An artificial intelligence method using 18F-FDG PET maximum intensity projections to predict 2-year time-to-progression in diffuse large B-cell lymphoma patients

doi:10.21203/rs.3.rs-2761494/v1

Download PDF

Article

An artificial intelligence method using ¹⁸F-FDG PET maximum intensity projections to predict 2-year time-to-progression in diffuse large B-cell lymphoma patients

https://doi.org/10.21203/rs.3.rs-2761494/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 11 Aug, 2023

Read the published version in Scientific Reports →

You are reading this latest preprint version

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 (MTV) 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 area under the curve (AUC) of 0.74, outperforming the IPI-based model (AUC = 0.68). A moderate association of CNN probabilities with MTV (r = 0.57) and Dmax_bulk (r = 0.52) was observed in the external dataset. 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.

Health sciences/Oncology/Cancer/Cancer imaging

Health sciences/Diseases/Cancer/Haematological cancer/Lymphoma/Non hodgkin lymphoma/B cell lymphoma

Diffuse large B-cell lymphoma (DLBCL) is an aggressive lymphoid malignancy which originates in the B lymphocytes and accounts for 30% of the total annual diagnoses of Non-Hodgkin lymphoma in western countries [1]. A widely used first-line therapy in DLBCL combines rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP). The number of R-CHOP cycles and/or initial usage of more intense chemotherapy regimens initially depends on the primary disease stage and the International Prognostic Index (IPI), which defines a patient's risk profile [2]. IPI score includes age, World Health Organization performance status, Ann Arbor stage, serum lactate dehydrogenase level, and number of extranodal sites of disease. ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography (PET) - computed tomography (CT) imaging allows highly accurate visualization of DLBCL tumours, which is therefore the essential modality for appropriate staging. Moreover, ¹⁸F-FDG PET is frequently used as an early outcome prediction marker, since complete metabolic response early throughout (i.e. interim) therapy allows de-escalation of treatment cycles [3]. This interim-PET adaptive treatment approach is increasingly integrated into national recommendations on DLBCL. Despite better identification of low-risk patients at baseline and early during treatment, overall, one-third of DLBCL patients do not respond to first-line treatment or relapse [1]. Therefore, early identification of high-risk patients is important as patients might benefit from a more tailored treatment strategy.

Quantitative parameters extracted from ¹⁸F-FDG PET/CT scans provide insight into the tumour characteristics. From these parameters, metabolic tumour volume (MTV) has been repeatedly reported as a promising prognostic factor in DLBCL [4–6]. The inclusion of dissemination features like the maximal distance between the largest lesion and any other lesion (Dmax_bulk), in combination with MTV, has further improved risk stratification of patients [5]. To obtain these parameters, tumour segmentation requires user interaction for each ¹⁸F-FDG PET/CT scan which can be time consuming and depends on the observers interpretation. The implementation of artificial intelligence (AI) and convolutional neural networks (CNNs) might be able to reduce and/or replace these tasks. CNNs can extract high-level features from multi-dimensional input data (i.e. images). In oncology, CNNs are already being investigated to automate different medical image classification tasks: diagnostics [7], tumour delineation and segmentation [8–10], extraction of PET features surrogates [11] and disease progression and/or treatment outcome prediction [12, 13]. Two of the main drawbacks of CNNs are the computational expense they entail and the high complexity of its extracted features, especially when it comes to 3D image analysis such as with PET/CT scans. Alternatively, maximum intensity projections (MIPs) of ¹⁸F-FDG PET/CT scans can be used as 2D inputs for the CNN, decreasing data dimension and complexity and, thereby, decreasing the computational load and cost [14, 15].

The aim of this study was to investigate the feasibility of a CNN for the prediction of 2-year time-to-progression (TTP) in DLBCL patients using MIP images derived from ¹⁸F-FDG PET/CT baseline scans. The models outcome is a binary prediction given by the probability of TTP longer than 2 years P(TTP0) or TTP shorter than 2 years P(TTP1), where TTP1 indicates an increased risk of tumour progression for the patient. TTP0 may indicate absence of tumour progression or absence of recurrence.

Datasets

In this study we used two different datasets of baseline DLBCL ¹⁸F-FDG PET/CT scans: the HOVON-84 dataset [16] was used to train the CNN model whereas the PETAL dataset [17] was used as an external validation of the models performance. All patients from both datasets provided written consent for the use of their data. After correction for IPI, there were no significant differences in survival between the PETAL and HOVON-84 study [18]. Both studies were approved by institutional review boards and all included patients provided informed consent. The use of all data within the PETRA imaging database has been approved by the institutional review board of the VU University Medical Center (JR /20140414).

HOVON-84. Three hundred seventy-three DLBCL patients who underwent baseline ¹⁸F-FDG PET/CT from the HOVON-84 trial (EudraCT, 2006–005,174 − 42) were included in this study. The main inclusion/exclusion criteria from this trial can be found elsewhere [16]. From these, 317 diagnosed DLBCL patients were included in this analysis. Missing essential DICOM (Digital Imaging and Communication in Medicine) information was the main reason for exclusion. Other reasons included: quality control (QC) outside of range, incomplete whole-body scans, no FDG-avid lesions or plasma glucose out of range. Furthermore, 7 patients were lost to follow-up within 2 years and 14 other patients died within 2 years of unrelated reasons. This led to a total of 296 DLBCL patients included in the study. Of which, 244 were classified as TTP0 and 52 as TTP1.

PETAL. The external validation was performed using diagnosed DLBCL patients from the PETAL trial (EudraCT 2006-001641-33) who underwent baseline ¹⁸F-FDG PET/CT. The eligibility for the PETAL trial is described elsewhere [17]. Initially, the trial consisted of 1098 PETAL patients. Reasons to exclude patients were as follows: diagnosis other than DLBCL, incomplete scans or with artefacts, missing DICOM information, QC or glucose out of range or no FDG-avid lesions. This led to a total of 395 DLBCL patients with associated ¹⁸F-FDG PET/CT baselines scans available for this study. Moreover, 12 underwent a different treatment to R-CHOP, 24 patients were lost for follow-up within 2 years, and 19 died without progression. This led to a total of 340 patients. From these, 279 were classified as TTP0 and 61 as TTP1.

Quality Control of scans. The participating sites provided the scans in DICOM format. The scans were subsequently anonymised. For QC we used criteria described by EANM guidelines: mean standardized uptake value (SUVmean) of the liver should be between 1.3 and 3.0 and the plasma glucose level lower than 11 mmol/L [3]. The QC criteria are described in detail elsewhere [4].

Maximum Intensity Projections

The ACCURATE tool was used to obtain the so-called lesion masks which are the images that contain only the lymphoma tumour(s) segmentation [19]. The segmentation of the tumours was performed using a SUV threshold of 4.0 [20]. Any physiological uptake adjacent to tumours was manually deleted. The conversion to MIP images was performed using an in-house developed CNN preprocessing tool which generates coronal and sagittal MIPs with size 275 x 200 x 1 and a pixel size of 4x4mm. MIPs were generated for (binary) lesion masks, lesion MIPs (MIPs containing only tumours) and for the complete PET scans. Examples of these coronal and sagittal MIPs can be found in Fig. 1.The MIPs were normalized by a fixed maximum intensity value (SUV = 40). The values above this maximum were truncated to avoid normalization to be driven by the SUV value of high uptake organs such as the bladder.

Data sampling scheme

Since the training dataset classes were highly unbalanced (244 TTP0 vs 52 TTP1), we applied a data sampling scheme were the TTP0s were divided into 5 equally stratified data subsets: 4 subsets of 49 patients (subsets A-D) and 1 subset of 48 patients (subset E). Additionally, 3 randomly selected TTP0 patients belonging to different subsets, were added to each of these subsets in order to achieve a total of 52 TTP0s per subset (subset E with 51 TTP0s), which matched the total number of TTP1s. Eventually, each subset contained a total of 104 patients with a prevalence of 50% for each class (subset E with 103 patients). The details of this scheme can be found in Fig. 2. Each subset (A to E) was trained using a 5-fold cross validation (for each cross validation run, the data was split into two sets: training set (80%) and internal validation set (20%)).

Convolutional Neural Network

The CNN consists of two branches, one receives the coronal MIP as input and the other one receives the sagittal MIP as input, which are merged as a last step to yield the final prediction. Coronal and sagittal MIPs are analysed independently but in parallel by an identical multi-layer architecture. The CNN design consists of 4 convolution layers, each one of these are followed by a max pooling layer. In a CNN, the convolution layer uses different filters over the image to extract low level features (e.g. edges, gradients) in earlier layers and high level features in deeper layers. In our CNN, the feature maps are doubled at each convolution layer, starting at 16 in the first layer and going up to 128 in the last layer, and their dimensions continuously decrease by (3,3). In each convolution layer the rectified linear unit (ReLU) activation function is applied. After each convolution layer, a dropout of 0.35 is applied, indicating that 35% of the network nodes and connections are randomly dropped from the CNN in order to prevent overfitting. Right after the dropout, a MaxPooling layer was implemented. The MaxPooling layer acts as the dimensionality reduction layer. There are 3 Maxpooling layers in our CNN, each of these with feature map sizes of (3,3), (3,3) and (2,2). After the last convolution and SpatialDropout layer, the CNN is connected to a GlobalAveragePooling2D (GAP2D) layer also known as ‘flattening’ layer. This layer ‘flattens’ the output from the convolution layers into a less complex shape (i.e. a 2D tensor). The coronal and sagittal outputs are then concatenated at the final dense layer or fully connected layer (FCL) which generates an output for two different classes: TTP0 and TTP1. A softmax function is introduced to generate a probability for each of these classes, which is the final CNN output the probability of TTP longer than 2 years, P(TTP0), and the probability of TTP shorter than 2 years, P(TTP1) both add up to 1. The classifier was compiled using the Adam optimizer with a learning rate (LRt) of 0.00005 and a decay rate (DR) of 0.000001. The CNN structure is illustrated in Fig. 3.

In this study we trained the model following 3 different training schemes. These are illustrated in Supplemental Fig. 1: training based on 1) only-lesion MIPs (Lesion MIP CNN); 2) lesion masks and regular MIPs (MIP CNN); 3) lesion masks and MIPs after removal of the brain, brain removed MIPs (BR-MIP CNN). The network architecture is kept the same. This approach was followed in order to explore if the model could be trained to recognize pathological patterns.

The lesion MIP CNN uses only the lesion MIPs as input (Fig. 1). These MIPs contain only the information and intensity of the tumours. The lesions MIPs of both coronal and sagittal views were used to train and validate the model for 200 epochs.

The MIP CNN uses both lesion masks and MIPs for the training of the model. The lesion masks contain only information of the tumours but not the intensity values. The training of the MIP CNN consisted in two subsequent steps. Firstly, the lesions masks of both coronal and sagittal MIPs were used to train and validate the model for 200 epochs. In the second step, the pre-trained model on the lesion masks was re-trained and re-validated for another 300 epochs, this time using the regular coronal and sagittal MIP images instead. The same patients were used for training of the two steps.

The BR-MIP CNN follows the same training process as the MIP CNN but instead of the original MIPs, it uses MIPs without brain (Fig. 1). MIP brain removal was performed in order to provide greater consistency across the dataset since not all scans included the head. The process of removing the brains is described in detail in Supplemental Material 1 [21].

In the case of the MIP CNN and the BR-MIP CNN, the classifier required only the MIP images to make the predictions. The idea behind these two CNNs was to generate a classifier where the prediction is free of the observer-dependent tumour segmentation.

Plausibility of the CNN

To better understand the CNN predictions, we further investigated the output of the model by exploring the association between P(TTP1) and two PET extracted features: MTV and Dmax_bulk since both have shown potential as prognostic markers in DLBCL. The process to extract PET features has been explained in previous studies [4]. Moreover, we synthetically removed the tumours from the MIP images to simulate tumour-free data and evaluated the CNN predictions on this data. The tumours were masked using the lesion masks generated using the in-house built CNN preprocessing tool. The voxel values corresponding to the tumours were replaced by an average of the voxel intensities excluding the background. This process is shown in Supplemental Fig. 2.

To facilitate representation, the TTP1 probabilities obtained through the CNN were calibrated by performing a logistic regression fit with the probabilities as input and the TTP0/TTP1 labels as outcome [22]. The obtained regression fit coefficients were then applied to the CNN TTP1 probabilities generated after removing the tumours in order to accordingly calibrate the tumour-free MIPs CNN TTP1 probabilities.

Statistical Analysis

The receiver operating characteristic (ROC) and the area under the curve (AUC) were used to evaluate the CNN performance. During training, the fold with the highest cross validated (CV-)AUC across the 5 folds was preserved. See Supplemental Material 2 for more details [23]. The PETAL dataset was used to externally validate the CNN model. This performance evaluation process was performed for the lesion MIP CNN, MIP CNN and BR-MIP CNN. In addition, the AUCs were compared to the performance of an IPI-based prediction model. This IPI model defines patients with risk factor of 4 or higher as high-risk patients (i.e. TTP1) [4]. The association of the TTP1 probabilities and the PET-extracted features (MTV and Dmax_bulk) was assessed using Pearson’s correlation coefficient.

A summary of the characteristics of the datasets can be found in Supplemental Table 1.

Internal validation

The results for the 5-fold CV for the 5 data subsets (A to E) can be found in the Supplemental Tables 2–4. The average performance of the models (associated with each subset) can be found in Supplemental Fig. 3 (a). The model trained with subset C was best performing in all cases (Supplemental Fig. 4). The AUCs for the model trained on subset C for Lesion MIP CNN, MIP CNN and BR-MIP CNN are illustrated in Fig. 4 (a) and the CV-AUCs are given in Table 1. These values are comparable or better to the ones obtained from the IPI prediction model (AUC was reported as 0.68 for the HOVON-84 dataset) [4].

Table 1

Cross-validation (± SD) of AUC, sensitivity, and specificity for training and internal validation (HOVON-84 dataset) for the model associated with subset C.
CNN	CV-AUC (SD)		Sensitivity (SD)		Specificity (SD)
CNN	Training	Validation	Training	Validation	Training	Validation
Lesion MIP	0.81 (0.02)	0.75 (0.07)	0.69 (0.06)	0.63 (0.19)	0.76 (0.02)	0.73 (0.08)
MIP	0.79 (0.03)	0.70 (0.06)	0.76 (0.04)	0.75 (0.08)	0.64 (0.06)	0.55 (0.10)
BR-MIP	0.77 (0.09)	0.72 (0.11)	0.73 (0.17)	0.71 (0.14)	0.69 (0.11)	0.63 (0.19)
Abbreviations: AUC: Area Under the Curve, SD: Standard Deviation

External validation

The PETAL dataset was used to externally validate the performance of the CNNs. The average performance of the models (associated with each subset) can be found in Supplemental Fig. 3 (b). The model trained with subset C was again the best performing trained model (Supplemental Fig. 5). The ROC plot with the corresponding AUC values for each CNN are shown in Fig. 4 (b). The BR-MIP CNN outperformed the Lesion MIP CNN and the MIP CNN (Table 2). Furthermore, it outperformed the IPI model with an AUC of 0.67 (sensitivity = 0.57, specificity = 0.68). We provided some examples of the BR-MIP CNN predictions in Supplemental Fig. 6.

Table 2

AUC, sensitivity and specificity for external validation data (PETAL dataset) for the model associated with subset C.
CNN	AUC	Sensitivity	Specificity
Lesion MIP	0.72	0.59	0.8
MIP	0.71	0.62	0.72
BR-MIP	0.74	0.54	0.81
Abbreviations: AUC: Area Under the Curve

Plausibility of the CNN

The BR-MIP CNN (trained using subset C) was used to further investigate the feasibility of the model for TTP prediction. A moderate association was found for MTV and Dmax_bulk with P(TTP1) as illustrated in Fig. 5 (a-d). In all scenarios, higher P(TTP1) seemed to be related to higher MTV and Dmax_bulk values. These features have been previously reported as promising prognostic factors in DLBCL [4–6].

After generating new probabilities for the tumour-free MIPs in the PETAL dataset. We found that these were generally lower when compared to the initial probabilities obtained from the images with tumours. This is the case, specially, for probabilities over 0.6, which, after tumour removal, were decreased to values below 0.4. Some examples are given in Supplemental Fig. 7 for different patients with decreased probabilities after tumour removal. The histogram of P(TTP1)s is shown in Fig. 6 for both datasets: original MIPs (with tumours) and tumour-free MIPs.

In this study, we investigated the feasibility of a CNN model for the prediction of progression after 2 years in DLBCL patients using ¹⁸F-FDG PET/CT MIP images. Our model was internally (HOVON-84 dataset) and externally validated (PETAL dataset) to assess the generalizability of the model in a different dataset [18]. Proper external validation is one of the main limitations seen across AI studies [24, 25]. Poor or insufficient validation causes misleading results and limits translation into the clinical setting [26]. In our external validation, the CNN showed an improved predictive performance compared to the IPI model (AUC of 0.74 vs 0.67). Features extracted from PET scans are currently being investigated to predict outcome of DLBCL patients with promising results. Eertink et. al. [4] developed a model that combined PET-extracted features (including MTV and Dmax_bulk) with clinical parameters, yielding a CV-AUC of 0.77 in an internal validation using scans from the HOVON-84 trial. Despite the potential of this model, it still requires time-consuming preprocessing steps and observer-dependent segmentation. Moreover, the calculations of PET extracted features are prone to variation depending on the reconstruction settings. Recent studies show how different reconstruction and segmentation methods yielded different MTVs and dissemination values extracted from the same lesion [27–29]. The implementation of automated tasks can potentially alleviate these effects. Several studies are exploring CNNs as segmentation tools and to calculate surrogates of these PET features. Blanc-Durand et. al. [9] trained a U-Net using 3D scans from both PET and CT for tumour segmentation, achieving a mean dice similarity score of 0.73 when compared to manually supervised segmentations. Girum et al. [11], instead, used MIP images for the same purpose. They supported their findings by proving that both surrogates for MTV and Dmax (largest distance between two lesions) were good prognosticators of progression free survival and overall survival. The use of MIPs has facilitated the implementation of deep learning for this purpose, as it allows to downsize the data from 3D to 2D which makes the process faster and less computationally expensive [15].

Previous studies [9, 11] (ref) show the potential of CNNs for improving the current state of treatment outcome prediction in DLBCL. However, these models are meant to address only one specific task (i.e. segmentation). Instead, our CNN is designed to replace every task involved in the process; end-to-end CNNs have a clear advantage over those which perform one specific task. There are few studies which have looked at the applications of end-to-end CNNs. Liu et al. [30] developed a multi task 3D U-Net for both tumour segmentation and prediction of progression-free survival (PFS) in DLBCL with outstanding results when compared to radiomic-based models and single task CNNs. Moreover, the use of MIPs for treatment outcome prediction in DLBCL is currently being investigated [15, 30, 31]. Rebaud et al. [31] trained a multi-task ranker neural network using coronal MIP images which performed as well as TMTV for DLBCL PFS prediction. However, these studies did not assess the generalizability of the model as indicated in the RELAINCE guidelines [26]. To our best knowledge, this is the first paper to investigate the feasibility of CNNs for ‘deep’ treatment outcome prediction in DLBCL patients using ¹⁸F-FDG PET/CT MIP images and to investigate its generalizability with external datasets.

We trained the CNN in 3 different ways: lesion MIP, MIP and BR-MIP CNN. In all cases, the CNN outperformed the IPI-based model. The BR-MIP CNN was kept for the CNN plausibility analysis as it predicted 2-year TTP with the highest AUC in the external dataset. Moreover, compared to the lesion MIP, BR-MIP CNN does not require prior tumour segmentation to make predictions and, compared to the MIP CNN, BR-MIP model uses MIP images without brain uptake. The main reason to remove the brain is that it brings consistency across the dataset as, in some PET studies, the head was not (fully) included during acquisition.

Regarding the limitations of this study, DLBCL patients can develop lesions near or within the brain which might complicate brain removal. Even after addressing this issue, there were around 1% of patients with truncated lesions which could not be solved. It is therefore important to consider clinician supervision in these cases.

In order to better understand the output of our model, we investigated whether there was any relation between the CNN outcome probabilities with MTV and Dmax_bulk, since both of them have prognostic capabilities for DLBCL. Our results suggests that the CNN is capturing information that might be related to tumour volume and dissemination but also that other image characteristics partly influence the CNN prediction. Deep learning methods tend to pay more attention to textural features [32]. In this context, conventional PET parameters, although easier to understand, might be missing relevant information for tumour progression. Another technique we used to examine the plausibility of the model is ‘ablating’ the tumours from the MIPs. The decrease in P(TTP1) values suggest that the CNN is indeed mainly using tumour information to make the predictions. Furthermore, when looking at a few examples of the CNN output (Supplemental Fig. 6) we found that patients with fewer tumours and lower dissemination are given lower probability values than patients with more tumours and higher dissemination. Even though further analysis is needed, these findings suggest that the model associates tumour dissemination with a higher risk of disease progression.

In this study we introduced a CNN model capable of predicting 2-year TTP in DLBCL patients using ¹⁸F-FDG PET/CT MIP images as input. The CNN model predicted 2-year TTP in DLBCL patients better than IPI scores. The CNN predictions are moderately associated with other PET derived predictive parameters (MTV and Dmax_bulk). Moreover, it was illustrated that the model prediction is affected by the presence or absence of lesions. Even though further investigations are necessary, our current findings suggest that CNNs using MIPs have potential as outcome prediction models.

DATA AVAILABILITY

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

ADDITIONAL INFORMATION

Competing Interests. This work was financially supported by the Hanarth Fonds Fund and the Dutch Cancer Society (#VU-2018-11648). M.C.F., S.S.V.G, J.J.E., B.M.d.V., S.E.W., G.J.C.Z., S.P., L.K., A.H., C.H., U.D., H.C.W.d.V. and R.B. declare no competing financial interests. P.J.L. received research funding from Takeda, Servier and Roche and received honoraria for advisory boards from Takeda, Servier, Genentech, Genmab, Celgene, Incyte and AbbVie. J.M.Z. received research funding from Roche and received honoraria for advisory boards from Takeda, Gilead, BMS and Roche. No other potential conflicts of interest relevant to this article exist.

Acknowledgements. This work was financially supported by the Hanarth Fonds Fund and the Dutch Cancer Society (#VU-2018-11648). The sponsor had no role in gathering, analyzing or interpreting the data. The authors thank all the patients who participated in the trials. Figures 2 and 3 were designed and drawn by M.C.F. The rest of the figures were generated with Python (version 3.9).

Ethics Declaration. All individual participants included in the study gave written informed consent to participate in the study. The HOVON-84 study was approved by the institutional review board of the Erasmus MC (2007–055) and was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. The PETAL study was approved by the Federal Institute for Drugs and Medical Devices and the ethics committees of all participating sites (University Hospital Essen and Deutsche Krebshilfe; ClinicalTrials.gov NCT00554164).

Consent to participate. Not applicable.

AUTHORS CONTRIBUTION

S.S.V.G. and R.B. contributed to the concept and design of the study. P.J.L. and U.D. were responsible for acquiring and collecting the data. C.H., S.P., J.J.E., S.E.W. and M.C.F. performed the data analysis. M.C.F. performed the training of the model, validation of the results and completed the first draft of the manuscript. P.J.L., U.D., L.K., A.H., C.H., S.P., S.E.W, B.M.d.V., J.J.E., S.S.V.G., R.B., G.J.C.Z. , H.C.W.V. and J.M.Z. reviewed and approved the manuscript.

Crump, M. et al. Outcomes in refractory diffuse large B-cell lymphoma: Results from the international scholar-1 study. Blood 130, 1800-1808 (2017).
Galaznik, A. et al. Predicting outcomes in patients with diffuse large B-cell lymphoma treated with standard of care. Cancer Inform 18, 1-16 (2019).
Boellaard, R. et al. FDG PET/CT: EANM procedure guidelines for tumour imaging: Version 2.0. Eur J Nucl Med Mol Imaging 42, 328-354 (2015).
Eertink, J. J. et al. (18)F-FDG PET baseline radiomics features improve the prediction of treatment outcome in diffuse large B-cell lymphoma. Eur J Nucl Med Mol Imaging 49, 932-942 (2022).
Cottereau, A. S. et al. (18)F-FDG PET dissemination features in diffuse large B-cell lymphoma are predictive of outcome. J Nucl Med 61, 40-45 (2020).
Schmitz, C. et al. Dynamic risk assessment based on positron emission tomography scanning in diffuse large B-cell lymphoma: Post-hoc analysis from the petal trial. Eur J Cancer 124, 25-36 (2020).
Bi, L. et al. Automatic detection and classification of regions of FDG uptake in whole-body PET/CT lymphoma studies. Comput Med Imaging Graph 60, 3-10 (2017).
Sibille, L. et al. (18) F-FDG PET/CT uptake classification in lymphoma and lung cancer by using deep convolutional neural networks. Radiology 294, 445-452 (2020).
Blanc-Durand, P. et al. Fully automatic segmentation of diffuse large B cell lymphoma lesions on 3D F-FDG PET/CT for total metabolic tumour volume prediction using a convolutional neural network. Eur J Nucl Med Mol Imaging 48, 1362-1370 (2021).
Chen, L. et al. Automatic PET cervical tumor segmentation by combining deep learning and anatomic prior. Phys Med Biol 64, 085019 (2019).
Girum, K. B. et al. (18)F-FDG PET maximum intensity projections and artificial intelligence: A win-win combination to easily measure prognostic biomarkers in DLBCL patients. J Nucl Med (2022).
Fujima, N. et al. Deep learning analysis using FDG-PET to predict treatment outcome in patients with oral cavity squamous cell carcinoma. Eur Radiol 30, 6322-6330 (2020).
Guo, R. et al. Weakly supervised deep learning for determining the prognostic value of (18)F-FDG PET/CT in extranodal natural killer/T cell lymphoma, nasal type. Eur J Nucl Med Mol Imaging 48, 3151-3161 (2021).
Takehiko, F. et al. Easy detection of tumor in oncologic whole-body PET by projection reconstruction images with maximum intensity projection algorithm. Ann Nucl Med 13, 199–203 (1999).
Boellaard, R. et al. Artificial Intelligence based outcome classification from baseline 18F-FDG PET/CT in de novo diffuse large B-cell lymphoma patients. European association of nuclear medicine october 20-23, 2021 virtual. Eur J Nucl Med Mol Imaging 48, 348 (2021).
Lugtenburg, P. J. et al. Rituximab-chop with early rituximab intensification for diffuse large B-cell lymphoma: A randomized phase III trial of the HOVON and the NORDIC lymphoma group (HOVON-84). J Clin Oncol 38, 3377-3387 (2020).
Duhrsen, U. et al. Positron emission tomography-guided therapy of aggressive non-hodgkin lymphomas (PETAL): A multicenter, randomized phase III trial. J Clin Oncol 36, 2024-2034 (2018).
Eertink, J. J. et al. Optimal timing and criteria of interim pet in DLBCL: A comparative study of 1692 patients. Blood Adv 5, 2375-2384 (2021).
Boellaard, R. Quantitative oncology molecular analysis suite: Accurate. J Nucl Med 59, 1753 (2018).
Barrington, S. F. et al. Automated segmentation of baseline metabolic total tumor burden in diffuse large B-cell lymphoma: Which method is most successful? A study on behalf of the PETRA consortium. J Nucl Med 62, 332-337 (2021).
Otsu, N. A threshold selection method from gray-level histograms. IEEE Trans. Systems Man Cybernet 9 (1979).
Eertink, J. J. et al. External validation: A simulation study to compare cross-validation versus holdout or external testing to assess the performance of clinical prediction models using PET data from DLBCL patients. EJNMMI Res 12, 58 (2022).
Levy, P. S. Clinical epidemiology: The essentials. JAMA 250, 1469-1469 (1983).
Corti, C. et al. Artificial intelligence for prediction of treatment outcomes in breast cancer: Systematic review of design, reporting standards, and bias. Cancer Treat Rev 108, 102410 (2022).
Frood, R. et al. Baseline PET/CT imaging parameters for prediction of treatment outcome in hodgkin and diffuse large B cell lymphoma: A systematic review. Eur J Nucl Med Mol Imaging 48, 3198-3220 (2021).
Jha, A. K. et al. Nuclear medicine and artificial intelligence: Best practices for evaluation (the relaince guidelines). J Nucl Med 63, 1288-1299 (2022).
Ferrandez, M. C. et al. Combatting the effect of image reconstruction settings on lymphoma [(18)F]FDG PET metabolic tumor volume assessment using various segmentation methods. EJNMMI Res 12, 44 (2022).
Eertink, J. J. et al. Quantitative radiomics features in diffuse large B-cell lymphoma: Does segmentation method matter? J Nucl Med 63, 389-395 (2022).
Driessen, J. et al. The impact of semi-automatic segmentation methods on metabolic tumor volume, intensity and dissemination radiomics in (18)F-FDG PET scans of patients with classical hodgkin lymphoma. J Nucl Med 63, 1424–1430 (2022).
Liu, P., Zhang, M., Gao, X., Li, B. & Zheng, G. Joint lymphoma lesion segmentation and prognosis prediction from baseline FDG PET images via multitask convolutional neural networks. IEEE Access 10, 81612-81623 (2022).
Rebaud, L. et al. Multitask learning-to-rank neural network for predicting survival of diffuse large b-cell lymphoma patients from their unsegmented baseline [18F]FDG PET/CT scans. J Nucl Med 63, 3250-3250 (2022).
Kubilius, J., Bracci, S. & Op de Beeck, H. P. Deep neural networks as a computational model for human shape sensitivity. PLoS Comput Biol 12, e1004896 (2016).

Competing interest reported. This work was financially supported by the Hanarth Fonds Fund and the Dutch Cancer Society (#VU-2018-11648). M.C.F., S.S.V.G, J.J.E., B.M.d.V., S.E.W., G.J.C.Z., S.P., L.K., A.H., C.H., U.D., H.C.W.d.V. and R.B. declare no competing financial interests. P.J.L. received research funding from Takeda, Servier and Roche and received honoraria for advisory boards from Takeda, Servier, Genentech, Genmab, Celgene, Incyte and AbbVie. J.M.Z. received research funding from Roche and received honoraria for advisory boards from Takeda, Gilead, BMS and Roche. No other potential conflicts of interest relevant to this article exist.

SupplementalMaterialScientificReports.docx

Download PDF

Journal Publication

published 11 Aug, 2023

Read the published version in Scientific Reports →

Editorial decision: Major revision
21 Jul, 2023
Reviews received at journal
20 Jul, 2023
Reviewers agreed at journal
11 Jul, 2023
Reviews received at journal
06 Jul, 2023
Reviewers agreed at journal
26 Jun, 2023
Reviewers invited by journal
25 May, 2023
Editor assigned by journal
25 May, 2023
Editor invited by journal
05 Apr, 2023
Submission checks completed at journal
05 Apr, 2023
First submitted to journal
31 Mar, 2023

You are reading this latest preprint version

An artificial intelligence method using ¹⁸F-FDG PET maximum intensity projections to predict 2-year time-to-progression in diffuse large B-cell lymphoma patients

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials And Methods

Datasets

Maximum Intensity Projections

Data sampling scheme

Convolutional Neural Network

Plausibility of the CNN

Statistical Analysis

Results

Internal validation

External validation

Plausibility of the CNN

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1