Fully automatic segmentation of diffuse large B cell lymphoma lesions on 3D FDG-PET/CT for total metabolic tumour volume prediction using a convolutional neural network

Automatic detection and segmentation of lesions in 18 F-FDG PET/CT imaging of patients with Hodgkin lymphoma using 3D dense U-Net

OBJECTIVE

The accuracy of automatic tumor segmentation in PET/computed tomography (PET/CT) images is crucial for the effective treatment and monitoring of Hodgkin lymphoma. This study aims to address the challenges faced by certain segmentation algorithms in accurately differentiating lymphoma from normal organ uptakes due to PET image resolution and tumor heterogeneity.

MATERIALS AND METHODS

Variants of the encoder-decoder architectures are state-of-the-art models for image segmentation. Among these kinds of architectures, U-Net is one of the most famous and predominant for medical image segmentation. In this study, we propose a fully automatic approach for Hodgkin lymphoma segmentation that combines U-Net and DenseNet architectures to reduce network loss for very small lesions, which is trained using the Tversky loss function. The hypothesis is that the fusion of these two deep learning models can improve the accuracy and robustness of Hodgkin lymphoma segmentation. A dataset with 141 samples was used to train our proposed network. Also, to test and evaluate the proposed network, we allocated two separate datasets of 20 samples.

RESULTS

We achieved 0.759 as the mean Dice similarity coefficient with a median value of 0.767, and interquartile range (0.647-0.837). A good agreement was observed between the ground truth of test images against the predicted volume with precision and recall scores of 0.798 and 0.763, respectively.

CONCLUSION

This study demonstrates that the integration of U-Net and DenseNet architectures, along with the Tversky loss function, can significantly enhance the accuracy of Hodgkin lymphoma segmentation in PET/CT images compared to similar studies.

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). Methods: 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, SUVpeak, age, and performance status. The PET model included metabolic tumor volume, maximum distance from the bulkiest lesion to another lesion, and SUVpeak Model performance was assessed using the area under the curve (AUC) and Kaplan-Meier curves. Results: The IPI yielded an AUC of 0.60 on all external data. The deep learning model yielded a significantly higher AUC of 0.66 (P < 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; P > 0.05). The PET model yielded the highest AUC of all models (AUC, 0.71; P < 0.05). Conclusion: 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.

Enhancing Lymphoma Diagnosis, Treatment, and Follow-Up Using 18F-FDG PET/CT Imaging: Contribution of Artificial Intelligence and Radiomics Analysis

Lymphoma, encompassing a wide spectrum of immune system malignancies, presents significant complexities in its early detection, management, and prognosis assessment since it can mimic post-infectious/inflammatory diseases. The heterogeneous nature of lymphoma makes it challenging to definitively pinpoint valuable biomarkers for predicting tumor biology and selecting the most effective treatment strategies. Although molecular imaging modalities, such as positron emission tomography/computed tomography (PET/CT), specifically 18F-FDG PET/CT, hold significant importance in the diagnosis of lymphoma, prognostication, and assessment of treatment response, they still face significant challenges. Over the past few years, radiomics and artificial intelligence (AI) have surfaced as valuable tools for detecting subtle features within medical images that may not be easily discerned by visual assessment. The rapid expansion of AI and its application in medicine/radiomics is opening up new opportunities in the nuclear medicine field. Radiomics and AI capabilities seem to hold promise across various clinical scenarios related to lymphoma. Nevertheless, the need for more extensive prospective trials is evident to substantiate their reliability and standardize their applications. This review aims to provide a comprehensive perspective on the current literature regarding the application of AI and radiomics applied/extracted on/from 18F-FDG PET/CT in the management of lymphoma patients.

Clinical Meaningfulness of an Algorithm-Based Service for Analyzing Treatment Response in Patients with Metastatic Cancer Using FDG PET/CT

Background/Objectives: Determining how a patient with metastatic cancer is responding to therapy can be difficult for medical oncologists, especially with text-only radiology reports. In this investigation, we assess the clinical usefulness of a new algorithm-based analysis that provides spatial location and quantification for each detected lesion region of interest (ROI) and compare it to information included in radiology reports in the United States. Methods: Treatment response radiology reports for FDG PET/CT scans were retrospectively gathered from 228 patients with metastatic cancers. Each radiology report was assessed for the presence of both qualitative and quantitative information. A subset of patients (N = 103) was further analyzed using an algorithm-based service that provides the clinician with comprehensive quantitative information, including change over time, of all detected ROI with visualization of anatomical location. For each patient, three medical oncologists from different practices independently rated the usefulness of the additional analysis overall and in four subcategories. Results: In the 228 radiology reports, quantitative information of size and uptake was provided for at least one lesion at one time point in 78% (size) and 95% (uptake) of patients. This information was reported for both analyzed time points (current scan and previous comparator) in 52% (size) and 66% (uptake) of patients. Only 7% of reports quantified the total number of lesions, and none of the reports quantified changes in all lesions for patients with more than a few lesions. In the assessment of the augmentative algorithm-based analysis, the majority of oncologists rated it as overall useful for 98% of patients (101/103). Within specific categories of use, the majority of oncologists voted to use it for making decisions regarding systemic therapy in 97% of patients, for targeted therapy decisions in 72% of patients, for spatial location information in 96% of patients, and for patient education purposes in 93% of patients. Conclusions: For patients with metastatic cancer, the algorithm-based analysis of all ROI would allow oncologists to better understand treatment response and support their work to more precisely optimize the patient's therapy.

Role of Novel Quantitative Imaging Techniques in Hematological Malignancies

Hematological malignancies exhibit a widespread distribution, necessitating evaluation of disease activity over the entire body. In clinical practice, visual analysis and semiquantitative parameters are used to assess 18F-FDGPET/CT imaging, which solely represents measurements of disease activity from limited area and may not adequately reflect global disease assessment. An efficient method for assessing the global disease burden of hematological malignancies is to employ PET/computed tomography based novel quantitative parameters. In this article, we explored novel quantitative parameters on PET/CT imaging for assessing global disease burden and the potential role of artificial intelligence (AI) to determine these parameters in evaluation of hematological malignancies.

A deep learning method for predicting the origins of cervical lymph node metastatic cancer on digital pathological images

The metastatic cancer of cervical lymph nodes presents complex shapes and poses significant challenges for doctors in determining its origin. We established a deep learning framework to predict the status of lymph nodes in patients with cervical lymphadenopathy (CLA) by hematoxylin and eosin (H&E) stained slides. This retrospective study utilized 1,036 cervical lymph node biopsy specimens at the First Affiliated Hospital of Sun Yat-Sen University (FAHSYSU). A multiple-instance learning algorithm designed for key region identification was applied, and cross-validation experiments were conducted in the dataset. Additionally, the model distinguished between primary lymphoma and metastatic cancer with high prediction accuracy. We also validated our model and other models on an external dataset. Our model showed better generalization and achieved the best results on both internal and external datasets. This algorithm offers an approach for evaluating cervical lymph node status before surgery, significantly aiding physicians in preoperative diagnosis and treatment planning.

Semi-supervised learning towards automated segmentation of PET images with limited annotations: application to lymphoma patients

Manual segmentation poses a time-consuming challenge for disease quantification, therapy evaluation, treatment planning, and outcome prediction. Convolutional neural networks (CNNs) hold promise in accurately identifying tumor locations and boundaries in PET scans. However, a major hurdle is the extensive amount of supervised and annotated data necessary for training. To overcome this limitation, this study explores semi-supervised approaches utilizing unlabeled data, specifically focusing on PET images of diffuse large B-cell lymphoma (DLBCL) and primary mediastinal large B-cell lymphoma (PMBCL) obtained from two centers. We considered 2-[18F]FDG PET images of 292 patients PMBCL (n = 104) and DLBCL (n = 188) (n = 232 for training and validation, and n = 60 for external testing). We harnessed classical wisdom embedded in traditional segmentation methods, such as the fuzzy clustering loss function (FCM), to tailor the training strategy for a 3D U-Net model, incorporating both supervised and unsupervised learning approaches. Various supervision levels were explored, including fully supervised methods with labeled FCM and unified focal/Dice loss, unsupervised methods with robust FCM (RFCM) and Mumford-Shah (MS) loss, and semi-supervised methods combining FCM with supervised Dice loss (MS + Dice) or labeled FCM (RFCM + FCM). The unified loss function yielded higher Dice scores (0.73 ± 0.11; 95% CI 0.67-0.8) than Dice loss (p value < 0.01). Among the semi-supervised approaches, RFCM + αFCM (α = 0.3) showed the best performance, with Dice score of 0.68 ± 0.10 (95% CI 0.45-0.77), outperforming MS + αDice for any supervision level (any α) (p < 0.01). Another semi-supervised approach with MS + αDice (α = 0.2) achieved Dice score of 0.59 ± 0.09 (95% CI 0.44-0.76) surpassing other supervision levels (p < 0.01). Given the time-consuming nature of manual delineations and the inconsistencies they may introduce, semi-supervised approaches hold promise for automating medical imaging segmentation workflows.

The impact of introducing deep learning based [18F]FDG PET denoising on EORTC and PERCIST therapeutic response assessments in digital PET/CT

BACKGROUND

[18F]FDG PET denoising by SubtlePET™ using deep learning artificial intelligence (AI) was previously found to induce slight modifications in lesion and reference organs' quantification and in lesion detection. As a next step, we aimed to evaluate its clinical impact on [18F]FDG PET solid tumour treatment response assessments, while comparing "standard PET" to "AI denoised half-duration PET" ("AI PET") during follow-up.

RESULTS

110 patients referred for baseline and follow-up standard digital [18F]FDG PET/CT were prospectively included. "Standard" EORTC and, if applicable, PERCIST response classifications by 2 readers between baseline standard PET1 and follow-up standard PET2 as a "gold standard" were compared to "mixed" classifications between standard PET1 and AI PET2 (group 1; n = 64), or between AI PET1 and standard PET2 (group 2; n = 46). Separate classifications were established using either standardized uptake values from ultra-high definition PET with or without AI denoising (simplified to "UHD") or EANM research limited v2 (EARL2)-compliant values (by Gaussian filtering in standard PET and using the same filter in AI PET). Overall, pooling both study groups, in 11/110 (10%) patients at least one EORTCUHD or EARL2 or PERCISTUHD or EARL2 mixed vs. standard classification was discordant, with 369/397 (93%) concordant classifications, unweighted Cohen's kappa = 0.86 (95% CI: 0.78-0.94). These modified mixed vs. standard classifications could have impacted management in 2% of patients.

CONCLUSIONS

Although comparing similar PET images is preferable for therapy response assessment, the comparison between a standard [18F]FDG PET and an AI denoised half-duration PET is feasible and seems clinically satisfactory.

What clinicians should know about surrogate end points in hematologic malignancies

ABSTRACT

Use of surrogates as primary end points is commonplace in hematology/oncology clinical trials. As opposed to prognostic markers, surrogates are end points that can be measured early and yet can still capture the full effect of treatment, because it would be captured by the true outcome (eg, overall survival). We discuss the level of evidence of the most commonly used end points in hematology and share recommendations on how to apply and evaluate surrogate end points in research and clinical practice. Based on the statistical literature, this clinician-friendly review intends to build a bridge between clinicians and surrogacy specialists.

Multicenter PET image harmonization using generative adversarial networks

PURPOSE

To improve reproducibility and predictive performance of PET radiomic features in multicentric studies by cycle-consistent generative adversarial network (GAN) harmonization approaches.

METHODS

GAN-harmonization was developed to harmonize whole-body PET scans to perform image style and texture translation between different centers and scanners. GAN-harmonization was evaluated by application to two retrospectively collected open datasets and different tasks. First, GAN-harmonization was performed on a dual-center lung cancer cohort (127 female, 138 male) where the reproducibility of radiomic features in healthy liver tissue was evaluated. Second, GAN-harmonization was applied to a head and neck cancer cohort (43 female, 154 male) acquired from three centers. Here, the clinical impact of GAN-harmonization was analyzed by predicting the development of distant metastases using a logistic regression model incorporating first-order statistics and texture features from baseline 18F-FDG PET before and after harmonization.

RESULTS

Image quality remained high (structural similarity: left kidney ≥ 0.800, right kidney ≥ 0.806, liver ≥ 0.780, lung ≥ 0.838, spleen ≥ 0.793, whole-body ≥ 0.832) after image harmonization across all utilized datasets. Using GAN-harmonization, inter-site reproducibility of radiomic features in healthy liver tissue increased at least by ≥ 5 ± 14% (first-order), ≥ 16 ± 7% (GLCM), ≥ 19 ± 5% (GLRLM), ≥ 16 ± 8% (GLSZM), ≥ 17 ± 6% (GLDM), and ≥ 23 ± 14% (NGTDM). In the head and neck cancer cohort, the outcome prediction improved from AUC 0.68 (95% CI 0.66-0.71) to AUC 0.73 (0.71-0.75) by application of GAN-harmonization.

CONCLUSIONS

GANs are capable of performing image harmonization and increase reproducibility and predictive performance of radiomic features derived from different centers and scanners.

TMTV-Net: fully automated total metabolic tumor volume segmentation in lymphoma PET/CT images - a multi-center generalizability analysis

PURPOSE

Total metabolic tumor volume (TMTV) segmentation has significant value enabling quantitative imaging biomarkers for lymphoma management. In this work, we tackle the challenging task of automated tumor delineation in lymphoma from PET/CT scans using a cascaded approach.

METHODS

Our study included 1418 2-[18F]FDG PET/CT scans from four different centers. The dataset was divided into 900 scans for development/validation/testing phases and 518 for multi-center external testing. The former consisted of 450 lymphoma, lung cancer, and melanoma scans, along with 450 negative scans, while the latter consisted of lymphoma patients from different centers with diffuse large B cell, primary mediastinal large B cell, and classic Hodgkin lymphoma cases. Our approach involves resampling PET/CT images into different voxel sizes in the first step, followed by training multi-resolution 3D U-Nets on each resampled dataset using a fivefold cross-validation scheme. The models trained on different data splits were ensemble. After applying soft voting to the predicted masks, in the second step, we input the probability-averaged predictions, along with the input imaging data, into another 3D U-Net. Models were trained with semi-supervised loss. We additionally considered the effectiveness of using test time augmentation (TTA) to improve the segmentation performance after training. In addition to quantitative analysis including Dice score (DSC) and TMTV comparisons, the qualitative evaluation was also conducted by nuclear medicine physicians.

RESULTS

Our cascaded soft-voting guided approach resulted in performance with an average DSC of 0.68 ± 0.12 for the internal test data from developmental dataset, and an average DSC of 0.66 ± 0.18 on the multi-site external data (n = 518), significantly outperforming (p < 0.001) state-of-the-art (SOTA) approaches including nnU-Net and SWIN UNETR. While TTA yielded enhanced performance gains for some of the comparator methods, its impact on our cascaded approach was found to be negligible (DSC: 0.66 ± 0.16). Our approach reliably quantified TMTV, with a correlation of 0.89 with the ground truth (p < 0.001). Furthermore, in terms of visual assessment, concordance between quantitative evaluations and clinician feedback was observed in the majority of cases. The average relative error (ARE) and the absolute error (AE) in TMTV prediction on external multi-centric dataset were ARE = 0.43 ± 0.54 and AE = 157.32 ± 378.12 (mL) for all the external test data (n = 518), and ARE = 0.30 ± 0.22 and AE = 82.05 ± 99.78 (mL) when the 10% outliers (n = 53) were excluded.

CONCLUSION

TMTV-Net demonstrates strong performance and generalizability in TMTV segmentation across multi-site external datasets, encompassing various lymphoma subtypes. A negligible reduction of 2% in overall performance during testing on external data highlights robust model generalizability across different centers and cancer types, likely attributable to its training with resampled inputs. Our model is publicly available, allowing easy multi-site evaluation and generalizability analysis on datasets from different institutions.

Total metabolic tumor volume on 18F-FDG PET/CT is a game-changer for patients with metastatic lung cancer treated with immunotherapy

PURPOSE

Because of atypical response imaging patterns in patients with metastatic non-small cell lung cancer (NSCLC) treated with immune checkpoint inhibitors (ICPIs), new biomarkers are needed for a better monitoring of treatment efficacy. The aim of this prospective study was to evaluate the prognostic value of volume-derived positron-emission tomography (PET) parameters on baseline and follow-up 18F-fluoro-deoxy-glucose PET (18F-FDG-PET) scans and compare it with the conventional PET Response Criteria in Solid Tumors (PERCIST).

METHODS

Patients with metastatic NSCLC were included in two different single-center prospective trials. 18F-FDG-PET studies were performed before the start of immunotherapy (PETbaseline), after 6-8 weeks (PETinterim1) and after 12-16 weeks (PETinterim2) of treatment, using PERCIST criteria for tumor response assessment. Different metabolic parameters were evaluated: absolute values of maximum standardized uptake value (SUVmax) of the most intense lesion, total metabolic tumor volume (TMTV), total lesion glycolysis (TLG), but also their percentage changes between PET studies (ΔSUVmax, ΔTMTV and ΔTLG). The median follow-up of patients was 31 (7.3-31.8) months. Prognostic values and optimal thresholds of PET parameters were estimated by ROC (Receiver Operating Characteristic) curve analysis of 12-month overall survival (12M-OS) and 6-month progression-free survival (6M-PFS). Tumor progression needed to be confirmed by a multidisciplinary tumor board, considering atypical response patterns on imaging.

RESULTS

110 patients were prospectively included. On PETbaseline, TMTV was predictive of 12M-OS [AUC (Area Under Curve) =0.64; 95% CI: 0.61 to 0.66] whereas SUVmax and TLG were not. On PETinterim1 and PETinterim2, all metabolic parameters were predictive for 12M-OS and 6M-PFS, the residual TMTV on PETinterim1 (TMTV1) being the strongest prognostic biomarker (AUC=0.83 and 0.82; 95% CI: 0.74 to 0.91, for 12M-OS and 6M-PFS, respectively). Using the optimal threshold by ROC curve to classify patients into three TMTV1 subgroups (0 cm3; 0-57 cm3; >57 cm3), TMTV1 prognostic stratification was independent of PERCIST criteria on both PFS and OS, and significantly outperformed them. Subgroup analysis demonstrated that TMTV1 remained a strong prognostic biomarker of 12M-OS for non-responding patients (p=0.0003) according to PERCIST criteria. In the specific group of patients with PERCIST progression on PETinterim1, low residual tumor volume (<57 cm3) was still associated with a very favorable patients' outcome (6M-PFS=73%; 24M-OS=55%).

CONCLUSION

The absolute value of residual metabolic tumor volume, assessed 6-8 weeks after the start of ICPI, is an optimal and independent prognostic measure, exceeding and complementing conventional PERCIST criteria. Oncologists should consider it in patients with first tumor progression according to PERCIST criteria, as it helps identify patients who benefit from continued treatment.

TRIAL REGISTRATION NUMBER

2018-A02116-49; NCT03584334.

Quarter-Century PET/CT Transformation of Oncology: Lymphoma

The clinical landscape of lymphomas has changed dramatically over the last 2 decades, including significant progress made in the understanding and utilization of imaging modalities and the available treatment options for both indolent and aggressive lymphomas. Since the introduction of hybrid PET/CT scanners in 2001, the indications of 18F-fluorodeoxyglucose (FDG) PET/CT in the management of lymphomas have grown rapidly. In today's clinical practice, FDG PET/CT is used in successful management of the vast majority patients with lymphomas.

Deep learning-based whole-body characterization of prostate cancer lesions on [68Ga]Ga-PSMA-11 PET/CT in patients with post-prostatectomy recurrence

PURPOSE

The automatic segmentation and detection of prostate cancer (PC) lesions throughout the body are extremely challenging due to the lesions' complexity and variability in appearance, shape, and location. In this study, we investigated the performance of a three-dimensional (3D) convolutional neural network (CNN) to automatically characterize metastatic lesions throughout the body in a dataset of PC patients with recurrence after radical prostatectomy.

METHODS

We retrospectively collected [68 Ga]Ga-PSMA-11 PET/CT images from 116 patients with metastatic PC at two centers: center 1 provided the data for fivefold cross validation (n = 78) and internal testing (n = 19), and center 2 provided the data for external testing (n = 19). PET and CT data were jointly input into a 3D U-Net to achieve whole-body segmentation and detection of PC lesions. The performance in both the segmentation and the detection of lesions throughout the body was evaluated using established metrics, including the Dice similarity coefficient (DSC) for segmentation and the recall, precision, and F1-score for detection. The correlation and consistency between tumor burdens (PSMA-TV and TL-PSMA) calculated from automatic segmentation and artificial ground truth were assessed by linear regression and Bland‒Altman plots.

RESULTS

On the internal test set, the DSC, precision, recall, and F1-score values were 0.631, 0.961, 0.721, and 0.824, respectively. On the external test set, the corresponding values were 0.596, 0.888, 0.792, and 0.837, respectively. Our approach outperformed previous studies in segmenting and detecting metastatic lesions throughout the body. Tumor burden indicators derived from deep learning and ground truth showed strong correlation (R2 ≥ 0.991, all P < 0.05) and consistency.

CONCLUSION

Our 3D CNN accurately characterizes whole-body tumors in relapsed PC patients; its results are highly consistent with those of manual contouring. This automatic method is expected to improve work efficiency and to aid in the assessment of tumor burden.

Deep learning for [18F]fluorodeoxyglucose-PET-CT classification in patients with lymphoma: a dual-centre retrospective analysis

BACKGROUND

The rising global cancer burden has led to an increasing demand for imaging tests such as [18F]fluorodeoxyglucose ([18F]FDG)-PET-CT. To aid imaging specialists in dealing with high scan volumes, we aimed to train a deep learning artificial intelligence algorithm to classify [18F]FDG-PET-CT scans of patients with lymphoma with or without hypermetabolic tumour sites.

METHODS

In this retrospective analysis we collected 16 583 [18F]FDG-PET-CTs of 5072 patients with lymphoma who had undergone PET-CT before or after treatment at the Memorial Sloa Kettering Cancer Center, New York, NY, USA. Using maximum intensity projection (MIP), three dimensional (3D) PET, and 3D CT data, our ResNet34-based deep learning model (Lymphoma Artificial Reader System [LARS]) for [18F]FDG-PET-CT binary classification (Deauville 1-3 vs 4-5), was trained on 80% of the dataset, and tested on 20% of this dataset. For external testing, 1000 [18F]FDG-PET-CTs were obtained from a second centre (Medical University of Vienna, Vienna, Austria). Seven model variants were evaluated, including MIP-based LARS-avg (optimised for accuracy) and LARS-max (optimised for sensitivity), and 3D PET-CT-based LARS-ptct. Following expert curation, areas under the curve (AUCs), accuracies, sensitivities, and specificities were calculated.

FINDINGS

In the internal test cohort (3325 PET-CTs, 1012 patients), LARS-avg achieved an AUC of 0·949 (95% CI 0·942-0·956), accuracy of 0·890 (0·879-0·901), sensitivity of 0·868 (0·851-0·885), and specificity of 0·913 (0·899-0·925); LARS-max achieved an AUC of 0·949 (0·942-0·956), accuracy of 0·868 (0·858-0·879), sensitivity of 0·909 (0·896-0·924), and specificity of 0·826 (0·808-0·843); and LARS-ptct achieved an AUC of 0·939 (0·930-0·948), accuracy of 0·875 (0·864-0·887), sensitivity of 0·836 (0·817-0·855), and specificity of 0·915 (0·901-0·927). In the external test cohort (1000 PET-CTs, 503 patients), LARS-avg achieved an AUC of 0·953 (0·938-0·966), accuracy of 0·907 (0·888-0·925), sensitivity of 0·874 (0·843-0·904), and specificity of 0·949 (0·921-0·960); LARS-max achieved an AUC of 0·952 (0·937-0·965), accuracy of 0·898 (0·878-0·916), sensitivity of 0·899 (0·871-0·926), and specificity of 0·897 (0·871-0·922); and LARS-ptct achieved an AUC of 0·932 (0·915-0·948), accuracy of 0·870 (0·850-0·891), sensitivity of 0·827 (0·793-0·863), and specificity of 0·913 (0·889-0·937).

INTERPRETATION

Deep learning accurately distinguishes between [18F]FDG-PET-CT scans of lymphoma patients with and without hypermetabolic tumour sites. Deep learning might therefore be potentially useful to rule out the presence of metabolically active disease in such patients, or serve as a second reader or decision support tool.

FUNDING

National Institutes of Health-National Cancer Institute Cancer Center Support Grant.

Tumor Location Relative to the Spleen Is a Prognostic Factor in Lymphoma Patients: A Demonstration from the REMARC Trial

Baseline [18F]FDG PET/CT radiomic features can improve the survival prediction in patients with diffuse large B-cell lymphoma (DLBCL). The purpose of this study was to investigate whether characterizing tumor locations relative to the spleen location in baseline [18F]FDG PET/CT images predicts survival in patients with DLBCL and improves the predictive value of total metabolic tumor volume (TMTV) and age-adjusted international prognostic index (IPI). Methods: This retrospective study included 301 DLBCL patients from the REMARC (NCT01122472) cohort. Physicians delineated the tumor regions, whereas the spleen was automatically segmented using an open-access artificial intelligence algorithm. We systematically measured the distance between the centroid of the spleen and all other lesions, defining the SD of these distances as the lesion spread (SpreadSpleen). We calculated the maximum distance between the spleen and another lesion (Dspleen) for each patient and normalized it with the body surface area, resulting in standardized Dspleen (sDspleen). The predictive value of each PET/CT feature for progression-free survival (PFS) and overall survival (OS) was evaluated through univariate and multivariate time-dependent Cox models and Kaplan-Meier analysis. Results: In total, 282 patients (mean age, 68.33 ± 5.41 y; 164 men) were evaluated. The artificial intelligence algorithm successfully segmented the spleen in 96% of the patients. SpreadSpleen, Dspleen, and sDspleen were correlated neither with TMTV (Pearson ρ < 0.23) nor with IPI (Pearson ρ < 0.15). When median values were used as the cutoff, SpreadSpleen, Dspleen, and sDspleen all significantly classified patients into 2 risk groups for PFS and OS (P < 0.001). They complemented TMTV and IPI to classify the patients into 3 risk groups for PFS and OS (P < 0.001). Integrating SpreadSpleen, Dspleen, or sDspleen into a Cox model on the basis of TMTV, IPI, and TMTV combined with IPI significantly improved the concordance index for PFS and OS (P < 0.05). Conclusion: Baseline PET/CT features that characterize tumor spread and dissemination relative to the spleen strongly predicted survival in patients with DLBCL. Integrating these features with TMTV and IPI further improved survival prediction.

Deep convolutional neural network for differentiating between sarcoidosis and lymphoma based on [18F]FDG maximum-intensity projection images

OBJECTIVES

To compare the [18F]FDG PET/CT findings of untreated sarcoidosis and malignant lymphoma (ML) and develop convolutional neural network (CNN) models to differentiate between these diseases using maximum intensity projection (MIP) [18F]FDG PET images.

METHODS

We retrospectively collected data on consecutive patients newly diagnosed with sarcoidosis and ML who underwent [18F]FDG PET/CT before treatment. Two nuclear radiologists reviewed the images. CNN models were created using MIP PET images and evaluated with k-fold cross-validation. The points of interest were visualized using gradient-weighted class activation mapping (Grad-CAM).

RESULTS

A total of 56 patients with sarcoidosis and 62 patients with ML were included. Patients with sarcoidosis had more prominent FDG accumulation in the mediastinal lymph nodes and lung lesions, while those with ML had more prominent accumulation in the cervical lymph nodes (all p < 0.001). For the mediastinal lymph nodes, sarcoidosis patients had significant FDG accumulation in the level 2, 4, 7, and 10 lymph nodes (all p < 0.01). Otherwise, the accumulation in ML patients tended to be in the level 1 lymph nodes (p = 0.08). The CNN model using frontal and lateral MIP images achieved an average accuracy of 0.890 (95% CI: 0.804-0.977), a sensitivity of 0.898 (95% CI: 0.782-1.000), a specificity of 0.907 (95% CI: 0.799-1.000), and an area under the curve of 0.963 (95% CI: 0.899-1.000). Grad-CAM showed that the model focused on the sites of abnormal FDG accumulation.

CONCLUSIONS

CNN models based on differences in FDG accumulation sites archive high performance in differentiating between sarcoidosis and ML.

CLINICAL RELEVANCE STATEMENT

We developed a CNN model using MIP images of [18F]FDG PET/CT to distinguish between sarcoidosis and malignant lymphoma. It achieved high performance and could be useful in diagnosing diseases with involvement across organs and lymph nodes.

KEY POINTS

• There are differences in FDG distribution when comparing whole-body [18F]FDG PET/CT findings in patients with sarcoidosis and malignant lymphoma before treatment. • Convolutional neural networks, a type of deep learning technique, trained with maximum-intensity projection PET images from two angles showed high performance. • A deep learning model that utilizes differences in FDG distribution may be helpful in differentiating between diseases with lesions that are characteristically widespread among organs and lymph nodes.

Review of artificial intelligence clinical applications in Nuclear Medicine

This paper provides an in-depth analysis of the clinical applications of artificial intelligence (AI) in Nuclear Medicine, focusing on three key areas: neurology, cardiology, and oncology. Beginning with neurology, specifically Alzheimer's disease and Parkinson's disease, the paper examines reviews on diagnosis and treatment planning. The same pattern is followed in cardiology studies. In the final section on oncology, the paper explores the various AI applications in multiple cancer types, including lung, head and neck, lymphoma, and pancreatic cancer.

Application of machine learning in the management of lymphoma: Current practice and future prospects

In the past decade, digitization of medical records and multiomics data analysis in lymphoma has led to the accessibility of high-dimensional records. The digitization of medical records, the visualization of extensive volume data extracted from medical images, and the integration of multiomics methods into clinical decision-making have produced many datasets. As a promising auxiliary tool, machine learning (ML) intends to extract homologous features in large-scale data sets and encode them into various patterns to complete complicated tasks. At present, artificial intelligence and digital mining have shown promising prospects in the field of lymphoma pathological image analysis. The paradigm shift from qualitative analysis to quantitative analysis makes the pathological diagnosis more intelligent and the results more accurate and objective. ML can promote accurate lymphoma diagnosis and provide patients with prognostic information and more individualized treatment options. Based on the above, this comprehensive review of the general workflow of ML highlights recent advances in ML techniques in the diagnosis, treatment, and prognosis of lymphoma, and clarifies the boundedness and future orientation of the ML technique in the clinical practice of lymphoma.

Segmentation of Dynamic Total-Body [18F]-FDG PET Images Using Unsupervised Clustering

Clustering time activity curves of PET images have been used to separate clinically relevant areas of the brain or tumours. However, PET image segmentation in multiorgan level is much less studied due to the available total-body data being limited to animal studies. Now, the new PET scanners providing the opportunity to acquire total-body PET scans also from humans are becoming more common, which opens plenty of new clinically interesting opportunities. Therefore, organ-level segmentation of PET images has important applications, yet it lacks sufficient research. In this proof of concept study, we evaluate if the previously used segmentation approaches are suitable for segmenting dynamic human total-body PET images in organ level. Our focus is on general-purpose unsupervised methods that are independent of external data and can be used for all tracers, organisms, and health conditions. Additional anatomical image modalities, such as CT or MRI, are not used, but the segmentation is done purely based on the dynamic PET images. The tested methods are commonly used building blocks of the more sophisticated methods rather than final methods as such, and our goal is to evaluate if these basic tools are suited for the arising human total-body PET image segmentation. First, we excluded methods that were computationally too demanding for the large datasets from human total-body PET scanners. These criteria filtered out most of the commonly used approaches, leaving only two clustering methods, k-means and Gaussian mixture model (GMM), for further analyses. We combined k-means with two different preprocessing approaches, namely, principal component analysis (PCA) and independent component analysis (ICA). Then, we selected a suitable number of clusters using 10 images. Finally, we tested how well the usable approaches segment the remaining PET images in organ level, highlight the best approaches together with their limitations, and discuss how further research could tackle the observed shortcomings. In this study, we utilised 40 total-body [18F] fluorodeoxyglucose PET images of rats to mimic the coming large human PET images and a few actual human total-body images to ensure that our conclusions from the rat data generalise to the human data. Our results show that ICA combined with k-means has weaker performance than the other two computationally usable approaches and that certain organs are easier to segment than others. While GMM performed sufficiently, it was by far the slowest one among the tested approaches, making k-means combined with PCA the most promising candidate for further development. However, even with the best methods, the mean Jaccard index was slightly below 0.5 for the easiest tested organ and below 0.2 for the most challenging organ. Thus, we conclude that there is a lack of accurate and computationally light general-purpose segmentation method that can analyse dynamic total-body PET images.

Dapagliflozin Attenuates Heart Failure With Preserved Ejection Fraction Remodeling and Dysfunction by Elevating β-Hydroxybutyrate-activated Citrate Synthase

ABSTRACT

Heart failure with preserved ejection fraction (HFpEF) is highly prevalent, accounting for 50% of all heart failure patients, and is associated with significant mortality. Sodium-glucose cotransporter subtype inhibitor (SGLT2i) is recommended in the AHA and ESC guidelines for the treatment of HFpEF, but the mechanism of SGLT2i to prevent and treat cardiac remodeling and dysfunction is currently unknown, hindering the understanding of the pathophysiology of HFpEF and the development of novel therapeutics. HFpEF model was induced by a high-fat diet (60% calories from lard) + N [w] -nitro- l -arginine methyl ester ( l -NAME-0.5 g/L) (2 Hit) in male Sprague Dawley rats to effectively recapture the myriad phenotype of HFpEF. This study's results showed that administration of dapagliflozin (DAPA, SGLT2 inhibitor) significantly limited the 2-Hit-induced cardiomyocyte hypertrophy, apoptosis, inflammation, oxidative stress, and fibrosis. It also improved cardiac diastolic and systolic dysfunction in a late-stage progression of HFpEF. Mechanistically, DAPA influences energy metabolism associated with fatty acid intake and mitochondrial dysfunction in HFpEF by increasing β-hydroxybutyric acid (β-OHB) levels, directing the activation of citrate synthase, reducing acetyl coenzyme A (acetyl-CoA) pools, modulating adenosine 5'-triphosphate production, and increasing the expression of mitochondrial oxidative phosphorylation system complexes I-V. In addition, following clinical DAPA therapy, the blood levels of β-OHB and citrate synthase increased and the levels of acetyl-CoA in the blood of HFpEF patients decreased. SGLT2i plays a beneficial role in the prevention and treatment of cardiac remodeling and dysfunction in HFpEF model by attenuating cardiometabolic dysregulation.

Whole-body tumor segmentation from PET/CT images using a two-stage cascaded neural network with camouflaged object detection mechanisms

BACKGROUND

Whole-body Metabolic Tumor Volume (MTVwb) is an independent prognostic factor for overall survival in lung cancer patients. Automatic segmentation methods have been proposed for MTV calculation. Nevertheless, most of existing methods for patients with lung cancer only segment tumors in the thoracic region.

PURPOSE

In this paper, we present a Two-Stage cascaded neural network integrated with Camouflaged Object Detection mEchanisms (TS-Code-Net) for automatic segmenting tumors from whole-body PET/CT images.

METHODS

Firstly, tumors are detected from the Maximum Intensity Projection (MIP) images of PET/CT scans, and tumors' approximate localizations along z-axis are identified. Secondly, the segmentations are performed on PET/CT slices that contain tumors identified by the first step. Camouflaged object detection mechanisms are utilized to distinguish the tumors from their surrounding regions that have similar Standard Uptake Values (SUV) and texture appearance. Finally, the TS-Code-Net is trained by minimizing the total loss that incorporates the segmentation accuracy loss and the class imbalance loss.

RESULTS

The performance of the TS-Code-Net is tested on a whole-body PET/CT image data-set including 480 Non-Small Cell Lung Cancer (NSCLC) patients with five-fold cross-validation using image segmentation metrics. Our method achieves 0.70, 0.76, and 0.70, for Dice, Sensitivity and Precision, respectively, which demonstrates the superiority of the TS-Code-Net over several existing methods related to metastatic lung cancer segmentation from whole-body PET/CT images.

CONCLUSIONS

The proposed TS-Code-Net is effective for whole-body tumor segmentation of PET/CT images. Codes for TS-Code-Net are available at: https://github.com/zyj19/TS-Code-Net.

Artificial Intelligence in Oncological Hybrid Imaging

BACKGROUND

 Artificial intelligence (AI) applications have become increasingly relevant across a broad spectrum of settings in medical imaging. Due to the large amount of imaging data that is generated in oncological hybrid imaging, AI applications are desirable for lesion detection and characterization in primary staging, therapy monitoring, and recurrence detection. Given the rapid developments in machine learning (ML) and deep learning (DL) methods, the role of AI will have significant impact on the imaging workflow and will eventually improve clinical decision making and outcomes.

METHODS AND RESULTS

 The first part of this narrative review discusses current research with an introduction to artificial intelligence in oncological hybrid imaging and key concepts in data science. The second part reviews relevant examples with a focus on applications in oncology as well as discussion of challenges and current limitations.

CONCLUSION

 AI applications have the potential to leverage the diagnostic data stream with high efficiency and depth to facilitate automated lesion detection, characterization, and therapy monitoring to ultimately improve quality and efficiency throughout the medical imaging workflow. The goal is to generate reproducible, structured, quantitative diagnostic data for evidence-based therapy guidance in oncology. However, significant challenges remain regarding application development, benchmarking, and clinical implementation.

KEY POINTS

  · Hybrid imaging generates a large amount of multimodality medical imaging data with high complexity and depth.. · Advanced tools are required to enable fast and cost-efficient processing along the whole radiology value chain.. · AI applications promise to facilitate the assessment of oncological disease in hybrid imaging with high quality and efficiency for lesion detection, characterization, and response assessment. The goal is to generate reproducible, structured, quantitative diagnostic data for evidence-based oncological therapy guidance.. · Selected applications in three oncological entities (lung, prostate, and neuroendocrine tumors) demonstrate how AI algorithms may impact imaging-based tasks in hybrid imaging and potentially guide clinical decision making..

Sensitivity of an AI method for [18F]FDG PET/CT outcome prediction of diffuse large B-cell lymphoma patients to image reconstruction protocols

BACKGROUND

Convolutional 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).

RESULTS

CNN 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]).

CONCLUSION

Our 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.

Machine Learning CT-Based Automatic Nodal Segmentation and PET Semi-Quantification of Intraoperative 68Ga-PSMA-11 PET/CT Images in High-Risk Prostate Cancer: A Pilot Study

High-resolution intraoperative PET/CT specimen imaging, coupled with prostate-specific membrane antigen (PSMA) molecular targeting, holds great potential for the rapid ex vivo identification of disease localizations in high-risk prostate cancer patients undergoing surgery. However, the accurate analysis of radiotracer uptake would require time-consuming manual volumetric segmentation of 3D images. The aim of this study was to test the feasibility of using machine learning to perform automatic nodal segmentation of intraoperative 68Ga-PSMA-11 PET/CT specimen images. Six (n = 6) lymph-nodal specimens were imaged in the operating room after an e.v. injection of 2.1 MBq/kg of 68Ga-PSMA-11. A machine learning-based approach for automatic lymph-nodal segmentation was developed using only open-source Python libraries (Scikit-learn, SciPy, Scikit-image). The implementation of a k-means clustering algorithm (n = 3 clusters) allowed to identify lymph-nodal structures by leveraging differences in tissue density. Refinement of the segmentation masks was performed using morphological operations and 2D/3D-features filtering. Compared to manual segmentation (ITK-SNAP v4.0.1), the automatic segmentation model showed promising results in terms of weighted average precision (97-99%), recall (68-81%), Dice coefficient (80-88%) and Jaccard index (67-79%). Finally, the ML-based segmentation masks allowed to automatically compute semi-quantitative PET metrics (i.e., SUVmax), thus holding promise for facilitating the semi-quantitative analysis of PET/CT images in the operating room.

3D PET/CT tumor segmentation based on nnU-Net with GCN refinement

Objective. Whole-body positron emission tomography/computed tomography (PET/CT) scans are an important tool for diagnosing various malignancies (e.g. malignant melanoma, lymphoma, or lung cancer), and accurate segmentation of tumors is a key part of subsequent treatment. In recent years, convolutional neural network based segmentation methods have been extensively investigated. However, these methods often give inaccurate segmentation results, such as oversegmentation and undersegmentation. To address these issues, we propose a postprocessing method based on a graph convolutional network (GCN) to refine inaccurate segmentation results and improve the overall segmentation accuracy.Approach. First, nnU-Net is used as an initial segmentation framework, and the uncertainty in the segmentation results is analyzed. Certain and uncertain pixels are used to establish the nodes of a graph. Each node and its 6 neighbors form an edge, and 32 nodes are randomly selected as uncertain nodes to form edges. The highly uncertain nodes are used as the subsequent refinement targets. Second, the nnU-Net results of the certain nodes are used as labels to form a semisupervised graph network problem, and the uncertain part is optimized by training the GCN to improve the segmentation performance. This describes our proposed nnU-Net + GCN segmentation framework.Main results.We perform tumor segmentation experiments with the PET/CT dataset from the MICCIA2022 autoPET challenge. Among these data, 30 cases are randomly selected for testing, and the experimental results show that the false-positive rate is effectively reduced with nnU-Net + GCN refinement. In quantitative analysis, there is an improvement of 2.1% for the average Dice score, 6.4 for the 95% Hausdorff distance (HD95), and 1.7 for the average symmetric surface distance.Significance. The quantitative and qualitative evaluation results show that GCN postprocessing methods can effectively improve the tumor segmentation performance.

Quantitative PET-based biomarkers in lymphoma: getting ready for primetime

The use of functional quantitative biomarkers extracted from routine PET-CT scans to characterize clinical responses in patients with lymphoma is gaining increased attention, and these biomarkers can outperform established clinical risk factors. Total metabolic tumour volume enables individualized estimation of survival outcomes in patients with lymphoma and has shown the potential to predict response to therapy suitable for  risk-adapted treatment approaches in clinical trials. The deployment of machine learning tools in molecular imaging research can assist in recognizing complex patterns and, with image classification, in tumour identification and segmentation of data from PET-CT scans. Initial studies using fully automated approaches to calculate metabolic tumour volume and other PET-based biomarkers have demonstrated appropriate correlation with calculations from experts, warranting further testing in large-scale studies. The extraction of computer-based quantitative tumour characterization through radiomics can provide a comprehensive view of phenotypic heterogeneity that better captures the molecular and functional features of the disease. Additionally, radiomics can be integrated with genomic data to provide more accurate prognostic information. Further improvements in PET-based biomarkers are imminent, although their incorporation into clinical decision-making currently has methodological shortcomings that need to be addressed with confirmatory prospective validation in selected patient populations. In this Review, we discuss the current knowledge, challenges and opportunities in the integration of quantitative PET-based biomarkers in clinical trials and the routine management of patients with lymphoma.

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

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.

Evaluation of Semiautomatic and Deep Learning-Based Fully Automatic Segmentation Methods on [18F]FDG PET/CT Images from Patients with Lymphoma: Influence on Tumor Characterization

The objective is to assess the performance of seven semiautomatic and two fully automatic segmentation methods on [18F]FDG PET/CT lymphoma images and evaluate their influence on tumor quantification. All lymphoma lesions identified in 65 whole-body [18F]FDG PET/CT staging images were segmented by two experienced observers using manual and semiautomatic methods. Semiautomatic segmentation using absolute and relative thresholds, k-means and Bayesian clustering, and a self-adaptive configuration (SAC) of k-means and Bayesian was applied. Three state-of-the-art deep learning-based segmentations methods using a 3D U-Net architecture were also applied. One was semiautomatic and two were fully automatic, of which one is publicly available. Dice coefficient (DC) measured segmentation overlap, considering manual segmentation the ground truth. Lymphoma lesions were characterized by 31 features. Intraclass correlation coefficient (ICC) assessed features agreement between different segmentation methods. Nine hundred twenty [18F]FDG-avid lesions were identified. The SAC Bayesian method achieved the highest median intra-observer DC (0.87). Inter-observers' DC was higher for SAC Bayesian than manual segmentation (0.94 vs 0.84, p < 0.001). Semiautomatic deep learning-based median DC was promising (0.83 (Obs1), 0.79 (Obs2)). Threshold-based methods and publicly available 3D U-Net gave poorer results (0.56 ≤ DC ≤ 0.68). Maximum, mean, and peak standardized uptake values, metabolic tumor volume, and total lesion glycolysis showed excellent agreement (ICC ≥ 0.92) between manual and SAC Bayesian segmentation methods. The SAC Bayesian classifier is more reproducible and produces similar lesion features compared to manual segmentation, giving the best concordant results of all other methods. Deep learning-based segmentation can achieve overall good segmentation results but failed in few patients impacting patients' clinical evaluation.

Advances in positron emission tomography and radiomics

Positron emission tomography is established for staging and response evaluation in lymphoma using visual evaluation and semi-quantitative analysis. Radiomic analysis involving quantitative imaging features at baseline, such as metabolic tumor volume and markers of disease dissemination and changes in the standardized uptake value during treatment are emerging as powerful biomarkers. The combination of radiomic features with clinical risk factors and genomic analysis offers the potential to improve clinical risk prediction. This review discusses the state of current knowledge, progress toward standardization of tumor delineation for radiomic analysis and argues that radiomic features, molecular markers and circulating tumor DNA should be included in clinical trial designs to enable the development of baseline and dynamic risk scores that could further advance the field to facilitate testing of novel treatments and personalized therapy in aggressive lymphomas.

PET/CT in Non-Hodgkin Lymphoma: An Update

Non-Hodgkin lymphomas represents a heterogeneous group of lymphoproliferative disorders characterized by different clinical courses, varying from indolent to highly aggressive. ¹⁸F-FDG-PET/CT is the current state-of-the-art diagnostic imaging, for the staging, restaging and evaluation of response to treatment in lymphomas with avidity for ¹⁸F-FDG, despite it is not routinely recommended for surveillance. PET-based response criteria (using five-point Deauville Score) are nowadays uniformly applied in FDG-avid lymphomas. In this review, a comprehensive overview of the role of ¹⁸F-FDG-PET in Non-Hodgkin lymphomas is provided, at each relevant point of patient management, particularly focusing on recent advances on diffuse large B-cell lymphoma and follicular lymphoma, with brief updates also on other histotypes (such as marginal zone, mantle cell, primary mediastinal- B cell lymphoma and T cell lymphoma). PET-derived semiquantitative factors useful for patient stratification and prognostication and emerging radiomics research are also presented.

Role of Artificial Intelligence in PET/CT Imaging for Management of Lymphoma

Our review shows that AI-based analysis of lymphoma whole-body FDG-PET/CT can inform all phases of clinical management including staging, prognostication, treatment planning, and treatment response evaluation. We highlight advancements in the role of neural networks for performing automated image segmentation to calculate PET-based imaging biomarkers such as the total metabolic tumor volume (TMTV). AI-based image segmentation methods are at levels where they can be semi-automatically implemented with minimal human inputs and nearing the level of a second-opinion radiologist. Advances in automated segmentation methods are particularly apparent in the discrimination of lymphomatous vs non-lymphomatous FDG-avid regions, which carries through to automated staging. Automated TMTV calculators, in addition to automated calculation of measures such as Dmax are informing robust models of progression-free survival which can then feed into improved treatment planning.

Baseline circulating tumour DNA and total metabolic tumour volume as early outcome predictors in aggressive large B-cell lymphoma. A real-world 112-patient cohort

Approximately 20%-50% of patients with large B-cell lymphoma (LBCL) experience poor outcomes. We aimed to evaluate the combined prognostic value of circulating tumour DNA (ctDNA) and total metabolic tumour volume (TMTV) in LBCL. This observational single-centre study included 112 newly diagnosed LBCL patients, receiving R-CHOP/R-CHOP-like chemotherapies. CtDNA load was calculated following next-generation sequencing of cell-free DNA (cfDNA) using a targeted 40-gene lymphopanel. TMTV was measured using a fully automated artificial intelligence-based method for lymphoma lesion segmentation. CtDNA was detected in cfDNA samples from 95 patients with a median concentration of 3.15 log haploid genome equivalents per mL. TMTV measurements were available for 102 patients. The median TMTV was 501 mL. High ctDNA load (>3.57 log hGE/mL) or high TMTV (>200 mL) were associated with shorter 1-year PFS (44% vs. 83%, p < 0.001 and 64% vs. 97%, p = 0.002, respectively). When combined, three prognostic groups were identified. The shortest PFS was observed when both TMTV and ctDNA load were high (p < 0.001). Even with a short follow up, combining ctDNA load with TMTV improved the risk stratification of patients with aggressive LBCL. In the near future, very high-risk patients could benefit from CAR T-cell therapy or bispecific antibodies as first-line treatments.

Clinical application of AI-based PET images in oncological patients

Based on the advantages of revealing the functional status and molecular expression of tumor cells, positron emission tomography (PET) imaging has been performed in numerous types of malignant diseases for diagnosis and monitoring. However, insufficient image quality, the lack of a convincing evaluation tool and intra- and interobserver variation in human work are well-known limitations of nuclear medicine imaging and restrict its clinical application. Artificial intelligence (AI) has gained increasing interest in the field of medical imaging due to its powerful information collection and interpretation ability. The combination of AI and PET imaging potentially provides great assistance to physicians managing patients. Radiomics, an important branch of AI applied in medical imaging, can extract hundreds of abstract mathematical features of images for further analysis. In this review, an overview of the applications of AI in PET imaging is provided, focusing on image enhancement, tumor detection, response and prognosis prediction and correlation analyses with pathology or specific gene mutations in several types of tumors. Our aim is to describe recent clinical applications of AI-based PET imaging in malignant diseases and to focus on the description of possible future developments.

Technologist-Based Implementation of Total Metabolic Tumor Volume into Clinical Practice

Metabolic tumor volume (MTV) is defined as the total metabolically active tumor volume seen on ¹⁸F-FDG PET/CT examinations. Calculating MTV is often time-consuming, requiring a high degree of manual input. In this study, the MTV calculations of a board-certified nuclear radiologist were compared with those of 2 nuclear medicine technologists. As part of the technologists' educational program, after their classroom time they were trained by the radiologist for 30 min. The technologists calculated MTV within 7.5% of the radiologist's calculations in a set of patients who had diffuse large B-cell lymphoma and were undergoing initial staging ¹⁸F-FDG PET/CT. These findings suggest that nuclear medicine technologists may help accelerate implementation of MTV into clinical practice with favorable accuracy, possibly as an initial step followed by validation by the interpreting physician. The aim of this study was to explore whether efficiency is improved by integrating nuclear medicine technologists into a semiautomated workflow to calculate total MTV.

Artificial Intelligence in Oncological Hybrid Imaging

BACKGROUND

Artificial intelligence (AI) applications have become increasingly relevant across a broad spectrum of settings in medical imaging. Due to the large amount of imaging data that is generated in oncological hybrid imaging, AI applications are desirable for lesion detection and characterization in primary staging, therapy monitoring, and recurrence detection. Given the rapid developments in machine learning (ML) and deep learning (DL) methods, the role of AI will have significant impact on the imaging workflow and will eventually improve clinical decision making and outcomes.

METHODS AND RESULTS

The first part of this narrative review discusses current research with an introduction to artificial intelligence in oncological hybrid imaging and key concepts in data science. The second part reviews relevant examples with a focus on applications in oncology as well as discussion of challenges and current limitations.

CONCLUSION

AI applications have the potential to leverage the diagnostic data stream with high efficiency and depth to facilitate automated lesion detection, characterization, and therapy monitoring to ultimately improve quality and efficiency throughout the medical imaging workflow. The goal is to generate reproducible, structured, quantitative diagnostic data for evidence-based therapy guidance in oncology. However, significant challenges remain regarding application development, benchmarking, and clinical implementation.

KEY POINTS

· Hybrid imaging generates a large amount of multimodality medical imaging data with high complexity and depth.. · Advanced tools are required to enable fast and cost-efficient processing along the whole radiology value chain.. · AI applications promise to facilitate the assessment of oncological disease in hybrid imaging with high quality and efficiency for lesion detection, characterization, and response assessment. The goal is to generate reproducible, structured, quantitative diagnostic data for evidence-based oncological therapy guidance.. · Selected applications in three oncological entities (lung, prostate, and neuroendocrine tumors) demonstrate how AI algorithms may impact imaging-based tasks in hybrid imaging and potentially guide clinical decision making..

CITATION FORMAT

· Feuerecker B, Heimer M, Geyer T et al. Artificial Intelligence in Oncological Hybrid Imaging. Fortschr Röntgenstr 2023; 195: 105 - 114.

Semi-automated 18F-FDG PET segmentation methods for tumor volume determination in Non-Hodgkin lymphoma patients: a literature review, implementation and multi-threshold evaluation

In the treatment of Non-Hodgkin lymphoma (NHL), multiple therapeutic options are available. Improving outcome predictions are essential to optimize treatment. The metabolic active tumor volume (MATV) has shown to be a prognostic factor in NHL. It is usually retrieved using semi-automated thresholding methods based on standardized uptake values (SUV), calculated from ¹⁸F-Fluorodeoxyglucose Positron Emission Tomography (¹⁸F-FDG PET) images. However, there is currently no consensus method for NHL. The aim of this study was to review literature on different segmentation methods used, and to evaluate selected methods by using an in house created software tool. A software tool, MUltiple SUV Threshold (MUST)-segmenter was developed where tumor locations are identified by placing seed-points on the PET images, followed by subsequent region growing. Based on a literature review, 9 SUV thresholding methods were selected and MATVs were extracted. The MUST-segmenter was utilized in a cohort of 68 patients with NHL. Differences in MATVs were assessed with paired t-tests, and correlations and distributions figures. High variability and significant differences between the MATVs based on different segmentation methods (p < 0.05) were observed in the NHL patients. Median MATVs ranged from 35 to 211 cc. No consensus for determining MATV is available based on the literature. Using the MUST-segmenter with 9 selected SUV thresholding methods, we demonstrated a large and significant variation in MATVs. Identifying the most optimal segmentation method for patients with NHL is essential to further improve predictions of toxicity, response, and treatment outcomes, which can be facilitated by the MUST-segmenter.

Evaluation of a convolution neural network for baseline total tumor metabolic volume on [18F]FDG PET in diffuse large B cell lymphoma

OBJECTIVES

New PET data-processing tools allow for automatic lesion selection and segmentation by a convolution neural network using artificial intelligence (AI) to obtain total metabolic tumor volume (TMTV) and total lesion glycolysis (TLG) routinely at the clinical workstation. Our objective was to evaluate an AI implemented in a new version of commercial software to verify reproducibility of results and time savings in a daily workflow.

METHODS

Using the software to obtain TMTV and TLG, two nuclear physicians applied five methods to retrospectively analyze data for 51 patients. Methods 1 and 2 were fully automated with exclusion of lesions ≤ 0.5 mL and ≤ 0.1 mL, respectively. Methods 3 and 4 were fully automated with physician review. Method 5 was semi-automated and used as reference. Time and number of clicks to complete the measurement were recorded for each method. Inter-instrument and inter-observer variation was assessed by the intra-class coefficient (ICC) and Bland-Altman plots.

RESULTS

Between methods 3 and 5, for the main user, the ICC was 0.99 for TMTV and 1.0 for TLG. Between the two users applying method 3, ICC was 0.97 for TMTV and 0.99 for TLG. Mean processing time (± standard deviation) was 20 s ± 9.0 for method 1, 178 s ± 125.7 for method 3, and 326 s ± 188.6 for method 5 (p < 0.05).

CONCLUSION

AI-enabled lesion detection software offers an automated, fast, reliable, and consistently performing tool for obtaining TMTV and TLG in a daily workflow.

KEY POINTS

• Our study shows that artificial intelligence lesion detection software is an automated, fast, reliable, and consistently performing tool for obtaining total metabolic tumor volume and total lesion glycolysis in a daily workflow.

The application of machine learning to imaging in hematological oncology: A scoping review

Background

Here, we conducted a scoping review to (i) establish which machine learning (ML) methods have been applied to hematological malignancy imaging; (ii) establish how ML is being applied to hematological cancer radiology; and (iii) identify addressable research gaps.

Methods

The review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Extension for Scoping Reviews guidelines. The inclusion criteria were (i) pediatric and adult patients with suspected or confirmed hematological malignancy undergoing imaging (population); (ii) any study using ML techniques to derive models using radiological images to apply to the clinical management of these patients (concept); and (iii) original research articles conducted in any setting globally (context). Quality Assessment of Diagnostic Accuracy Studies 2 criteria were used to assess diagnostic and segmentation studies, while the Newcastle-Ottawa scale was used to assess the quality of observational studies.

Results

Of 53 eligible studies, 33 applied diverse ML techniques to diagnose hematological malignancies or to differentiate them from other diseases, especially discriminating gliomas from primary central nervous system lymphomas (n=18); 11 applied ML to segmentation tasks, while 9 applied ML to prognostication or predicting therapeutic responses, especially for diffuse large B-cell lymphoma. All studies reported discrimination statistics, but no study calculated calibration statistics. Every diagnostic/segmentation study had a high risk of bias due to their case-control design; many studies failed to provide adequate details of the reference standard; and only a few studies used independent validation.

Conclusion

To deliver validated ML-based models to radiologists managing hematological malignancies, future studies should (i) adhere to standardized, high-quality reporting guidelines such as the Checklist for Artificial Intelligence in Medical Imaging; (ii) validate models in independent cohorts; (ii) standardize volume segmentation methods for segmentation tasks; (iv) establish comprehensive prospective studies that include different tumor grades, comparisons with radiologists, optimal imaging modalities, sequences, and planes; (v) include side-by-side comparisons of different methods; and (vi) include low- and middle-income countries in multicentric studies to enhance generalizability and reduce inequity.

Multi-scale feature similarity-based weakly supervised lymphoma segmentation in PET/CT images

Accurate lymphoma segmentation in PET/CT images is important for evaluating Diffuse Large B-Cell Lymphoma (DLBCL) prognosis. As systemic multiple lymphomas, DLBCL lesions vary in number and size for different patients, which makes DLBCL labeling labor-intensive and time-consuming. To reduce the reliance on accurately labeled datasets, a weakly supervised deep learning method based on multi-scale feature similarity is proposed for automatic lymphoma segmentation. Weak labeling was performed by randomly dawning a small and salient lymphoma volume for the patient without accurate labels. A 3D V-Net is used as the backbone of the segmentation network and image features extracted in different convolutional layers are fused with the Atrous Spatial Pyramid Pooling (ASPP) module to generate multi-scale feature representations of input images. By imposing multi-scale feature consistency constraints on the predicted tumor regions as well as the labeled tumor regions, weakly labeled data can also be effectively used for network training. The cosine similarity, which has strong generalization, is exploited here to measure feature distances. The proposed method is evaluated with a PET/CT dataset of 147 lymphoma patients. Experimental results show that when using data, half of which have accurate labels and the other half have weak labels, the proposed method performed similarly to a fully supervised segmentation network and achieved an average Dice Similarity Coefficient (DSC) of 71.47%. The proposed method is able to reduce the requirement for expert annotations in deep learning-based lymphoma segmentation.

18F-FDG PET Maximum-Intensity Projections and Artificial Intelligence: A Win-Win Combination to Easily Measure Prognostic Biomarkers in DLBCL Patients

Total metabolic tumor volume (TMTV) and tumor dissemination (Dmax) calculated from baseline ¹⁸F-FDG PET/CT images are prognostic biomarkers in diffuse large B-cell lymphoma (DLBCL) patients. Yet, their automated calculation remains challenging. The purpose of this study was to investigate whether TMTV and Dmax features could be replaced by surrogate features automatically calculated using an artificial intelligence (AI) algorithm from only 2 maximum-intensity projections (MIPs) of the whole-body ¹⁸F-FDG PET images. Methods: Two cohorts of DLBCL patients from the REMARC (NCT01122472) and LNH073B (NCT00498043) trials were retrospectively analyzed. Experts delineated lymphoma lesions from the baseline whole-body ¹⁸F-FDG PET/CT images, from which TMTV and Dmax were measured. Coronal and sagittal MIP images and associated 2-dimensional reference lesion masks were calculated. An AI algorithm was trained on the REMARC MIP data to segment lymphoma regions. The AI algorithm was then used to estimate surrogate TMTV (sTMTV) and surrogate Dmax (sDmax) on both datasets. The ability of the original and surrogate TMTV and Dmax to stratify patients was compared. Results: Three hundred eighty-two patients (mean age ± SD, 62.1 y ± 13.4 y; 207 men) were evaluated. sTMTV was highly correlated with TMTV for REMARC and LNH073B datasets (Spearman r = 0.878 and 0.752, respectively), and so were sDmax and Dmax (r = 0.709 and 0.714, respectively). The hazard ratios for progression free survival of volume and MIP-based features derived using AI were similar, for example, TMTV: 11.24 (95% CI: 2.10-46.20), sTMTV: 11.81 (95% CI: 3.29-31.77), and Dmax: 9.0 (95% CI: 2.53-23.63), sDmax: 12.49 (95% CI: 3.42-34.50). Conclusion: Surrogate TMTV and Dmax calculated from only 2 PET MIP images are prognostic biomarkers in DLBCL patients and can be automatically estimated using an AI algorithm.

Interim FDG-PET analysis to identify patients with aggressive non-Hodgkin lymphoma who benefit from treatment intensification: a post-hoc analysis of the PETAL trial

The randomized PETAL trial failed to demonstrate a benefit of interim FDG-PET (iPET)-based treatment intensification over continued standard therapy with CHOP (plus rituximab (R) in CD20-positive lymphomas). We hypothesized that PET analysis of all lymphoma manifestations may identify patients who benefitted from treatment intensification. A previously developed neural network was employed for iPET analysis to identify the highest pathological FDG uptake (max-SUVAI) and the mean FDG uptake of all lymphoma manifestations (mean-SUVAI). High mean-SUVAI uptake was determined separately for iPET-positive and iPET-negative patients. The endpoint was time-to-progression (TTP). There was a significant interaction of additional rituximab and mean-SUVAI in the iPET-negative group (HR = 0.6, p < 0.05). Patients with high mean-SUVAI had significantly prolonged TTP when treated with 6xR-CHOP + 2 R (not reached versus 52 months, p < 0.05), whereas max-SUVmanual failed to show an impact of additional rituximab. In the iPET-positive group, patients with high mean-SUVAI had a significantly longer TTP with (R-)CHOP than with the Burkitt protocol (14 versus 4 months, p < 0.01). Comprehensive iPET evaluation may provide new prognosticators in aggressive lymphoma. Additional application of rituximab was associated with prolonged TTP in iPET-negative patients with high mean-SUVAI. Comprehensive iPET interpretation could identify high-risk patients who benefit from study-specific interventions.

PSR-Nets: Deep neural networks with prior shift regularization for PET/CT based automatic, accurate, and calibrated whole-body lymphoma segmentation

Lymphoma is a type of lymphatic tissue originated cancer. Automatic and accurate lymphoma segmentation is critical for its diagnosis and prognosis yet challenging due to the severely class-imbalanced problem. Generally, deep neural networks trained with class-observation-frequency based re-weighting loss functions are used to address this problem. However, the majority class can be under-weighted by them, due to the existence of data overlap. Besides, they are more mis-calibrated. To resolve these, we propose a neural network with prior-shift regularization (PSR-Net), which comprises a UNet-like backbone with re-weighting loss functions, and a prior-shift regularization (PSR) module including a prior-shift layer (PSL), a regularizer generation layer (RGL), and an expected prediction confidence updating layer (EPCUL). We first propose a trainable expected prediction confidence (EPC) for each class. Periodically, PSL shifts a prior training dataset to a more informative dataset based on EPCs; RGL presents a generalized informative-voxel-aware (GIVA) loss with EPCs and calculates it on the informative dataset for model finetuning in back-propagation; and EPCUL updates EPCs to refresh PSL and RRL in next forward-propagation. PSR-Net is trained in a two- stage manner. The backbone is first trained with re-weighting loss functions, then we reload the best saved model for the backbone and continue to train it with the weighted sum of the re-weighting loss functions, the GIVA regularizer and the L2 loss function of EPCs for regularization fine-tuning. Extensive experiments are performed based on PET/CT volumes with advanced stage lymphomas. Our PSR-Net achieves 95.12% sensitivity and 87.18% Dice coefficient, demonstrating the effectiveness of PSR-Net, when compared to the baselines and the state-of-the-arts.

A Deep Learning-Aided Automated Method for Calculating Metabolic Tumor Volume in Diffuse Large B-Cell Lymphoma

Metabolic tumor volume (MTV) is a robust prognostic biomarker in diffuse large B-cell lymphoma (DLBCL). The available semiautomatic software for calculating MTV requires manual input limiting its routine application in clinical research. Our objective was to develop a fully automated method (AM) for calculating MTV and to validate the method by comparing its results with those from two nuclear medicine (NM) readers. The automated method designed for this study employed a deep convolutional neural network to segment normal physiologic structures from the computed tomography (CT) scans that demonstrate intense avidity on positron emission tomography (PET) scans. The study cohort consisted of 100 patients with newly diagnosed DLBCL who were randomly selected from the Alliance/CALGB 50,303 (NCT00118209) trial. We observed high concordance in MTV calculations between the AM and readers with Pearson's correlation coefficients and interclass correlations comparing reader 1 to AM of 0.9814 (p < 0.0001) and 0.98 (p < 0.001; 95%CI = 0.96 to 0.99), respectively; and comparing reader 2 to AM of 0.9818 (p < 0.0001) and 0.98 (p < 0.0001; 95%CI = 0.96 to 0.99), respectively. The Bland-Altman plots showed only relatively small systematic errors between the proposed method and readers for both MTV and maximum standardized uptake value (SUVmax). This approach may possess the potential to integrate PET-based biomarkers in clinical trials.

A whole-body FDG-PET/CT Dataset with manually annotated Tumor Lesions

We describe a publicly available dataset of annotated Positron Emission Tomography/Computed Tomography (PET/CT) studies. 1014 whole body Fluorodeoxyglucose (FDG)-PET/CT datasets (501 studies of patients with malignant lymphoma, melanoma and non small cell lung cancer (NSCLC) and 513 studies without PET-positive malignant lesions (negative controls)) acquired between 2014 and 2018 were included. All examinations were acquired on a single, state-of-the-art PET/CT scanner. The imaging protocol consisted of a whole-body FDG-PET acquisition and a corresponding diagnostic CT scan. All FDG-avid lesions identified as malignant based on the clinical PET/CT report were manually segmented on PET images in a slice-per-slice (3D) manner. We provide the anonymized original DICOM files of all studies as well as the corresponding DICOM segmentation masks. In addition, we provide scripts for image processing and conversion to different file formats (NIfTI, mha, hdf5). Primary diagnosis, age and sex are provided as non-imaging information. We demonstrate how this dataset can be used for deep learning-based automated analysis of PET/CT data and provide the trained deep learning model.

Full automation of total metabolic tumor volume from FDG-PET/CT in DLBCL for baseline risk assessments

Background

Current radiological assessments of ¹⁸fluorodeoxyglucose-positron emission tomography (FDG-PET) imaging data in diffuse large B-cell lymphoma (DLBCL) can be time consuming, do not yield real-time information regarding disease burden and organ involvement, and hinder the use of FDG-PET to potentially limit the reliance on invasive procedures (e.g. bone marrow biopsy) for risk assessment.

Methods

Our aim is to enable real-time assessment of imaging-based risk factors at a large scale and we propose a fully automatic artificial intelligence (AI)-based tool to rapidly extract FDG-PET imaging metrics in DLBCL. On availability of a scan, in combination with clinical data, our approach generates clinically informative risk scores with minimal resource requirements. Overall, 1268 patients with previously untreated DLBCL from the phase III GOYA trial (NCT01287741) were included in the analysis (training: n = 846; hold-out: n = 422).

Results

Our AI-based model comprising imaging and clinical variables yielded a tangible prognostic improvement compared to clinical models without imaging metrics. We observed a risk increase for progression-free survival (PFS) with hazard ratios [HR] of 1.87 (95% CI: 1.31–2.67) vs 1.38 (95% CI: 0.98–1.96) (C-index: 0.59 vs 0.55), and a risk increase for overall survival (OS) (HR: 2.16 (95% CI: 1.37–3.40) vs 1.40 (95% CI: 0.90–2.17); C-index: 0.59 vs 0.55). The combined model defined a high-risk population with 35% and 42% increased odds of a 4-year PFS and OS event, respectively, versus the International Prognostic Index components alone. The method also identified a subpopulation with a 2-year Central Nervous System (CNS)-relapse probability of 17.1%.

Conclusion

Our tool enables an enhanced risk stratification compared with IPI, and the results indicate that imaging can be used to improve the prediction of central nervous system relapse in DLBCL. These findings support integration of clinically informative AI-generated imaging metrics into clinical workflows to improve identification of high-risk DLBCL patients.

Trial Registration

Registered clinicaltrials.gov number: NCT01287741.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s40644-022-00476-0.

Automatic classification of lymphoma lesions in FDG-PET–Differentiation between tumor and non-tumor uptake

Introduction

The automatic classification of lymphoma lesions in PET is a main topic of ongoing research. An automatic algorithm would enable the swift evaluation of PET parameters, like texture and heterogeneity markers, concerning their prognostic value for patients outcome in large datasets. Moreover, the determination of the metabolic tumor volume would be facilitated. The aim of our study was the development and evaluation of an automatic algorithm for segmentation and classification of lymphoma lesions in PET.

Methods

Pre-treatment PET scans from 60 Hodgkin lymphoma patients from the EuroNet-PHL-C1 trial were evaluated. A watershed algorithm was used for segmentation. For standardization of the scan length, an automatic cropping algorithm was developed. All segmented volumes were manually classified into one of 14 categories. The random forest method and a nested cross-validation was used for automatic classification and evaluation.

Results

Overall, 853 volumes were segmented and classified. 203/246 tumor lesions and 554/607 non-tumor volumes were classified correctly by the automatic algorithm, corresponding to a sensitivity, a specificity, a positive and a negative predictive value of 83%, 91%, 79% and 93%. In 44/60 (73%) patients, all tumor lesions were correctly classified. In ten out of the 16 patients with misclassified tumor lesions, only one false-negative tumor lesion occurred. The automatic classification of focal gastrointestinal uptake, brown fat tissue and composed volumes consisting of more than one tissue was challenging.

Conclusion

Our algorithm, trained on a small number of patients and on PET information only, showed a good performance and is suitable for automatic lymphoma classification.

Deep Neural Networks and Machine Learning Radiomics Modelling for Prediction of Relapse in Mantle Cell Lymphoma

Simple Summary

Mantle cell lymphoma (MCL) is an aggressive lymphoid tumour with a poor prognosis. There exist no routine biomarkers for the early prediction of relapse. Our study compared the potential of radiomics-based machine learning and 3D deep learning models as non-invasive biomarkers to risk-stratify MCL patients, thus promoting precision imaging in clinical oncology.

Abstract

Mantle cell lymphoma (MCL) is a rare lymphoid malignancy with a poor prognosis characterised by frequent relapse and short durations of treatment response. Most patients present with aggressive disease, but there exist indolent subtypes without the need for immediate intervention. The very heterogeneous behaviour of MCL is genetically characterised by the translocation t(11;14)(q13;q32), leading to Cyclin D1 overexpression with distinct clinical and biological characteristics and outcomes. There is still an unfulfilled need for precise MCL prognostication in real-time. Machine learning and deep learning neural networks are rapidly advancing technologies with promising results in numerous fields of application. This study develops and compares the performance of deep learning (DL) algorithms and radiomics-based machine learning (ML) models to predict MCL relapse on baseline CT scans. Five classification algorithms were used, including three deep learning models (3D SEResNet50, 3D DenseNet, and an optimised 3D CNN) and two machine learning models based on K-nearest Neighbor (KNN) and Random Forest (RF). The best performing method, our optimised 3D CNN, predicted MCL relapse with a 70% accuracy, better than the 3D SEResNet50 (62%) and the 3D DenseNet (59%). The second-best performing method was the KNN-based machine learning model (64%) after principal component analysis for improved accuracy. Our optimised CNN developed by ourselves correctly predicted MCL relapse in 70% of the patients on baseline CT imaging. Once prospectively tested in clinical trials with a larger sample size, our proposed 3D deep learning model could facilitate clinical management by precision imaging in MCL.