PET segmentation of bulky tumors: Strategies and workflows to improve inter-observer variability

Identification of impactful imaging biomarker: Clinical applications for breast and prostate carcinoma

BACKGROUND

 Imaging biomarkers are quantitative parameters from imaging modalities, which are collected noninvasively, allow conclusions about physiological and pathophysiological processes, and may consist of single (monoparametric) or multiple parameters (bi- or multiparametric).

METHOD

 This review aims to present the state of the art for the quantification of multimodal and multiparametric imaging biomarkers. Here, the use of biomarkers using artificial intelligence will be addressed and the clinical application of imaging biomarkers in breast and prostate cancers will be explained. For the preparation of the review article, an extensive literature search was performed based on Pubmed, Web of Science and Google Scholar. The results were evaluated and discussed for consistency and generality.

RESULTS AND CONCLUSION

 Different imaging biomarkers (multiparametric) are quantified based on the use of complementary imaging modalities (multimodal) from radiology, nuclear medicine, or hybrid imaging. From these techniques, parameters are determined at the morphological (e. g., size), functional (e. g., vascularization or diffusion), metabolic (e. g., glucose metabolism), or molecular (e. g., expression of prostate specific membrane antigen, PSMA) level. The integration and weighting of imaging biomarkers are increasingly being performed with artificial intelligence, using machine learning algorithms. In this way, the clinical application of imaging biomarkers is increasing, as illustrated by the diagnosis of breast and prostate cancers.

KEY POINTS

· Imaging biomarkers are quantitative parameters to detect physiological and pathophysiological processes.. · Imaging biomarkers from multimodality and multiparametric imaging are integrated using artificial intelligence algorithms.. · Quantitative imaging parameters are a fundamental component of diagnostics for all tumor entities, such as for mammary and prostate carcinomas..

CITATION FORMAT

· Bäuerle T, Dietzel M, Pinker K et al. Identification of impactful imaging biomarker: Clinical applications for breast and prostate carcinoma. Fortschr Röntgenstr 2024; 196: 354 - 362.

ASE-Net: A tumor segmentation method based on image pseudo enhancement and adaptive-scale attention supervision module

Fluorine 18(¹⁸F) fluorodeoxyglucose positron emission tomography and Computed Tomography (PET/CT) is the preferred imaging method of choice for the diagnosis and treatment of many cancers. However, factors such as low-contrast organ and tissue images, and the original scale of tumors pose huge obstacles to the accurate segmentation of tumors. In this work, we propose a novel model ASE-Net which is used for multimodality tumor segmentation. Firstly, we propose a pseudo-enhanced CT image generation method based on metabolic intensity to generate pseudo-enhanced CT images as additional input, which reduces the learning of the network in the spatial position of PET/CT and increases the discriminability of the corresponding structural positions of the high and low metabolic region. Second, unlike previous networks that directly segment tumors of all scales, we propose an Adaptive-Scale Attention Supervision Module at the skip connections, after combining the results of all paths, tumors of different scales will be given different receptive fields. Finally, Dual Path Block is used as the backbone of our network to leverage the ability of residual learning for feature reuse and dense connection for exploring new features. Our experimental results on two clinical PET/CT datasets demonstrate the effectiveness of our proposed network and achieve 78.56% and 72.57% in Dice Similarity Coefficient, respectively, which has better performance compared to state-of-the-art network models, whether for large or small tumors. The proposed model will help pathologists formulate more accurate diagnoses by providing reference opinions during diagnosis, consequently improving patient survival rate.

Development of scalable lymphatic system in the 4D XCAT phantom: Application to quantitative evaluation of lymphoma PET segmentations

BACKGROUND

Digital anthropomorphic phantoms, such as the 4D extended cardiac-torso (XCAT) phantom, are actively used to develop, optimize, and evaluate a variety of imaging applications, allowing for realistic patient modeling and knowledge of ground truth. The XCAT phantom defines the activity and attenuation for a simulated patient, which includes a complete set of organs, muscle, bone, and soft tissue, while also accounting for cardiac and respiratory motion. However, the XCAT phantom does not currently include the lymphatic system, critical for evaluating medical imaging tasks such as sentinel node detection, node density measurement, and radiation dosimetry.

PURPOSE

In this study, we aimed to develop a scalable lymphatic system in the XCAT phantom, to facilitate improved research of the lymphatic system in medical imaging. Using this scalable lymphatic system, we modeled the lymph node conglomerate pathology that is characteristically observed in primary mediastinal B-cell lymphoma (PMBCL). As an extended application, we evaluated positron emission tomography (PET) image quantification of metabolic tumor volume (MTV) and total lesion glycolysis (TLG) of these simulated lymphomas, though the phantoms may be applied to other imaging modalities and study design paradigms (e.g., image quality, detection).

METHODS

A template model for the lymphatic system was developed based on anatomical data from the Visible Human Project of the National Library of Medicine. The segmented nodes and vessels were fit with non-uniform rational basis spline surfaces, and multichannel large deformation diffeomorphic metric mapping was used to propagate the template to different XCAT anatomies. To model conglomerates observed in PMBCL, lymph nodes were enlarged, converged within the mediastinum, and tracer concentration was increased. We used the phantoms as inputs to a PET simulation tool, which generated images using ordered subsets expectation maximization reconstruction with 2-8 mm Gaussian filters. Fixed thresholding (FT) and gradient segmentation were used to determine MTV and TLG. Percent bias (%Bias) and coefficient of variation (COV) were computed as measures of accuracy and precision, respectively, for each MTV and TLG measurement.

RESULTS

Using the methodology described above, we introduced a scalable lymphatic system in the XCAT phantom, which allows for the radioactivity and attenuation ground truth to be generated in 116 ± 2.5 s using a 2.3 GHz processor. Within the Rhinoceros interface, lymph node anatomy and function were modified to create a cohort of 10 phantoms with lymph node conglomerates. Using the lymphoma phantoms to evaluate PET quantification of MTV, mean %Bias values were -9.3%, -41.3%, and 20.9%, while COV values were 4.08%, 7.6%, and 3.4% using 25% FT, 40% FT, and gradient segmentations, respectively. Comparatively for TLG, mean %Bias values were -27.4%, -45.8%, and -16.0%, while COV values were 1.9%, 5.7%, and 1.4%, for the 25% FT, 40% FT, and gradient segmentations, respectively.

CONCLUSIONS

In this work, we upgraded the XCAT phantom to include a lymphatic system, comprised of a network of 276 scalable lymph nodes and corresponding vessels. As an application, we created a cohort of phantoms with lymph node conglomerates to evaluate lymphoma quantification in PET imaging, which highlights an important application of this work.

Standardised lesion segmentation for imaging biomarker quantitation: a consensus recommendation from ESR and EORTC

BACKGROUND

Lesion/tissue segmentation on digital medical images enables biomarker extraction, image-guided therapy delivery, treatment response measurement, and training/validation for developing artificial intelligence algorithms and workflows. To ensure data reproducibility, criteria for standardised segmentation are critical but currently unavailable.

METHODS

A modified Delphi process initiated by the European Imaging Biomarker Alliance (EIBALL) of the European Society of Radiology (ESR) and the European Organisation for Research and Treatment of Cancer (EORTC) Imaging Group was undertaken. Three multidisciplinary task forces addressed modality and image acquisition, segmentation methodology itself, and standards and logistics. Devised survey questions were fed via a facilitator to expert participants. The 58 respondents to Round 1 were invited to participate in Rounds 2-4. Subsequent rounds were informed by responses of previous rounds.

RESULTS/CONCLUSIONS

Items with ≥ 75% consensus are considered a recommendation. These include system performance certification, thresholds for image signal-to-noise, contrast-to-noise and tumour-to-background ratios, spatial resolution, and artefact levels. Direct, iterative, and machine or deep learning reconstruction methods, use of a mixture of CE marked and verified research tools were agreed and use of specified reference standards and validation processes considered essential. Operator training and refreshment were considered mandatory for clinical trials and clinical research. Items with a 60-74% agreement require reporting (site-specific accreditation for clinical research, minimal pixel number within lesion segmented, use of post-reconstruction algorithms, operator training refreshment for clinical practice). Items with ≤ 60% agreement are outside current recommendations for segmentation (frequency of system performance tests, use of only CE-marked tools, board certification of operators, frequency of operator refresher training). Recommendations by anatomical area are also specified.

Reliability of gradient-based segmentation for measuring metabolic parameters influenced by uptake time on 18F-PSMA-1007 PET/CT for prostate cancer

Purpose

To determine an optimal setting for functional contouring and quantification of prostate cancer lesions with minimal variation by evaluating metabolic parameters on 18F-PSMA-1007 PET/CT measured by threshold-based and gradient-based methods under the influence of varying uptake time.

Methods and materials

Dual time point PET/CT was chosen to mimic varying uptake time in clinical setting. Positive lesions of patients who presented with newly diagnosed disease or biochemical recurrence after total prostatectomy were reviewed retrospectively. Gradient-based and threshold-based tools at 40%, 50% and 60% of lesion SUVmax (MIM 6.9) were used to create contours on PET. Contouring was considered completed if the target lesion, with its hottest voxel, was delineated from background tissues and nearby lesions under criteria specific to their operations. The changes in functional tumour volume (FTV) and metabolic tumour burden (MTB, defined as the product of SUVmean and FTV) were analysed. Lesion uptake patterns (increase/decrease/stable) were determined by the percentage change in tumour SUVmax at ±10% limit.

Results

A total of 275 lesions (135 intra-prostatic lesions, 65 lymph nodes, 45 bone lesions and 30 soft tissue lesions in pelvic region) in 68 patients were included. Mean uptake time of early and delayed imaging were 94 and 144 minutes respectively. Threshold-based method using 40% to 60% delineated only 85 (31%), 110 (40%) and 137 (50%) of lesions which all were contoured by gradient-based method. Although the overall percentage change using threshold at 50% was the smallest among other threshold levels in FTV measurement, it was still larger than gradient-based method (median: 50%=-7.6% vs gradient=0%). The overall percentage increase in MTB of gradient-based method (median: 6.3%) was compatible with the increase in tumour SUVmax. Only a small proportion of intra-prostatic lesions (<2%), LN (<4%), bone lesions (0%) and soft tissue lesions (<4%) demonstrated decrease uptake patterns.

Conclusions

With a high completion rate, gradient-based method is reliable for prostate cancer lesion contouring on 18F-PSMA-1007 PET/CT. Under the influence of varying uptake time, it has smaller variation than threshold-based method for measuring volumetric parameters. Therefore, gradient-based method is recommended for tumour delineation and quantification on 18F-PSMA-1007 PET/CT.

Diagnosis of Heart Failure Complicated with Sleep Apnea Syndrome by Thoracic Computerized Tomography under Artificial Intelligence Algorithm

The aim of this study was to explore the application effect of thoracic computerized tomography (CT) under single threshold segmentation algorithm in the diagnosis of heart failure (HF) complicated with sleep apnea syndrome. 30 patients diagnosed with HF complicated with sleep apnea syndrome were chosen for the research. Another 30 patients without sleep apnea syndrome were selected as the control group, whose age, height, and weight were similar to those of the experimental group. Then, a model for thoracic CT image segmentation was proposed under the single threshold segmentation algorithm, and the faster region convolutional neural network (Faster RCNN) was applied to label the thoracic respiratory lesions. All the patients underwent thoracic CT examination, and the obtained images were processed using the algorithm model above. After that, the morphology of the patient's respiratory tract after treatment was observed. The results suggested that the improved single threshold segmentation algorithm was effective for the image segmentation of patient lesions, and the Faster RCNN could effectively finish the labeling of the lesion area in the CT image. The classification accuracy of the Faster RCNN was about 0.966, and the loss value was about 0.092. With CT scanning under the algorithm, it was found that the airway collapse of the posterior palatal area, retrolingual area, and laryngopharyngeal area of the sleep apnea syndrome patients was significantly greater than that of the control group (P < 0.05). But there was no significant difference of the collapse of the nasopharyngeal area between the two groups (P > 0.05). The single threshold segmentation algorithm had a better segmentation accuracy for thoracic CT images in patients with HF and sleep apnea syndrome, so it had a highly promising application prospect in the diagnosis of the disease.

Deep Learning With Radiomics for Disease Diagnosis and Treatment: Challenges and Potential

The high-throughput extraction of quantitative imaging features from medical images for the purpose of radiomic analysis, i.e., radiomics in a broad sense, is a rapidly developing and emerging research field that has been attracting increasing interest, particularly in multimodality and multi-omics studies. In this context, the quantitative analysis of multidimensional data plays an essential role in assessing the spatio-temporal characteristics of different tissues and organs and their microenvironment. Herein, recent developments in this method, including manually defined features, data acquisition and preprocessing, lesion segmentation, feature extraction, feature selection and dimension reduction, statistical analysis, and model construction, are reviewed. In addition, deep learning-based techniques for automatic segmentation and radiomic analysis are being analyzed to address limitations such as rigorous workflow, manual/semi-automatic lesion annotation, and inadequate feature criteria, and multicenter validation. Furthermore, a summary of the current state-of-the-art applications of this technology in disease diagnosis, treatment response, and prognosis prediction from the perspective of radiology images, multimodality images, histopathology images, and three-dimensional dose distribution data, particularly in oncology, is presented. The potential and value of radiomics in diagnostic and therapeutic strategies are also further analyzed, and for the first time, the advances and challenges associated with dosiomics in radiotherapy are summarized, highlighting the latest progress in radiomics. Finally, a robust framework for radiomic analysis is presented and challenges and recommendations for future development are discussed, including but not limited to the factors that affect model stability (medical big data and multitype data and expert knowledge in medical), limitations of data-driven processes (reproducibility and interpretability of studies, different treatment alternatives for various institutions, and prospective researches and clinical trials), and thoughts on future directions (the capability to achieve clinical applications and open platform for radiomics analysis).

A novel semiautomated method for background activity and biological tumour volume definition to improve standardisation of 18F-FET PET imaging in glioblastoma

Background

Multicentre clinical trials evaluating the role of ¹⁸F-Fluoroethyl-l-tyrosine (¹⁸F-FET) PET as a diagnostic biomarker in glioma management have highlighted a need for standardised methods of data analysis. ¹⁸F-FET uptake normalised against background in the contralateral brain is a standard imaging technique to delineate the biological tumour volume (BTV). Quantitative analysis of ¹⁸F-FET PET images requires a consistent and robust background activity. Currently, defining background activity involves the manual selection of an arbitrary region of interest, a process that is subject to large variability. This study aims to eliminate methodological errors in background activity definition through the introduction of a semiautomated method for region of interest selection. A new method for background activity definition, involving the semiautomated generation of mirror-image (MI) reference regions, was compared with the current state-of-the-art method, involving manually drawing crescent-shape (gCS) reference regions. The MI and gCS methods were tested by measuring values of background activity and resulting BTV of ¹⁸F-FET PET scans of ten patients with recurrent glioblastoma multiforme generated from inputs provided by seven readers. To assess intra-reader variability, each scan was evaluated six times by each reader. Intra- and inter-reader variability in background activity and BTV definition was assessed by means of coefficient of variation.

Results

Compared to the gCS method, the MI method showed significantly lower intra- and inter-reader variability both in background activity and in BTV definition.

Conclusions

The proposed semiautomated MI method minimises intra- and inter-reader variability, providing a valuable approach for standardisation of ¹⁸F-FET PET quantitative parameters.

Trial registration ANZCTR, ACTRN12618001346268. Registered 9 August 2018, https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=374253

Supplementary Information

The online version contains supplementary material available at 10.1186/s40658-022-00438-2.

Locoregional Recurrence Prediction Using a Deep Neural Network of Radiological and Radiotherapy Images

Radiation therapy (RT) is an important and potentially curative modality for head and neck squamous cell carcinoma (HNSCC). Locoregional recurrence (LR) of HNSCC after RT is ranging from 15% to 50% depending on the primary site and stage. In addition, the 5-year survival rate of patients with LR is low. To classify high-risk patients who might develop LR, a deep learning model for predicting LR needs to be established. In this work, 157 patients with HNSCC who underwent RT were analyzed. Based on the National Cancer Institute’s multi-institutional TCIA data set containing FDG-PET/CT/dose, a 3D deep learning model was proposed to predict LR without time-consuming segmentation or feature extraction. Our model achieved an averaged area under the curve (AUC) of 0.856. Adding clinical factors into the model improved the AUC to an average of 0.892 with the highest AUC of up to 0.974. The 3D deep learning model could perform individualized risk quantification of LR in patients with HNSCC without time-consuming tumor segmentation.

Artificial Intelligence in Lymphoma PET Imaging:: A Scoping Review (Current Trends and Future Directions)

Malignant lymphomas are a family of heterogenous disorders caused by clonal proliferation of lymphocytes. 18F-FDG-PET has proven to provide essential information for accurate quantification of disease burden, treatment response evaluation, and prognostication. However, manual delineation of hypermetabolic lesions is often a time-consuming and impractical task. Applications of artificial intelligence (AI) may provide solutions to overcome this challenge. Beyond segmentation and detection of lesions, AI could enhance tumor characterization and heterogeneity quantification, as well as treatment response prediction and recurrence risk stratification. In this scoping review, we have systematically mapped and discussed the current applications of AI (such as detection, classification, segmentation as well as the prediction and prognostication) in lymphoma PET.

Lung Cancer Segmentation With Transfer Learning: Usefulness of a Pretrained Model Constructed From an Artificial Dataset Generated Using a Generative Adversarial Network

Purpose: The purpose of this study was to develop and evaluate lung cancer segmentation with a pretrained model and transfer learning. The pretrained model was constructed from an artificial dataset generated using a generative adversarial network (GAN).

Materials and Methods: Three public datasets containing images of lung nodules/lung cancers were used: LUNA16 dataset, Decathlon lung dataset, and NSCLC radiogenomics. The LUNA16 dataset was used to generate an artificial dataset for lung cancer segmentation with the help of the GAN and 3D graph cut. Pretrained models were then constructed from the artificial dataset. Subsequently, the main segmentation model was constructed from the pretrained models and the Decathlon lung dataset. Finally, the NSCLC radiogenomics dataset was used to evaluate the main segmentation model. The Dice similarity coefficient (DSC) was used as a metric to evaluate the segmentation performance.

Results: The mean DSC for the NSCLC radiogenomics dataset improved overall when using the pretrained models. At maximum, the mean DSC was 0.09 higher with the pretrained model than that without it.

Conclusion: The proposed method comprising an artificial dataset and a pretrained model can improve lung cancer segmentation as confirmed in terms of the DSC metric. Moreover, the construction of the artificial dataset for the segmentation using the GAN and 3D graph cut was found to be feasible.

Artificial intelligence in molecular imaging

AI has, to varying degrees, affected all aspects of molecular imaging, from image acquisition to diagnosis. During the last decade, the advent of deep learning in particular has transformed medical image analysis. Although the majority of recent advances have resulted from neural-network models applied to image segmentation, a broad range of techniques has shown promise for image reconstruction, image synthesis, differential-diagnosis generation, and treatment guidance. Applications of AI for drug design indicate the way forward for using AI to facilitate molecular-probe design, which is still in its early stages. Deep-learning models have demonstrated increased efficiency and image quality for PET reconstruction from sinogram data. Generative adversarial networks (GANs), which are paired neural networks that are jointly trained to generate and classify images, have found applications in modality transformation, artifact reduction, and synthetic-PET-image generation. Some AI applications, based either partly or completely on neural-network approaches, have demonstrated superior differential-diagnosis generation relative to radiologists. However, AI models have a history of brittleness, and physicians and patients may not trust AI applications that cannot explain their reasoning. To date, the majority of molecular-imaging applications of AI have been confined to research projects, and are only beginning to find their ways into routine clinical workflows via commercialization and, in some cases, integration into scanner hardware. Evaluation of actual clinical products will yield more realistic assessments of AI’s utility in molecular imaging.

Interobserver agreement in automated metabolic tumor volume measurements of Deauville score 4 and 5 lesions at interim 18F-FDG PET in DLBCL

Introduction: Metabolic tumor volume (MTV) on interim-PET (I-PET) is a potential prognostic biomarker for diffuse large B-cell lymphoma (DLBCL). Implementation of MTV on I-PET requires consensus which semi-automated segmentation method delineates lesions most successfully with least user interaction. Methods used for baseline PET are not necessarily optimal for I-PET due to lower lesional standardized uptake values (SUV) at I-PET. Therefore, we aimed to evaluate which method provides the best delineation quality of Deauville-score (DS) 4-5 DLBCL lesions on I-PET at best interobserver agreement on delineation quality and, secondly, to assess the effect of lesional SUV_max on delineation quality and performance agreements. Methods: DS4-5 lesions from 45 I-PET scans were delineated using six semi-automated methods i) SUV 2.5, ii) SUV 4.0, iii) adaptive threshold [A50%peak], iv) 41% of maximum SUV [41%max], v) majority vote including voxels detected by ≥2 methods [MV2] and vi) detected by ≥3 methods [MV3]. Delineation quality per MTV was rated by three independent observers as acceptable or non-acceptable. For each method, observer scores on delineation quality, specific agreements and MTV were assessed for all lesions, and per category of lesional SUV_max (<5, 5-10, >10). Results: In 60 DS4-5 lesions on I-PET, MV3 performed best, with acceptable delineation in 90% of lesions, with a positive agreement (PA) of 93%. Delineation quality scores and agreements per method strongly depended on lesional SUV: the best delineation quality scores were obtained using MV3 in lesions with SUV_max<10 and SUV4.0 in more FDG-avid lesions. Consequently, overall delineation quality and PA improved by applying the most preferred method per SUV category instead of using MV3 as single best method. MV3- and SUV4.0-derived MTVs of lesions with SUV_max>10, were comparable after excluding visually failed MV3 contouring. For lesions with SUV_max<10, MTVs using different methods correlated poorly. Conclusion: On I-PET, MV3 performed best and provided the highest interobserver agreement regarding acceptable delineations of DS4-5 DLBCL lesions. However, delineation method preference strongly depended on lesional SUV. Therefore, we suggest to explore an approach that identifies the optimal delineation method per lesion as function of tumor FDG uptake characteristics, i.e. SUV_max.

Plausibility and redundancy analysis to select FDG-PET textural features in non-small cell lung cancer

Background

Radiomics refers to the extraction of a large number of image biomarker describing the tumor phenotype displayed in a medical image. Extracted from positron emission tomography (PET) images, radiomics showed diagnostic and prognostic value for several cancer types. However, a large number of radiomic features are nonreproducible or highly correlated with conventional PET metrics. Moreover, radiomic features used in the clinic should yield relevant information about tumor texture. In this study, we propose a framework to identify technical and clinical meaningful features and exemplify our results using a PET non‐small cell lung cancer (NSCLC) dataset.

Materials and methods

The proposed selection procedure consists of several steps. A priori, we only include features that were found to be reproducible in a multicenter setting. Next, we apply a voxel randomization step to identify features that reflect actual textural information, that is, that yield in 90% of the patient scans a value significantly different from random texture. Finally, the remaining features were correlated with standard PET metrics to further remove redundancy with common PET metrics. The selection procedure was performed for different volume ranges, that is, excluding lesions with smaller volumes in order to assess the effect of tumor size on the results. To exemplify our procedure, the selected features were used to predict 1‐yr survival in a dataset of 150 NSCLC patients. A predictive model was built using volume as predictive factor for smaller, and one of the selected features as predictive factor for bigger lesions. The prediction accuracy of the both models were compared with the prediction accuracy of volume.

Results

The number of selected features depended on the lesion size included in the analysis. When including the whole dataset, from 19 features reflecting actual texture only two were found to be not strongly correlated with conventional PET metrics. When excluding lesions smaller than 11.49 and 33.10 mL (25 and 50 percentile of the dataset), four out of 27 features and 13 out of 29 features remained after eliminating features highly correlated with standard PET metrics. When excluding lesions smaller than 103.9 mL (75 percentile), 33 out of 53 features remained. For larger lesions, some of these features outperformed volume in terms of classification accuracy (increase of 4–10%). The combination of using volume as predictor for smaller and one of the selected features for larger lesions also improved the accuracy when compared with volume only (increase from 72% to 76%).

Conclusion

When performing radiomic analysis for smaller lesions, it should be first carefully investigated if a textural feature reflects actual heterogeneity information. Next, verification of the absence of correlation with all conventional PET metrics is essential in order to assess the additional value of radiomic features. Radiomic analysis with lesions larger than 11.4 mL might give additional information to conventional metrics while at the same time reflecting actual tumor texture. Using a combination of volume and one of the selected features for prediction yields promise to increase accuracy and reliability of a radiomic model.

Intensity threshold based solid tumour segmentation method for Positron Emission Tomography (PET) images: A review

Accurate, robust and reproducible delineation of tumour in Positron Emission Tomography (PET) is essential for diagnosis, treatment planning and response assessment. Since standardized uptake value (SUV) – a normalized semiquantitative parameter used in PET is represented by the intensity of the PET images and related to the radiotracer uptake, a SUV based threshold method is a natural choice to delineate the tumour. However, determination of an optimum threshold value is a challenging task due to low spatial resolution, and signal-to-noise ratio (SNR) along with finite image sampling constraint. The aim of the review is to summarize different fixed and adaptive threshold-based PET image segmentation approaches under a common mathematical framework Advantages and disadvantages of different threshold based methods are also highlighted from the perspectives of diagnosis, treatment planning and response assessment. Several fixed threshold values (30%–70% of the maximum SUV of the tumour (SUV_maxT)) have been investigated. It has been reported that the fixed threshold-based method is very much dependent on the SNR, tumour to background ratio (TBR) and the size of the tumour. Adaptive threshold-based method, an alternative to fixed threshold, can minimize these dependencies by accounting for tumour to background ratio (TBR) and tumour size. However, the parameters for the adaptive methods need to be calibrated for each PET camera system (e.g., scanner geometry, image acquisition protocol, reconstruction algorithm etc.) and it is not straight forward to implement the same procedure to other PET systems to obtain similar results. It has been reported that the performance of the adaptive methods is also not optimum for smaller volumes with lower TBR and SNR. Statistical analysis carried out on the NEMA thorax phantom images also indicates that regions segmented by the fixed threshold method are significantly different for all cases. On the other hand, the adaptive method provides significantly different segmented regions only for low TBR with different SNR. From this viewpoint, a robust threshold based segmentation method that will be less sensitive to SUVmaxT, SNR, TBR and volume needs to be developed. It was really challenging to compare the performance of different threshold-based methods because the performance of each method was tested on dissimilar data set with different data acquisition and reconstruction protocols along with different TBR, SNR and volumes. To avoid such difficulties, it will be desirable to have a common database of clinical PET images acquired with different image acquisition protocols and different PET cameras to compare the performance of automatic segmentation methods. It is also suggested to report the changes in SNR and TBR while reporting the response using threshold based methods.

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

PURPOSE

Lymphoma lesion detection and segmentation on whole-body FDG-PET/CT are a challenging task because of the diversity of involved nodes, organs or physiological uptakes. We sought to investigate the performances of a three-dimensional (3D) convolutional neural network (CNN) to automatically segment total metabolic tumour volume (TMTV) in large datasets of patients with diffuse large B cell lymphoma (DLBCL).

METHODS

The dataset contained pre-therapy FDG-PET/CT from 733 DLBCL patients of 2 prospective LYmphoma Study Association (LYSA) trials. The first cohort (n = 639) was used for training using a 5-fold cross validation scheme. The second cohort (n = 94) was used for external validation of TMTV predictions. Ground truth masks were manually obtained after a 41% SUVmax adaptive thresholding of lymphoma lesions. A 3D U-net architecture with 2 input channels for PET and CT was trained on patches randomly sampled within PET/CTs with a summed cross entropy and Dice similarity coefficient (DSC) loss. Segmentation performance was assessed by the DSC and Jaccard coefficients. Finally, TMTV predictions were validated on the second independent cohort.

RESULTS

Mean DSC and Jaccard coefficients (± standard deviation) in the validations set were 0.73 ± 0.20 and 0.68 ± 0.21, respectively. An underestimation of mean TMTV by - 12 mL (2.8%) ± 263 was found in the validation sets of the first cohort (P = 0.27). In the second cohort, an underestimation of mean TMTV by - 116 mL (20.8%) ± 425 was statistically significant (P = 0.01).

CONCLUSION

Our CNN is a promising tool for automatic detection and segmentation of lymphoma lesions, despite slight underestimation of TMTV. The fully automatic and open-source features of this CNN will allow to increase both dissemination in routine practice and reproducibility of TMTV assessment in lymphoma patients.