was used to construct hybrid nomograms based on different RS and clinical variables. The performances of hybrid nomograms were evaluated using the time-dependent receiver operator characteristic (ROC) curve and the Hosmer-Lemeshow test. The clinical utilities of prediction nomograms were determined via decision curve analysis. The predictive efficiency of different RS, clinical variables, and hybrid nomograms was compared.ResultsThe RS and IPI were identified as independent predictors of PFS and OS, and were selected to construct hybrid nomograms. Both TMTV- and MBV-based hybrid nomograms had significantly higher values of area under the curve (AUC) than IPI in training and validation cohorts (all P < 0.05), while no significant difference was found between TMTV- and MBV-based hybrid nomograms (P > 0.05). The Hosmer-Lemeshow test showed that both TMTV- and MBV-based hybrid nomograms calibrated well in the training and validation cohorts (all P > 0.05). Decision curve analysis indicated that hybrid nomograms had higher net benefits than IPI.ConclusionThe hybrid nomograms combining RS with IPI could significantly improve survival prediction in DLBCL. Radiomic analysis on MBV may serve as a potential approach for prognosis assessment in DLBCL.Trial registrationNCT04317313. Registered March 16, 2020. Public site: https://clinicaltrials.gov/ct2/show/NCT04317313.

Project description:Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous and prevalent subtype of aggressive non-Hodgkin lymphoma that poses diagnostic and prognostic challenges, particularly in predicting drug responsiveness. In this study, we used digital pathology and deep learning to predict responses to immunochemotherapy in patients with DLBCL. We retrospectively collected 251 slide images from 216 DLBCL patients treated with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP), with their immunochemotherapy response labels. The digital pathology images were processed using contrastive learning for feature extraction. A multi-modal prediction model was developed by integrating clinical data and pathology image features. Knowledge distillation was employed to mitigate overfitting on gigapixel histopathology images to create a model that predicts responses based solely on pathology images. Based on the importance derived from the attention mechanism of the model, we extracted histological features that were considered key textures associated with drug responsiveness. The multi-modal prediction model achieved an impressive area under the ROC curve of 0.856, demonstrating significant associations with clinical variables such as Ann Arbor stage, International Prognostic Index, and bulky disease. Survival analyses indicated their effectiveness in predicting relapse-free survival. External validation using TCGA datasets supported the model's ability to predict survival differences. Additionally, pathology-based predictions show promise as independent prognostic indicators. Histopathological analysis identified centroblastic and immunoblastic features to be associated with treatment response, aligning with previous morphological classifications and highlighting the objectivity and reproducibility of artificial intelligence-based diagnosis. This study introduces a novel approach that combines digital pathology and clinical data to predict the response to immunochemotherapy in patients with DLBCL. This model shows great promise as a diagnostic and prognostic tool for clinical management of DLBCL. Further research and genomic data integration hold the potential to enhance its impact on clinical practice, ultimately improving patient outcomes.

Dataset Information

An artificial intelligence method using FDG PET to predict treatment outcome in diffuse large B cell lymphoma patients.

Publications

An artificial intelligence method using FDG PET to predict treatment outcome in diffuse large B cell lymphoma patients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets