PET-guided delineation of radiation therapy treatment volumes: a survey of image segmentation techniques

The Evolution of Artificial Intelligence in Nuclear Medicine

Nuclear medicine has continuously evolved since its beginnings, constantly improving the diagnosis and treatment of various diseases. The integration of artificial intelligence (AI) is one of the latest revolutionizing chapters, promising significant advancements in diagnosis, prognosis, segmentation, image quality enhancement, and theranostics. Early AI applications in nuclear medicine focused on improving diagnostic accuracy, leveraging machine learning algorithms for disease classification and outcome prediction. Advances in deep learning, including convolutional and more recently transformer-based neural networks, have further enabled more precise diagnosis and image segmentation as well as low-dose imaging, and patient-specific dosimetry for personalized treatment. Generative AI, driven by large language models and diffusion techniques, is now allowing the process, interpretation, and generation of complex medical language and images. Despite these achievements, challenges such as data scarcity, heterogeneity, and ethical concerns remain barriers to clinical translation. Addressing these issues through interdisciplinary collaboration will pave the way for a broader adoption of AI in nuclear medicine, potentially enhancing patient care and optimizing diagnosis and therapeutic outcomes.

Deep Learning-Powered CT-Less Multitracer Organ Segmentation From PET Images: A Solution for Unreliable CT Segmentation in PET/CT Imaging

PURPOSE

The common approach for organ segmentation in hybrid imaging relies on coregistered CT (CTAC) images. This method, however, presents several limitations in real clinical workflows where mismatch between PET and CT images are very common. Moreover, low-dose CTAC images have poor quality, thus challenging the segmentation task. Recent advances in CT-less PET imaging further highlight the necessity for an effective PET organ segmentation pipeline that does not rely on CT images. Therefore, the goal of this study was to develop a CT-less multitracer PET segmentation framework.

PATIENTS AND METHODS

We collected 2062 PET/CT images from multiple scanners. The patients were injected with either 18 F-FDG (1487) or 68 Ga-PSMA (575). PET/CT images with any kind of mismatch between PET and CT images were detected through visual assessment and excluded from our study. Multiple organs were delineated on CT components using previously trained in-house developed nnU-Net models. The segmentation masks were resampled to coregistered PET images and used to train 4 different deep learning models using different images as input, including noncorrected PET (PET-NC) and attenuation and scatter-corrected PET (PET-ASC) for 18 F-FDG (tasks 1 and 2, respectively using 22 organs) and PET-NC and PET-ASC for 68 Ga tracers (tasks 3 and 4, respectively, using 15 organs). The models' performance was evaluated in terms of Dice coefficient, Jaccard index, and segment volume difference.

RESULTS

The average Dice coefficient over all organs was 0.81 ± 0.15, 0.82 ± 0.14, 0.77 ± 0.17, and 0.79 ± 0.16 for tasks 1, 2, 3, and 4, respectively. PET-ASC models outperformed PET-NC models ( P < 0.05) for most of organs. The highest Dice values were achieved for the brain (0.93 to 0.96 in all 4 tasks), whereas the lowest values were achieved for small organs, such as the adrenal glands. The trained models showed robust performance on dynamic noisy images as well.

CONCLUSIONS

Deep learning models allow high-performance multiorgan segmentation for 2 popular PET tracers without the use of CT information. These models may tackle the limitations of using CT segmentation in PET/CT image quantification, kinetic modeling, radiomics analysis, dosimetry, or any other tasks that require organ segmentation masks.

PET radiomics in lung cancer: advances and translational challenges

Radiomics is an emerging field of medical imaging that aims at improving the accuracy of diagnosis, prognosis, treatment planning and monitoring non-invasively through the automated or semi-automated quantitative analysis of high-dimensional image features. Specifically in the field of nuclear medicine, radiomics utilizes imaging methods such as positron emission tomography (PET) and single photon emission computed tomography (SPECT) to evaluate biomarkers related to metabolism, blood flow, cellular activity and some biological pathways. Lung cancer ranks among the leading causes of cancer-related deaths globally, and radiomics analysis has shown great potential in guiding individualized therapy, assessing treatment response, and predicting clinical outcomes. In this review, we summarize the current state-of-the-art radiomics progress in lung cancer, highlighting the potential benefits and existing limitations of this approach. The radiomics workflow was introduced first including image acquisition, segmentation, feature extraction, and model building. Then the published literatures were described about radiomics-based prediction models for lung cancer diagnosis, differentiation, prognosis and efficacy evaluation. Finally, we discuss current challenges and provide insights into future directions and potential opportunities for integrating radiomics into routine clinical practice.

Development and evaluation of two open-source nnU-Net models for automatic segmentation of lung tumors on PET and CT images with and without respiratory motion compensation

OBJECTIVES

In lung cancer, one of the main limitations for the optimal integration of the biological and anatomical information derived from Positron Emission Tomography (PET) and Computed Tomography (CT) is the time and expertise required for the evaluation of the different respiratory phases. In this study, we present two open-source models able to automatically segment lung tumors on PET and CT, with and without motion compensation.

MATERIALS AND METHODS

This study involved time-bin gated (4D) and non-gated (3D) PET/CT images from two prospective lung cancer cohorts (Trials 108237 and 108472) and one retrospective. For model construction, the ground truth (GT) was defined by consensus of two experts, and the nnU-Net with 5-fold cross-validation was applied to 560 4D-images for PET and 100 3D-images for CT. The test sets included 270 4D- images and 19 3D-images for PET and 80 4D-images and 27 3D-images for CT, recruited at 10 different centres.

RESULTS

In the performance evaluation with the multicentre test sets, the Dice Similarity Coefficients (DSC) obtained for our PET model were DSC(4D-PET) = 0.74 ± 0.06, improving 19% relative to the DSC between experts and DSC(3D-PET) = 0.82 ± 0.11. The performance for CT was DSC(4D-CT) = 0.61 ± 0.28 and DSC(3D-CT) = 0.63 ± 0.34, improving 4% and 15% relative to DSC between experts.

CONCLUSIONS

Performance evaluation demonstrated that the automatic segmentation models have the potential to achieve accuracy comparable to manual segmentation and thus hold promise for clinical application. The resulting models can be freely downloaded and employed to support the integration of 3D- or 4D- PET/CT and to facilitate the evaluation of its impact on lung cancer clinical practice.

CLINICAL RELEVANCE STATEMENT

We provide two open-source nnU-Net models for the automatic segmentation of lung tumors on PET/CT to facilitate the optimal integration of biological and anatomical information in clinical practice. The models have superior performance compared to the variability observed in manual segmentations by the different experts for images with and without motion compensation, allowing to take advantage in the clinical practice of the more accurate and robust 4D-quantification.

KEY POINTS

Lung tumor segmentation on PET/CT imaging is limited by respiratory motion and manual delineation is time consuming and suffer from inter- and intra-variability. Our segmentation models had superior performance compared to the manual segmentations by different experts. Automating PET image segmentation allows for easier clinical implementation of biological information.

Challenges and limitations in applying radiomics to PET imaging: Possible opportunities and avenues for research

Radiomics, the high-throughput extraction of quantitative imaging features from medical images, holds immense potential for advancing precision medicine in oncology and beyond. While radiomics applied to positron emission tomography (PET) imaging offers unique insights into tumor biology and treatment response, it is imperative to elucidate the challenges and constraints inherent in this domain to facilitate their translation into clinical practice. This review examines the challenges and limitations of applying radiomics to PET imaging, synthesizing findings from the last five years (2019-2023) and highlights the significance of addressing these challenges to realize the full clinical potential of radiomics in oncology and molecular imaging. A comprehensive search was conducted across multiple electronic databases, including PubMed, Scopus, and Web of Science, using keywords relevant to radiomics issues in PET imaging. Only studies published in peer-reviewed journals were eligible for inclusion in this review. Although many studies have highlighted the potential of radiomics in predicting treatment response, assessing tumor heterogeneity, enabling risk stratification, and personalized therapy selection, various challenges regarding the practical implementation of the proposed models still need to be addressed. This review illustrates the challenges and limitations of radiomics in PET imaging across various cancer types, encompassing both phantom and clinical investigations. The analyzed studies highlight the importance of reproducible segmentation methods, standardized pre-processing and post-processing methodologies, and the need to create large multicenter studies registered in a centralized database to promote the continuous validation and clinical integration of radiomics into PET imaging.

A head-to-head comparison of [68Ga]Ga-DOTATATE and [68Ga]Ga-FAPI PET/CT in patients with nasopharyngeal carcinoma: a single-center, prospective study

PURPOSE

We aimed to compare the staging efficiency of [68Ga]Ga-DOTATATE and [68Ga]Ga-FAPI PET/CT in nasopharyngeal carcinoma (NPC) patients.

METHODS

Thirty-nine patients with pathologically confirmed NPC were enrolled in this prospective study. Each patient underwent paired [68Ga]Ga-DOTATATE and [68Ga]Ga-FAPI PET/CT on 2 successive days. The accuracy of two PET/CT for assessing T, N, and M stages was compared by using head-and-neck MRI, histopathologic diagnosis and follow-up results as reference standards. The radiotracer uptake derived from two PETs was also compared.

RESULTS

For treatment-naïve patients, [68Ga]Ga-DOTATATE PET/CT showed identical sensitivity for the primary tumours but clearer tumor delineation induced by higher tumour-to-background (TBR) ratio (19.1 ± 8.7 vs. 12.4 ± 7.7, P = 0.003), compared with [68Ga]Ga-FAPI PET/CT. Regarding cervical lymph node (CLN) metastases, [68Ga]Ga-DOTATATE PET had significantly better sensitivity and accuracy based on neck sides (98% vs. 82%, P < 0.001; 99% vs. 88% P = 0.008), neck levels (98% vs. 78%, 99% vs. 97%; both P < 0.001) and individual nodes (89% vs. 56%, 91% vs. 76%; both P < 0.001), and higher TBR (8.1 ± 4.1 vs. 6.3 ± 3.7, P < 0.001). Additionally, [68Ga]Ga-DOTATATE PET/CT revealed higher sensitivity and accuracy for distant metastases (96% vs. 53%, 95% vs. 52%; both P < 0.001), particularly in bone metastases (99% vs. 49%, 97% vs. 49%; both P < 0.001). For post-treatment patients, [68Ga]Ga-DOTATATE PET/CT identified one more true-negative case than [68Ga]Ga-FAPI PET/CT.

CONCLUSION

[68Ga]Ga-DOTATATE PET/CT performed better than [68Ga]Ga-FAPI PET/CT in visualizing the primary tumours, detecting the metastatic lesions and identifying the local recurrence, suggesting [68Ga]Ga-DOTATATE PET/CT may be superior to [68Ga]Ga-FAPI PET/CT for NPC staging.

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.

Semi-automatic fine delineation scheme for pancreatic cancer

BACKGROUND

Pancreatic cancer fine delineation in medical images by physicians is a major challenge due to the vast volume of medical images and the variability of patients.

PURPOSE

A semi-automatic fine delineation scheme was designed to assist doctors in accurately and quickly delineating the cancer target region to improve the delineation accuracy of pancreatic cancer in computed tomography (CT) images and effectively reduce the workload of doctors.

METHODS

A target delineation scheme in image blocks was also designed to provide more information for the deep learning delineation model. The start and end slices of the image block were manually delineated by physicians, and the cancer in the middle slices were accurately segmented using a three-dimensional Res U-Net model. Specifically, the input of the network is the CT image of the image block and the delineation of the cancer in the start and end slices, while the output of the network is the cancer area in the middle slices of the image block. Meanwhile, the model performance of pancreatic cancer delineation and the workload of doctors in different image block sizes were studied.

RESULTS

We used 37 3D CT volumes for training, 11 volumes for validating and 11 volumes for testing. The influence of different image block sizes on doctors' workload was compared quantitatively. Experimental results showed that the physician's workload was minimal when the image block size was 5, and all cancer could be accurately delineated. The Dice similarity coefficient was 0.894 ± 0.029, the 95% Hausdorff distance was 3.465 ± 0.710 mm, the normalized surface Dice was 0.969 ± 0.019. By completing the accurate delineation of all the CT images, the speed of the new method is 2.16 times faster than that of manual sketching.

CONCLUSION

Our proposed 3D semi-automatic delineative method based on the idea of block prediction could accurately delineate CT images of pancreatic cancer and effectively deal with the challenges of class imbalance, background distractions, and non-rigid geometrical features. This study had a significant advantage in reducing doctors' workload, and was expected to help doctors improve their work efficiency in clinical application.

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.

Application of a smart 18F-FDG-PET adaptive threshold segmentation algorithm for the biological target volume delineation in head and neck cancer

BACKGROUND

The aim of the present study is to evaluate the reliability of a 18F-fluorodeoxyglucose (18F-FDG) PET adaptive threshold segmentation (ATS) algorithm, previously validated in a preclinical setting on several scanners, for the biological target volume (BTV) delineation of head and neck radiotherapy planning.

METHODS

[18F]FDG PET ATS algorithm was studied in treatment plans of head and neck squamous cell carcinoma on a dedicated workstation (iTaRT, Tecnologie Avanzate, Turin, Italy). BTVs segmented by the present ATS algorithm (BTVATS) were compared with those manually segmented for the original radiotherapy treatment planning (BTVVIS). We performed a qualitative and quantitative volumetric analysis with a comparison tool within the ImSimQA TM software package (Oncology Systems Limited, Shrewsbury, UK). We reported figures of merit (FOMs) to convey complementary information: Dice Similarity Coefficient, Sensitivity Index, and Inclusiveness Index.

RESULTS

The study was conducted on 32 treatment plans. Median BTVATS was 11 cm3 while median BTVVIS was 14 cm3. The median Dice Similarity Coefficient, Sensitivity Index, Inclusiveness Index were 0.72, 63%, 88%, respectively. Interestingly, the median volume and the median distance of the voxels that are over contoured by ATS were respectively 1 cm3 and 1 mm.

CONCLUSIONS

ATS algorithm could be a smart and an independent operator tool when implemented for 18F-FDG-PET-based tumor volume delineation. Furthermore, it might be relevant in case of BTV-based dose painting.

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.

Evaluation of 11 C-Choline PET/CT for T Staging and Tumor Volume Delineation in Nasopharyngeal Cancer Patients in Comparison to 18 F-FDG PET/CT

PURPOSE

Accurate determination of the primary tumor extension of nasopharyngeal carcinoma (NPC) by 18 F-FDG PET/CT is limited by the high physiological 18 F-FDG uptake in the surrounding area, especially in the brain tissue. We aimed to assess whether 11 C-choline PET/CT could improve the accuracy of T staging and tumor volume delineation for NPC patients.

METHODS

Patients with pathologically confirmed diagnosis of NPC were enrolled. The primary tumor extension of each patient was evaluated by 11 C-choline PET/CT, 18 F-FDG PET/CT, and contrast-enhanced MRI. The PET/CT-based tumor volume ( VPET ) was measured by 3 threshold methods, including the threshold of SUV 2.5 (Th 2.5 ), 40% of maximal SUV (Th 40% ), and the relative background-dependent threshold (Th bgd ). Tumor volume and Dice similarity coefficient were compared among VPET with different segmentation methods and VMR .

RESULTS

Thirty-three patients with treatment-naive NPC and 6 patients with suspicious recurrent disease were enrolled. The NPC lesions were avid for both 11 C-choline and 18 F-FDG. Visual analysis showed that 11 C-choline PET/CT had better contrast and higher discernability than 18 F-FDG PET/CT for intracranial, skull base, and orbital involvement. 11 C-choline PET/CT also exhibited advantage over MRI for differentiation between local recurrence and radiation-induced alterations. For the tumor delineated, the VMR was larger than VPET in general, except for 18 F-FDG PET/CT with Th 2.5 threshold. For all 3 threshold methods applied, 11 C-choline PET/CT produced more consistent and comparable tumor volume to MRI than 18 F-FDG PET/CT. 11 C-choline PET/CT with Th bgd threshold showed the closest tumor volume and highest similarity to MRI.

CONCLUSIONS

11 C-choline PET/CT provides a higher accuracy than 18 F-FDG PET/CT in mapping tumor extension in locally advanced NPC and may be a promising complement to MRI in delineating the primary tumor.

Role of 18F-choline and 18F-fluorodeoxyglucose positron emission tomography in combination with magnetic resonance imaging in brachytherapy planning for locally advanced cervical cancer: A pilot study

Background and purpose

This pilot study aims to describe the advantages of combining metabolic and anatomic imaging modalities in brachytherapy (BT) planning for locally advanced cervical cancer (LACC) and to evaluate the supplementary value of Fluoro(F)-Choline positron emission tomography/computed tomography (PET/CT) in comparison to 18F-fluorodeoxyglucose (FDG) in this setting.

Materials and methods

A prospective cohort of six patients with LACC was included in this study. Each patient underwent BT planning CT scan, magnetic resonance imaging (MRI), and both FDG and F-Choline PET/CT scans on the same day, with BT applicators in place. Patients were treated according to the standard of care. Metabolic target volumes (TV) were generated retrospectively and compared with the anatomic volumes using Dice coefficients and absolute volume comparison.

Results

The threshold at which the metabolic and anatomic volumes were the most concordant was found to be 35% maximum standardized uptake value (SUV max) for both PET/CT scans. Amongst the six patients in this cohort, three in the FDG cohort and four in the F-Choline cohort were found to have more than ten percent ratio of excess (increase) in their MRI gross tumor volumes (GTV) when incorporating the metabolic information from the PET/CT scans. However, no significant changes were needed in the high risk-clinical target volumes (CTVHR) for both PET tracers.

Conclusions

FDG and F-Choline PET/CT scans can substantially modify the BT GTV on MRI, without affecting the CTVHR. F-Choline is potentially more informative than FDG in assessing residual TV, particularly in cases with significant post-radiation inflammatory changes.

Automated detection, delineation and quantification of whole-body bone metastasis using FDG-PET/CT images

Non-small cell lung cancer (NSCLC) patients with the metastatic spread of disease to the bone have high morbidity and mortality. Stereotactic ablative body radiotherapy increases the progression free survival and overall survival of these patients with oligometastases. FDG-PET/CT, a functional imaging technique combining positron emission tomography (PET) with 18 F-fluorodeoxyglucose (FDG) and computer tomography (CT) provides improved staging and identification of treatment response. It is also associated with reduction in size of the radiotherapy tumour volume delineation compared with CT based contouring in radiotherapy, thus allowing for dose escalation to the target volume with lower doses to the surrounding organs at risk. FDG-PET/CT is increasingly being used for the clinical management of NSCLC patients undergoing radiotherapy and has shown high sensitivity and specificity for the detection of bone metastases in these patients. Here, we present a software tool for detection, delineation and quantification of bone metastases using FDG-PET/CT images. The tool extracts standardised uptake values (SUV) from FDG-PET images for auto-segmentation of bone lesions and calculates volume of each lesion and associated mean and maximum SUV. The tool also allows automatic statistical validation of the auto-segmented bone lesions against the manual contours of a radiation oncologist. A retrospective review of FDG-PET/CT scans of more than 30 candidate NSCLC patients was performed and nine patients with one or more metastatic bone lesions were selected for the present study. The SUV threshold prediction model was designed by splitting the cohort of patients into a subset of 'development' and 'validation' cohorts. The development cohort yielded an optimum SUV threshold of 3.0 for automatic detection of bone metastases using FDG-PET/CT images. The validity of the derived optimum SUV threshold on the validation cohort demonstrated that auto-segmented and manually contoured bone lesions showed strong concordance for volume of bone lesion (r = 0.993) and number of detected lesions (r = 0.996). The tool has various applications in radiotherapy, including but not limited to studies determining optimum SUV threshold for accurate and standardised delineation of bone lesions and in scientific studies utilising large patient populations for instance for investigation of the number of metastatic lesions that can be treated safety with an ablative dose of radiotherapy without exceeding the normal tissue toxicity.

Mapping early tumor response to radiotherapy using diffusion kurtosis imaging*

PURPOSE

In this pilot study, DKI measures of diffusivity and kurtosis were compared in active tumor regions and correlated to radiologic response to radiotherapy after completion of 2 weeks of treatment to derive potential early measures of tumor response.

METHODS

MRI and Magnetic Resonance Spectroscopic Imaging (MRSI) data were acquired before the beginning of RT (pre-RT) and 2 weeks after the initiation of treatment (during-RT) in 14 glioblastoma patients. The active tumor region was outlined as the union of the residual contrast-enhancing region and metabolically active tumor region. Average and standard deviation of mean, axial, and radial diffusivity (MD, AD, RD) and mean, axial, and radial kurtosis (MK, AK, RK) values were calculated for the active tumor VOI from images acquired pre-RT and during-RT and paired t-tests were executed to estimate pairwise differences. Receiver operating characteristic (ROC) curve analysis was conducted to evaluate the predictive capabilities of changes in diffusion metrics for progression-free survival (PFS).

RESULTS

Analysis showed significant pairwise differences for AD (p = 0.035; Cohen's d of 0.659) and AK (p = 0.019; Cohen's d of 0.753) in diffusion measures after 2 weeks of RT. ROC curve analysis showed that percentage change differences in AD and AK between pre-RT and during-RT scans provided an Area Under the Curve (AUC) of 0.524 and 0.762, respectively, in discriminating responders (PFS>180 days) and non-responders (PFS<180 days).

CONCLUSION

This pilot study, although preliminary in nature, showed significant changes in AD and AK maps, with kurtosis derived AK maps showing an increased sensitivity in mapping early changes in the active tumor regions.

Does consensus contours improve robustness and accuracy on 18 F-FDG PET imaging tumor delineation?

PURPOSE

The aim of this study is to explore the robustness and accuracy of consensus contours with 225 nasopharyngeal carcinoma (NPC) clinical cases and 13 extended cardio-torso simulated lung tumors (XCAT) based on 2-deoxy-2-[18 F]fluoro-D-glucose (18 F-FDG) PET imaging.

METHODS

Primary tumor segmentation was performed with two different initial masks on 225 NPC18 F-FDG PET datasets and 13 XCAT simulations using methods of automatic segmentation with active contour, affinity propagation (AP), contrast-oriented thresholding (ST), and 41% maximum tumor value (41MAX), respectively. Consensus contours (ConSeg) were subsequently generated based on the majority vote rule. The metabolically active tumor volume (MATV), relative volume error (RE), Dice similarity coefficient (DSC) and their respective test-retest (TRT) metrics between different masks were adopted to analyze the results quantitatively. The nonparametric Friedman and post hoc Wilcoxon tests with Bonferroni adjustment for multiple comparisons were performed withP < 0.05 considered to be significant.

RESULTS

AP presented the highest variability for MATV in different masks, and ConSeg presented much better TRT performances in MATV compared with AP, and slightly poorer TRT in MATV compared with ST or 41MAXin most cases. Similar trends were also found in RE and DSC with the simulated data. The average of four segmentation results (AveSeg) showed better or comparable results in accuracy for most cases with respect to ConSeg. AP, AveSeg and ConSeg presented better RE and DSC in irregular masks as compared with rectangle masks. Additionally, all methods underestimated the tumour boundaries in relation to the ground truth for XCAT including respiratory motion.

CONCLUSIONS

The consensus method could be a robust approach to alleviate segmentation variabilities, but did not seem to improve the accuracy of segmentation results on average. Irregular initial masks might be at least in some cases attributable to mitigate the segmentation variability as well.

CGBO-Net: Cruciform structure guided and boundary-optimized lymphoma segmentation network

Lymphoma segmentation plays an important role in the diagnosis and treatment of lymphocytic tumor. Most current existing automatic segmentation methods are difficult to give precise tumor boundary and location. Semi-automatic methods are usually combined with manually added features such as bounding box or points to locate the tumor. Inspired by this, we propose a cruciform structure guided and boundary-optimized lymphoma segmentation network(CGBS-Net). The method uses a cruciform structure extracted based on PET images as an additional input to the network, while using a boundary gradient loss function to optimize the boundary of the tumor. Our method is divided into two main stages: In the first stage, we use the proposed axial context-based cruciform structure extraction (CCE) method to extract the cruciform structures of all tumor slices. In the second stage, we use PET/CT and the corresponding cruciform structure as input in the designed network (CGBO-Net) to extract tumor structure and boundary information. The Dice, Precision, Recall, IOU and RVD are 90.7%, 89.4%, 92.5%, 83.1% and 4.5%, respectively. Validate on the lymphoma dataset and publicly available head and neck data, our proposed approach is better than the other state-of-the-art semi-segmentation methods, which produces promising segmentation results.

Beyond Imaging and Genetic Signature in Glioblastoma: Radiogenomic Holistic Approach in Neuro-Oncology

Glioblastoma (GBM) is a malignant brain tumor exhibiting rapid and infiltrative growth, with less than 10% of patients surviving over 5 years, despite aggressive and multimodal treatments. The poor prognosis and the lack of effective pharmacological treatments are imputable to a remarkable histological and molecular heterogeneity of GBM, which has led, to date, to the failure of precision oncology and targeted therapies. Identification of molecular biomarkers is a paradigm for comprehensive and tailored treatments; nevertheless, biopsy sampling has proved to be invasive and limited. Radiogenomics is an emerging translational field of research aiming to study the correlation between radiographic signature and underlying gene expression. Although a research field still under development, not yet incorporated into routine clinical practice, it promises to be a useful non-invasive tool for future personalized/adaptive neuro-oncology. This review provides an up-to-date summary of the recent advancements in the use of magnetic resonance imaging (MRI) radiogenomics for the assessment of molecular markers of interest in GBM regarding prognosis and response to treatments, for monitoring recurrence, also providing insights into the potential efficacy of such an approach for survival prognostication. Despite a high sensitivity and specificity in almost all studies, accuracy, reproducibility and clinical value of radiomic features are the Achilles heel of this newborn tool. Looking into the future, investigators' efforts should be directed towards standardization and a disciplined approach to data collection, algorithms, and statistical analysis.

Comparison of protocols with respiratory-gated (4D) motion compensation in PET/CT: open-source package for quantification of phantom image quality

BACKGROUND

Patient's breathing affects the quality of chest images acquired with positron emission tomography/computed tomography (PET/CT) studies. Movement correction is required to optimize PET quantification in clinical settings. We present a reproducible methodology to compare the impact of different movement compensation protocols on PET image quality. Static phantom images were set as reference values, and recovery coefficients (RCs) were calculated from motion compensated images for the phantoms in respiratory movement. Image quality was evaluated in terms of: (1) volume accuracy (VA) with the NEMA phantom; (2) concentration accuracy (CA) by six refillable inserts within the electron density CIRS phantom; and (3) spatial resolution (R) with the Jaszczak phantom. Three different respiratory patterns were applied to the phantoms. We developed an open-source package to automatically analyze VA, CA and R. We compared 10 different movement compensation protocols available in the Philips Gemini TF-64 PET/CT (4-, 6-, 8- and 10-time bins, 20%-, 30%-, 40%-window width in Inhale and Exhale).

RESULTS

The homemade package provided RC values for VA, CA and R of 102 PET images in less than 5 min. Results of the comparison of the 10 different protocols demonstrated the feasibility of the proposed method for quantifying the variations observed qualitatively. Overall, prospective protocols showed better motion compensation than retrospective. The best performance was obtained for the protocol Exhale 30% (0.3 s after maximum Exhale position and window width of 30%) with RC[Formula: see text], RC[Formula: see text] and RC[Formula: see text]. Among retrospective protocols, 8 Phase protocol showed the best performance.

CONCLUSION

We provided an open-source package able to automatically evaluate the impact of motion compensation methods on PET image quality. A setup based on commonly available experimental phantoms is recommended. Its application for the comparison of 10 time-based approaches showed that Exhale 30% protocol had the best performance. The proposed framework is not specific to the phantoms and protocols presented on this study.

Isolation and characterization of phosphate solubilizing bacteria naturally colonizing legumes rhizosphere in Morocco

Low-cost and environmentally friendly agricultural practices have received increasing attention in recent years. Developing microbial inoculants containing phosphate (P) solubilizing bacteria (PSB) represents an emerging biological solution to improve rhizosphere P availability. The present study aims to explore PSB strains isolated from soils located at different bioclimatic stages in Morocco and present in various legumes rhizosphere to improve agronomic microbial fertilizer’s effectiveness. It was also aimed to test the isolated strains for their ability to solubilize P in NBRIP medium with Tricalcium P (Ca₃ (PO₄)₂) (TCP), rock phosphate (RP), and their combination as a source of phosphorus, by (2²) experiment design. Bacterial strains with a high P solubility index (PSI) were selected, characterized, and compared to commercial control. The vanadate-molybdate method was used to estimate P solubilization activity. Stress tolerance to salinity, acidity, drought, and temperature was tested. From all isolated strains (64), 12 were screened as promising biotechnological interest because of their P solubilization and their good resistance to different drastic conditions. Besides, the strain WJEF15 showed the most P solubility efficiency in NBRIP solid medium with a PSI of 4.1; while the WJEF61 strain was located as the most efficient strain in NBRIP-TCP liquid medium by releasing 147.62 mg.l^–1 of soluble P. In contrast, in the NBRIP-RP medium, the strain WJEF15 presented maximum solubilization with 25.16 mg.l^–1. The experiment design showed that a combination of RP and TCP with max level progressively increases P solubilization by 20.58%, while the WJEF63 strain has the most efficient concentration of 102.69 mg.l^–1. Indeed, among the selected strains, four strains were able to limit tested fungi growth. Thus, results reveal a potential effect of selecting PSBs to support cropping cultures as plant growth-promoting rhizobacteria (PGPR).

Joint EANM/SNMMI/ESTRO practice recommendations for the use of 2-[18F]FDG PET/CT external beam radiation treatment planning in lung cancer V1.0

PURPOSE

2-[18F]FDG PET/CT is of utmost importance for radiation treatment (RT) planning and response monitoring in lung cancer patients, in both non-small and small cell lung cancer (NSCLC and SCLC). This topic has been addressed in guidelines composed by experts within the field of radiation oncology. However, up to present, there is no procedural guideline on this subject, with involvement of the nuclear medicine societies.

METHODS

A literature review was performed, followed by a discussion between a multidisciplinary team of experts in the different fields involved in the RT planning of lung cancer, in order to guide clinical management. The project was led by experts of the two nuclear medicine societies (EANM and SNMMI) and radiation oncology (ESTRO).

RESULTS AND CONCLUSION

This guideline results from a joint and dynamic collaboration between the relevant disciplines for this topic. It provides a worldwide, state of the art, and multidisciplinary guide to 2-[18F]FDG PET/CT RT planning in NSCLC and SCLC. These practical recommendations describe applicable updates for existing clinical practices, highlight potential flaws, and provide solutions to overcome these as well. Finally, the recent developments considered for future application are also reviewed.

Artificial Intelligence in Radiation Therapy

Artificial intelligence (AI) has great potential to transform the clinical workflow of radiotherapy. Since the introduction of deep neural networks, many AI-based methods have been proposed to address challenges in different aspects of radiotherapy. Commercial vendors have started to release AI-based tools that can be readily integrated to the established clinical workflow. To show the recent progress in AI-aided radiotherapy, we have reviewed AI-based studies in five major aspects of radiotherapy including image reconstruction, image registration, image segmentation, image synthesis, and automatic treatment planning. In each section, we summarized and categorized the recently published methods, followed by a discussion of the challenges, concerns, and future development. Given the rapid development of AI-aided radiotherapy, the efficiency and effectiveness of radiotherapy in the future could be substantially improved through intelligent automation of various aspects of radiotherapy.

Radiomics for precision medicine in glioblastoma

INTRODUCTION

Being the most common primary brain tumor, glioblastoma presents as an extremely challenging malignancy to treat with dismal outcomes despite treatment. Varying molecular epidemiology of glioblastoma between patients and intra-tumoral heterogeneity explains the failure of current one-size-fits-all treatment modalities. Radiomics uses machine learning to identify salient features of the tumor on brain imaging and promises patient-specific management in glioblastoma patients.

METHODS

We performed a comprehensive review of the available literature on studies investigating the role of radiomics and radiogenomics models for the diagnosis, stratification, prognostication as well as treatment planning and monitoring of glioblastoma.

RESULTS

Classifiers based on a combination of various MRI sequences, genetic information and clinical data can predict non-invasive tumor diagnosis, overall survival and treatment response with reasonable accuracy. However, the use of radiomics for glioblastoma treatment remains in infancy as larger sample sizes, standardized image acquisition and data extraction techniques are needed to develop machine learning models that can be translated effectively into clinical practice.

CONCLUSION

Radiomics has the potential to transform the scope of glioblastoma management through personalized medicine.

Fully Automated Gross Tumor Volume Delineation From PET in Head and Neck Cancer Using Deep Learning Algorithms

PURPOSE

The availability of automated, accurate, and robust gross tumor volume (GTV) segmentation algorithms is critical for the management of head and neck cancer (HNC) patients. In this work, we evaluated 3 state-of-the-art deep learning algorithms combined with 8 different loss functions for PET image segmentation using a comprehensive training set and evaluated its performance on an external validation set of HNC patients.

PATIENTS AND METHODS

18F-FDG PET/CT images of 470 patients presenting with HNC on which manually defined GTVs serving as standard of reference were used for training (340 patients), evaluation (30 patients), and testing (100 patients from different centers) of these algorithms. PET image intensity was converted to SUVs and normalized in the range (0-1) using the SUVmax of the whole data set. PET images were cropped to 12 × 12 × 12 cm3 subvolumes using isotropic voxel spacing of 3 × 3 × 3 mm3 containing the whole tumor and neighboring background including lymph nodes. We used different approaches for data augmentation, including rotation (-15 degrees, +15 degrees), scaling (-20%, 20%), random flipping (3 axes), and elastic deformation (sigma = 1 and proportion to deform = 0.7) to increase the number of training sets. Three state-of-the-art networks, including Dense-VNet, NN-UNet, and Res-Net, with 8 different loss functions, including Dice, generalized Wasserstein Dice loss, Dice plus XEnt loss, generalized Dice loss, cross-entropy, sensitivity-specificity, and Tversky, were used. Overall, 28 different networks were built. Standard image segmentation metrics, including Dice similarity, image-derived PET metrics, first-order, and shape radiomic features, were used for performance assessment of these algorithms.

RESULTS

The best results in terms of Dice coefficient (mean ± SD) were achieved by cross-entropy for Res-Net (0.86 ± 0.05; 95% confidence interval [CI], 0.85-0.87), Dense-VNet (0.85 ± 0.058; 95% CI, 0.84-0.86), and Dice plus XEnt for NN-UNet (0.87 ± 0.05; 95% CI, 0.86-0.88). The difference between the 3 networks was not statistically significant (P > 0.05). The percent relative error (RE%) of SUVmax quantification was less than 5% in networks with a Dice coefficient more than 0.84, whereas a lower RE% (0.41%) was achieved by Res-Net with cross-entropy loss. For maximum 3-dimensional diameter and sphericity shape features, all networks achieved a RE ≤ 5% and ≤10%, respectively, reflecting a small variability.

CONCLUSIONS

Deep learning algorithms exhibited promising performance for automated GTV delineation on HNC PET images. Different loss functions performed competitively when using different networks and cross-entropy for Res-Net, Dense-VNet, and Dice plus XEnt for NN-UNet emerged as reliable networks for GTV delineation. Caution should be exercised for clinical deployment owing to the occurrence of outliers in deep learning-based algorithms.

Artificial intelligence in radiation oncology: A review of its current status and potential application for the radiotherapy workforce

OBJECTIVE

Radiation oncology is a continually evolving speciality. With the development of new imaging modalities and advanced imaging processing techniques, there is an increasing amount of data available to practitioners. In this narrative review, Artificial Intelligence (AI) is used as a reference to machine learning, and its potential, along with current problems in the field of radiation oncology, are considered from a technical position.

KEY FINDINGS

AI has the potential to harness the availability of data for improving patient outcomes, reducing toxicity, and easing clinical burdens. However, problems including the requirement of complexity of data, undefined core outcomes and limited generalisability are apparent.

CONCLUSION

This original review highlights considerations for the radiotherapy workforce, particularly therapeutic radiographers, as there will be an increasing requirement for their familiarity with AI due to their unique position as the interface between imaging technology and patients.

IMPLICATIONS FOR PRACTICE

Collaboration between AI experts and the radiotherapy workforce are required to overcome current issues before clinical adoption. The development of educational resources and standardised reporting of AI studies may help facilitate this.

Optimisation of CT scan parameters to increase the accuracy of gross tumour volume identification in brain radiotherapy

Aim:

This study aimed to optimise computed tomography (CT) simulation scan parameters to increase the accuracy for gross tumour volume identification in brain radiotherapy. For this purpose, high-contrast scan protocols were assessed.

Materials and methods:

A CT accreditation phantom (ACR Gammex 464) was used to optimise brain CT scan parameters on a Toshiba Alexion 16-row multislice CT scanner. Dose, tube voltage, tube current–time and CT dose index (CTDI) were varied to create five image quality enhancement (IQE) protocols. They were assessed in terms of contrast-to-noise ratio (CNR), signal-to-noise ratio (SNR) and noise level and compared with a standard clinical protocol. Finally, the ability of the selected protocols to identify low-contrast objects was examined based on a subjective method.

Results:

Among the five IQE protocols, the one with the highest tube current–time product (250 mA) and lowest tube voltage (100 kVp) showed higher CNR, while another with a tube current–time product of 150 mA and a tube voltage of 135 kVp had improved SNR and lower noise level compared to the standard protocol. In contouring low-contrast objects, the protocol with the highest milliampere and lowest peak kilovoltage exhibited the lowest error rate (1%) compared to the standard protocol (25%).

Findings:

CT image quality should be optimised using the high-dose parameters created in this study to provide better soft tissue contrast. This could lead to an accurate identification of gross tumour volume recognition in the planning of radiotherapy treatment.

Diffuse large B-cell lymphoma segmentation in PET-CT images via hybrid learning for feature fusion

PURPOSE

Diffuse large B-cell lymphoma (DLBCL) is an aggressive type of lymphoma with high mortality and poor prognosis that especially has a high incidence in Asia. Accurate segmentation of DLBCL lesions is crucial for clinical radiation therapy. However, manual delineation of DLBCL lesions is tedious and time-consuming. Automatic segmentation provides an alternative solution but is difficult for diffuse lesions without the sufficient utilization of multimodality information. Our work is the first study focusing on positron emission tomography and computed tomography (PET-CT) feature fusion for the DLBCL segmentation issue. We aim to improve the fusion performance of complementary information contained in PET-CT imaging with a hybrid learning module in the supervised convolutional neural network.

METHODS

First, two encoder branches extract single-modality features, respectively. Next, the hybrid learning component utilizes them to generate spatial fusion maps which can quantify the contribution of complementary information. Such feature fusion maps are then concatenated with specific-modality (i.e., PET and CT) feature maps to obtain a representation of the final fused feature maps in different scales. Finally, the reconstruction part of our network creates a prediction map of DLBCL lesions by integrating and up-sampling the final fused feature maps from encoder blocks in different scales.

RESULTS

The ability of our method was evaluated to detect foreground and segment lesions in three independent body regions (nasopharynx, chest, and abdomen) of a set of 45 PET-CT scans. Extensive ablation experiments compared our method to four baseline techniques for multimodality fusion (input-level (IL) fusion, multichannel (MC) strategy, multibranch (MB) strategy, and quantitative weighting (QW) fusion). The results showed that our method achieved a high detection accuracy (99.63% in the nasopharynx, 99.51% in the chest, and 99.21% in the abdomen) and had the superiority in segmentation performance with the mean dice similarity coefficient (DSC) of 73.03% and the modified Hausdorff distance (MHD) of 4.39 mm, when compared with the baselines (DSC: IL: 53.08%, MC: 63.59%, MB: 69.98%, and QW: 72.19%; MHD: IL: 12.16 mm, MC: 6.46 mm, MB: 4.83 mm, and QW: 4.89 mm).

CONCLUSIONS

A promising segmentation method has been proposed for the challenging DLBCL lesions in PET-CT images, which improves the understanding of complementary information by feature fusion and may guide clinical radiotherapy. The statistically significant analysis based on P-value calculation has indicated a degree of significant difference between our proposed method and other baselines (almost metrics: P < 0.05). This is a preliminary research using a small sample size, and we will collect data continually to achieve the larger verification study.

Total Lesion Glycolysis Estimated by a Radiomics Model From CT Image Alone

Purpose

In this study, total lesion glycolysis (TLG) on positron emission tomography images was estimated by a trained and validated CT radiomics model, and its prognostic ability was explored among lung cancer (LC) and esophageal cancer patients (EC).

Methods

Using the identical features between the combined and thin-section CT, the estimation model of SUVsum (summed standard uptake value) was trained from the lymph nodes (LNs) of LC patients (n = 1239). Besides LNs of LC patients from other centers, the validation cohorts also included LNs and primary tumors of LC/EC from the same center. After calculating TLG (accumulated SUVsum of each individual) based on the model, the prognostic ability of the estimated and measured values was compared and analyzed.

Results

In the training cohort, the model of 3 features was trained by the deep learning and linear regression method. It performed well in all validation cohorts (n = 5), and a linear regression could correct the bias from different scanners. Additionally, the absolute biases of the model were not significantly affected by the evaluated factors whether they included LN metastasis or not. Between the estimated natural logarithm of TLG (elnTLG) and the measured values (mlnTLG), significant difference existed among both LC (n = 137, bias = 0.510 ± 0.519, r = 0.956, P<0.001) and EC patients (n = 56, bias = 0.251± 0.463, r = 0.934, P<0.001). However, for both cancers, the overall shapes of the curves of hazard ratio (HR) against elnTLG or mlnTLG were quite alike.

Conclusion

Total lesion glycolysis can be estimated by three CT features with particular coefficients for different scanners, and it similar to the measured values in predicting the outcome of cancer patients.

Diagnostic efficiency of hybrid imaging using PSMA ligands, PET/CT, PET/MRI and MRI in identifying malignant prostate lesions

OBJECTIVE

The objective of this study was to assess the accuracy of 68Ga-PSMA-11 PET/MRI, 18F-PSMA-1007 PET/CT, 68Ga-PSMA-11 PET/CT, and multiparametric (mp)MRI for the delineating of dominant intraprostatic lesions (IPL).

MATERIALS AND METHODS

35 patients with organ-confined prostate cancer who were assigned to definitive radiotherapy (RT) were divided into three groups based on imaging techniques: 68Ga-PSMA-PET/MRI (n = 9), 18F-PSMA-PET/CT (n = 16) and 68Ga-PSMA-PET/CT (n = 10). All patients without PSMA-PET/MRI received an additional mpMRI. PSMA-PET-based automatic isocontours and manual contours of the dominant IPLs were generated for each modality. The biopsy results were then used to validate whether any of the prostate biopsies were positive in the marked lesion using Dice similarity coefficient (DSC), Youden index (YI), sensitivity and specificity. Factors that can predict the accuracy of IPLs contouring were analysed.

RESULTS

Diagnostic performance was significantly superior both for manual and automatic IPLs contouring using 68Ga-PSMA-PET/MRI (DSC/YI SUV70%-0.62/0.51), 18F-PSMA-PET/CT (DSC/YI SUV70%-0.67/0.53) or 68Ga-PSMA-PET/CT (DSC/YI SUV70%-0.63/0.51) compared to mpMRI (DSC/YI-0.47/0.41; p < 0.001). The accuracy for delineating IPLs was not improved by combination of PET/CT and mpMRI images compared to PET/CT alone. Significantly superior diagnostic accuracy was found for large prostate lesions (at least 15% from the prostate volume) and higher Gleason score (at least 7b) comparing to smaller lesions with lower GS.

CONCLUSION

IPL localization was significantly improved when using PSMA-imaging procedures compared to mpMRI. No significant difference for delineating IPLs was found between hybrid method PSMA-PET/MRI and PSMA-PET/CT. PSMA-based imaging technique should be considered for the diagnostics of IPLs and focal treatment modality.

Automated brain structures segmentation from PET/CT images based on landmark-constrained dual-modality atlas registration

Automated brain structures segmentation in positron emission tomography (PET) images has been widely investigated to help brain disease diagnosis and follow-up. To relieve the burden of a manual definition of volume of interest (VOI), automated atlas-based VOI definition algorithms were developed, but these algorithms mostly adopted a global optimization strategy which may not be particularly accurate for local small structures (especially the deep brain structures). This paper presents a PET/CT-based brain VOI segmentation algorithm combining anatomical atlas, local landmarks, and dual-modality information. The method incorporates local deep brain landmarks detected by the Deep Q-Network (DQN) to constrain the atlas registration process. Dual-modality PET/CT image information is also combined to improve the registration accuracy of the extracerebral contour. We compare our algorithm with the representative brain atlas registration methods based on 86 clinical PET/CT images. The proposed algorithm obtained accurate delineation of brain VOIs with an average Dice similarity score of 0.79, an average surface distance of 0.97 mm (sub-pixel level), and a volume recovery coefficient close to 1. The main advantage of our method is that it optimizes both global-scale brain matching and local-scale small structure alignment around the key landmarks, it is fully automated and produces high-quality parcellation of the brain structures from brain PET/CT images.

Motion-compensated FDG PET/CT for oesophageal cancer

PURPOSE

Respiratory-induced motion of oesophageal tumours and lymph nodes can influence positron-emission tomography/computed tomography (PET/CT). The aim was to compare standard three-dimensional (3D) and motion-compensated PET/CT regarding standardized uptake value (SUV), metabolic tumour volume (MTV) and detection of lymph node metastases.

METHODS

This prospective observational study (NCT02424864) included 37 newly diagnosed oesophageal cancer patients. Diagnostic PET/CT was reconstructed in 3D and motion-compensated PET/CT. MTVs of the primary tumour were calculated using an automated region-growing algorithm with SUV thresholds of 2.5 (MTV2.5) and ≥ 50% of SUVmax (MTV50%). Blinded for reconstruction method, a nuclear medicine physician assessed all lymph nodes showing ¹⁸F‑fluorodeoxyglucose uptake for their degree of suspicion.

RESULTS

The mean (95% CI) SUVmax of the primary tumour was 13.1 (10.6-15.5) versus 13.0 (10.4-15.6) for 3D and motion-compensated PET/CT, respectively. MTVs were also similar between the two techniques. Bland-Altman analysis showed mean differences between both measurements (95% limits of agreement) of 0.08 (-3.60-3.75), -0.26 (-2.34-1.82), 4.66 (-29.61-38.92) cm³ and -0.95 (-19.9-18.0) cm³ for tumour SUVmax, lymph node SUVmax, MTV2.5 and MTV50%, respectively. Lymph nodes were classified as highly suspicious (30/34 nodes), suspicious (20/22) and dubious (66/59) for metastases on 3D/motion-compensated PET/CT. No additional lymph node metastases were found on motion-compensated PET/CT. SUVmax of the most intense lymph nodes was similar for both scans: mean (95% CI) 6.6 (4.3-8.8) and 6.8 (4.5-9.1) for 3D and motion-compensated, respectively.

CONCLUSION

SUVmax of the primary oesophageal tumour and lymph nodes was comparable on 3D and motion-compensated PET/CT. The use of motion-compensated PET/CT did not improve lymph node detection.

Comparison of the Gross Target Volumes Based on Diagnostic PET/CT for Primary Esophageal Cancer

Background

Clinically, many esophageal cancer patients who planned for radiation therapy have already undergone diagnostic Positron-emission tomography/computed tomography (PET/CT) imaging, but it remains unclear whether these imaging results can be used to delineate the gross target volume (GTV) of the primary tumor for thoracic esophageal cancer (EC).

Methods

Seventy-two patients diagnosed with thoracic EC had undergone prior PET/CT for diagnosis and three-dimensional CT (3DCT) for simulation. The GTV_3D was contoured on the 3DCT image without referencing the PET/CT image. The GTV_PET-ref was contoured on the 3DCT image referencing the PET/CT image. The GTV_PET-reg was contoured on the deformed registration image derived from 3DCT and PET/CT. Differences in the position, volume, length, conformity index (CI), and degree of inclusion (DI) among the target volumes were determined.

Results

The centroid distance in the three directions between two different GTVs showed no significant difference (P > 0.05). No significant difference was found among the groups in the tumor volume (P > 0.05). The median DI values of the GTV_PET-reg and GTV_PET-ref in the GTV_3D were 0.82 and 0.86, respectively (P = 0.006). The median CI values of the GTV_3D in the GTV_PET-reg and GTV_PET-ref were 0.68 and 0.72, respectively (P = 0.006).

Conclusions

PET/CT can be used to optimize the definition of the target volume in EC. However, no significant difference was found between the GTVs delineated based on visual referencing or deformable registration whether using the volume or position. So, in the absence of planning PET–CT images, it is also feasible to delineate the GTV of primary thoracic EC with reference to the diagnostic PET–CT image.

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.

Imaging for Target Delineation in Head and Neck Cancer Radiotherapy

The definition of tumor involved volumes in patients with head and neck cancer poses great challenges with the increasing use of highly conformal radiotherapy techniques eg, volumetric modulated arc therapy and intensity modulated proton therapy. The risk of underdosing the tumor might increase unless great care is taken in the process. The information gained from imaging is increasing with both PET and MRI becoming readily available for the definition of targets. The information gained from these techniques is indeed multidimensional as one often acquire data on eg, metabolism, diffusion, and hypoxia together with anatomical and structural information. Nevertheless, much work remains to fully exploit the available information on a patient-specific level. Multimodality target definition in radiotherapy is a chain of processes that must be individually scrutinized, optimized and quality assured. Any uncertainties or errors in image acquisition, reconstruction, interpretation, and delineation are systematic errors and hence will potentially have a detrimental effect on the entire radiotherapy treatment and hence; the chance of cure or the risk of unnecessary side effects. Common guidelines and procedures create a common minimum standard and ground for evaluation and development. In Denmark, the treatment of head and neck cancer is organized within the multidisciplinary Danish Head and Neck Cancer Group (DAHANCA). The radiotherapy quality assurance group of DAHANCA organized a workshop in January 2020 with participants from oncology, radiology, and nuclear medicine from all centers in Denmark, treating patients with head and neck cancer. The participants agreed on a national guideline on imaging for target delineation in head and neck cancer radiotherapy, which has been approved by the DAHANCA group. The guidelines are available in the Supplementary. The use of multimodality imaging is being recommended for the planning of all radical treatments with a macroscopic tumor. 2-[¹⁸F]FDG-PET/CT should be available, preferable in the treatment position. The recommended MRI sequences are T1, T2 with and without fat suppression, and T1 with contrast enhancement, preferable in the treatment position. The interpretation of clinical information, including thorough physical examination as well as imaging, should be done in a multidisciplinary setting with an oncologist, radiologist, and nuclear medicine specialist.

Convolutional neural networks for brain tumour segmentation

The introduction of quantitative image analysis has given rise to fields such as radiomics which have been used to predict clinical sequelae. One growing area of interest for analysis is brain tumours, in particular glioblastoma multiforme (GBM). Tumour segmentation is an important step in the pipeline in the analysis of this pathology. Manual segmentation is often inconsistent as it varies between observers. Automated segmentation has been proposed to combat this issue. Methodologies such as convolutional neural networks (CNNs) which are machine learning pipelines modelled on the biological process of neurons (called nodes) and synapses (connections) have been of interest in the literature. We investigate the role of CNNs to segment brain tumours by firstly taking an educational look at CNNs and perform a literature search to determine an example pipeline for segmentation. We then investigate the future use of CNNs by exploring a novel field-radiomics. This examines quantitative features of brain tumours such as shape, texture, and signal intensity to predict clinical outcomes such as survival and response to therapy.

Deep multi-modality collaborative learning for distant metastases predication in PET-CT soft-tissue sarcoma studies

Soft-tissue Sarcomas (STS) are a heterogeneous group of malignant neoplasms with a relatively high mortality rate from distant metastases. Early prediction or quantitative evaluation of distant metastases risk for patients with STS is an important step which can provide better-personalized treatments and thereby improve survival rates. Positron emission tomography-computed tomography (PET-CT) image is regarded as the imaging modality of choice for the evaluation, staging and assessment of STS. Radiomics, which refers to the extraction and analysis of the quantitative of high-dimensional mineable data from medical images, is foreseen as an important prognostic tool for cancer risk assessment. However, conventional radiomics methods that depend heavily on hand-crafted features (e.g. shape and texture) and prior knowledge (e.g. tuning of many parameters) therefore cannot fully represent the semantic information of the image. In addition, convolutional neural networks (CNN) based radiomics methods present capabilities to improve, but currently, they are mainly designed for single modality e.g., CT or a particular body region e.g., lung structure. In this work, we propose a deep multi-modality collaborative learning to iteratively derive optimal ensembled deep and conventional features from PET-CT images. In addition, we introduce an end-to-end volumetric deep learning architecture to learn complementary PET-CT features optimised for image radiomics. Our experimental results using public PET-CT dataset of STS patients demonstrate that our method has better performance when compared with the state-of-the-art methods.

Radiomics and deep learning in lung cancer

Lung malignancies have been extensively characterized through radiomics and deep learning. By providing a three-dimensional characterization of the lesion, models based on radiomic features from computed tomography (CT) and positron-emission tomography (PET) have been developed to detect nodules, distinguish malignant from benign lesions, characterize their histology, stage, and genotype. Deep learning models have been applied to automatically segment organs at risk in lung cancer radiotherapy, stratify patients according to the risk for local and distant recurrence, and identify patients candidate for molecular targeted therapy and immunotherapy. Moreover, radiomics has also been applied successfully to predict side effects such as radiation- and immunotherapy-induced pneumonitis and differentiate lung injury from recurrence. Radiomics could also untap the potential for further use of the cone beam CT acquired for treatment image guidance, four-dimensional CT, and dose-volume data from radiotherapy treatment plans. Radiomics is expected to increasingly affect the clinical practice of treatment of lung tumors, optimizing the end-to-end diagnosis-treatment-follow-up chain. The main goal of this article is to provide an update on the current status of lung cancer radiomics.

Tumor Subregion Evolution-Based Imaging Features to Assess Early Response and Predict Prognosis in Oropharyngeal Cancer

The incidence of oropharyngeal squamous cell carcinoma (OPSCC) has been rapidly increasing. Disease stage and smoking history are often used in current clinical trials to select patients for deintensification therapy, but these features lack sufficient accuracy for predicting disease relapse. Our purpose was to develop an imaging signature to assess early response and predict outcomes of OPSCC. Methods: We retrospectively analyzed 162 OPSCC patients treated with concurrent chemoradiotherapy, equally divided into separate training and validation cohorts with similar clinical characteristics. A robust consensus clustering approach was used to spatially partition the primary tumor and involved lymph nodes into subregions (i.e., habitats) based on 18F-FDG PET and contrast CT imaging. We proposed quantitative image features to characterize the temporal volumetric change of the habitats and peritumoral/nodal tissue between baseline and midtreatment. The reproducibility of these features was evaluated. We developed an imaging signature to predict progression-free survival (PFS) by fitting an L1-regularized Cox regression model. Results: We identified 3 phenotypically distinct intratumoral habitats: metabolically active and heterogeneous, enhancing and heterogeneous, and metabolically inactive and homogeneous. The final Cox model consisted of 4 habitat evolution-based features. In both cohorts, this imaging signature significantly outperformed traditional imaging metrics, including midtreatment metabolic tumor volume for predicting PFS, with a C-index of 0.72 versus 0.67 (training) and 0.66 versus 0.56 (validation). The imaging signature stratified patients into high-risk versus low-risk groups with 2-y PFS rates of 59.1% versus 89.4% (hazard ratio, 4.4; 95% confidence interval, 1.4-13.4 [training]) and 61.4% versus 87.8% (hazard ratio, 4.6; 95% confidence interval, 1.7-12.1 [validation]). The imaging signature remained an independent predictor of PFS in multivariable analysis adjusting for stage, human papillomavirus status, and smoking history. Conclusion: The proposed imaging signature allows more accurate prediction of disease progression and, if prospectively validated, may refine OPSCC patient selection for risk-adaptive therapy.

PET-Derived Quantitative Metrics for Response and Prognosis in Lymphoma

Lymphoma is a potentially curable disease; however, the clinical challenge lies in further improvement of outcomes. PET with fludeoxyglucose is an effective imaging tool. PET-derived quantitative metrics have raised significant interest to be used as a prognostic factor to complement clinical parameters for treatment decisions. The most optimized use of these quantitative PET metrics, however, will be possible with the standardization of imaging procedures. In this article, we review the technical and methodological considerations related to PET-derived quantitative metrics, and the relevant published data to emphasize the potential value of these metrics in patient prognosis and treatment response in lymphoma.

A hybrid region growing tumour segmentation method for low contrast and high noise Nuclear Medicine (NM) images by combining a novel non-linear diffusion filter and global gradient measure (HNDF-GGM-RG)

Poor spatial resolution and low signal-to-noise ratio (SNR) along with the finite image sampling constraint make lesion segmentation on Nuclear Medicine (NM) images (e.g., PET-Positron Emission Tomography) a challenging task. Since the size, signal-to-background ratio (SBR) and SNR of lesion vary within and between patients, performance of the conventional segmentation methods are not consistent against statistical fluctuations. To overcome these limitations, a hybrid region growing segmentation method is proposed combining non-linear diffusion filter and global gradient measure (HNDF-GGM-RG). The performance of the algorithm is validated on PET images and compared with the 40%-fixed threshold and a state-of-the-art active contour (AC) methods. Segmented volume, dice similarity coefficient (DSC) and percentage classification error (% CE) were used as the quantitative figures of merit (FOM) using the torso NEMA phantom that contains six different sizes of spheres. A 2:1 SBR was created between the spheres and background and the phantom was scanned with a Siemens TrueV PET-CT scanner. 40T method is SNR dependent and overestimates the volumes ( ≈ 4.5  times ) . AC volumes match with the true volumes only for the largest three spheres. On the other hand, the proposed HNDF-GGM-RG volumes match closely with the true volumes irrespective of the size and SNR. Average DSC of 0.32 and 0.66 and % CE of 700% and 160% were achieved by the 40T and AC methods respectively. Conversely, average DSC and %CE are 0.70 and 60% for HNDF-GGM-RG and less dependent on SNR. Since two-sample t-test indicates that the performance of AC and HNDF-GGM-RG are statistically significant for the smallest three spheres and similar for the rest, HNDF-GGM-RG can be applied where the size, SBR and SNR are subject to change either due to alterations in the radiotracer uptake because of treatment or uptake variability of different radiotracers because of differences in their molecular pathways.

Quantitative imaging: Erring patterns in manual delineation of PET-imaged lung lesions

BACKGROUND AND PURPOSE

Uncertainty and variability in manual contouring of PET-imaged tumor targets are well recognized; however, the error patterns associated with it were little known and rarely investigated. The present study is aimed to quantitatively assess the erring patterns inherent to manual delineation of PET-imaged lung lesions in a setting with complete ground truth.

MATERIALS AND METHODS

Images being used for assessment consisted of 26 synthetic PET datasets created by using the anthropomorphic Zubal phantom in conjunction with the Monte Carlo based SimSET computational package. Each dataset included one PET-positive lesion differing in shape, dimension, uptake heterogeneity, and anatomical location inside the lung. Target contours were provided by 10 raters and the contour accuracy was evaluated using 12 metrics from five categories - spatial overlap, pair counting, information theory, distance, and volume.

RESULTS

In terms of spatial overlap, manual contouring results intersect substantially with the ground truth whereas tend to oversegment the lesions. Shapes of the segmented tumor volumes are in general geometrically consistent with the ground truth but lack sensitivity in characterizing topographical details. No complete consensus could be achieved between manual contours and the ground truth for any of the given lesions being examined when assessing using pair counting- and informatics-based metrics thus indicating an intrinsic stochastic component of manual contouring. Evaluation based on metrics related to distance and volume demonstrated that it is at the borderline areas between the lesions and the normal tissues where the majority part of manual delineation errors occurred and the extent of volume being identified false positively as cancerous by the raters is appreciable.

CONCLUSION

Quantification of segmentation errors associated with expert manual contouring of PET positive lesion in the lung reveals general patterns in what otherwise might be thought of as randomness. Findings from the current study may allow for the formation of new hypotheses towards improving the accuracy and precision of manual delineation of PET positive tumor targets in the lung.

Textural features on 18F-FDG PET/CT and dynamic contrast-enhanced MR imaging for predicting treatment response and survival of patients with hypopharyngeal carcinoma

The utility of multimodality molecular imaging for predicting treatment response and survival of patients with hypopharyngeal carcinoma remains unclear. Here, we sought to investigate whether the combination of different molecular imaging parameters may improve outcome prediction in this patient group.Patients with pathologically proven hypopharyngeal carcinoma scheduled to undergo chemoradiotherapy (CRT) were deemed eligible. Besides clinical data, parameters obtained from pretreatment 2-deoxy-2-[fluorine-18]fluoro-D-glucose positron emission tomography/computed tomography (F-FDG PET/CT), dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI), and diffusion-weighted MRI were analyzed in relation to treatment response, recurrence-free survival (RFS), and overall survival (OS).A total of 61 patients with advanced-stage disease were examined. After CRT, 36% of the patients did not achieve a complete response. Total lesion glycolysis (TLG) and texture feature entropy were found to predict treatment response. The transfer constant (K), TLG, and entropy were associated with RFS, whereas K, blood plasma volume (Vp), standardized uptake value (SUV), and entropy were predictors of OS. Different scoring systems based on the sum of PET- or MRI-derived prognosticators enabled patient stratification into distinct prognostic groups (P <.0001). The complete response rate of patients with a score of 2 was significantly lower than those of patients with a score 1 or 0 (14.7% vs 58.9% vs 75.7%, respectively, P = .007, respectively). The combination of PET- and DCE-MRI-derived independent risk factors allowed a better survival stratification than the TNM staging system (P <.0001 vs .691, respectively).Texture features on F-FDG PET/CT and DCE-MRI are clinically useful to predict treatment response and survival in patients with hypopharyngeal carcinoma. Their combined use in prognostic scoring systems may help these patients benefit from tailored treatment and obtain better oncological results.

A phantom study to assess the reproducibility, robustness and accuracy of PET image segmentation methods against statistical fluctuations

Background

Automatic and semi-automatic segmentation methods for PET serve as alternatives to manual delineation and eliminate observer variability. The robustness of these segmentation methods against statistical fluctuations arising from variable size, contrast and noise are vital for providing reliable clinical outcomes for diagnosis and treatment response assessment. In this study, the performances of several segmentation methods have been investigated using the torso NEMA phantom against statistical fluctuations.

Methods

The six hot spheres (0.5-27ml) and the background of the phantom were filled with different activities of ¹⁸F to yield 2:1 and 4:1 contrast ratios. The phantom was scanned on a TrueV PET-CT scanner for 120 minutes. The images were reconstructed using OSEM (4iterations-21subsets) for different durations (15, 20, 34 and 67 minutes) to represent different noise levels and smoothed with a 4-mm Gaussian filter. Each sphere with different settings was delineated using a fixed 40% threshold (40T), fuzzy clustering mean (FCM), adaptive threshold and region based variational (C-V) segmentation methods and compared with the gold standard volume, which was estimated from the known diameter and position of each sphere.

Results

The smallest three spheres at the 2:1 contrast level are not evaluable for the 40T method. For the other spheres, the 40T method grossly overestimates the volumes and the segmented volumes are highly dependent on the statistical variations. These volumes are the least reproducible (80%) with a mean Dice Similarity Coefficient (DSC) of 0.67 and 90% classification error (CE). The other three methods reduce the dependency on noise and contrast in a similar manner by providing low bias (<10%) and CE (<25%) as well as a high DSC (0.88) and reproducibility (30%) for objects >17mm in diameter. However, for the smallest three spheres at a 2:1 contrast level, the performances of all three methods were significantly low, with the adaptive method being superior to the FCM and C-V (mean bias 168% and 350%, mean DSC 0.65 and 0.50, mean CE 227% and 454% for the adaptive and other two methods (approximately similar for FCM and C-V), respectively).

Conclusions

The segmentation accuracy of the fixed threshold-based method depends on size, contrast and noise. The intensity thresholds determined by the adaptive threshold methods are less sensitive to noise and therefore, the segmented volumes are more reproducible across different acquisition durations. A similar performance can be achieved with the FCM and C-V methods. Though, for small lesions (< 2cm diameter) with low counts and contrast, the adaptive threshold-based method outperforms the FCM and C-V methods, and the performance of neither of these methods is optimal for volumes <2cm in diameter. These three methods can only reliably be used to delineate tumours for diagnostic and monitoring purposes provided that the contrast between the tumour and background is not below a 2:1 ratio and the size of the tumour does not fall not below 2cm in diameter in response to treatment. They can also be used for different radiotracers with variable uptake. However, the FCM and C-V methods have the advantage of not requiring calibrations for different scanners and settings.

Radiomics in nuclear medicine: robustness, reproducibility, standardization, and how to avoid data analysis traps and replication crisis

Radiomics in nuclear medicine is rapidly expanding. Reproducibility of radiomics studies in multicentre settings is an important criterion for clinical translation. We therefore performed a meta-analysis to investigate reproducibility of radiomics biomarkers in PET imaging and to obtain quantitative information regarding their sensitivity to variations in various imaging and radiomics-related factors as well as their inherent sensitivity. Additionally, we identify and describe data analysis pitfalls that affect the reproducibility and generalizability of radiomics studies. After a systematic literature search, 42 studies were included in the qualitative synthesis, and data from 21 were used for the quantitative meta-analysis. Data concerning measurement agreement and reliability were collected for 21 of 38 different factors associated with image acquisition, reconstruction, segmentation and radiomics-specific processing steps. Variations in voxel size, segmentation and several reconstruction parameters strongly affected reproducibility, but the level of evidence remained weak. Based on the meta-analysis, we also assessed inherent sensitivity to variations of 110 PET image biomarkers. SUV_mean and SUV_max were found to be reliable, whereas image biomarkers based on the neighbourhood grey tone difference matrix and most biomarkers based on the size zone matrix were found to be highly sensitive to variations, and should be used with care in multicentre settings. Lastly, we identify 11 data analysis pitfalls. These pitfalls concern model validation and information leakage during model development, but also relate to reporting and the software used for data analysis. Avoiding such pitfalls is essential for minimizing bias in the results and to enable reproduction and validation of radiomics studies.

Machine-learned target volume delineation of 18F-FDG PET images after one cycle of induction chemotherapy

Biological tumour volume (GTV_PET) delineation on ¹⁸F-FDG PET acquired during induction chemotherapy (ICT) is challenging due to the reduced metabolic uptake and volume of the GTV_PET. Automatic segmentation algorithms applied to ¹⁸F-FDG PET (PET-AS) imaging have been used for GTV_PET delineation on ¹⁸F-FDG PET imaging acquired before ICT. However, their role has not been investigated in ¹⁸F-FDG PET imaging acquired after ICT. In this study we investigate PET-AS techniques, including ATLAAS a machine learned method, for accurate delineation of the GTV_PET after ICT. Twenty patients were enrolled onto a prospective phase I study (FiGaRO). PET/CT imaging was acquired at baseline and 3 weeks following 1 cycle of induction chemotherapy. The GTV_PET was manually delineated by a nuclear medicine physician and clinical oncologist. The resulting GTV_PET was used as the reference contour. The ATLAAS original statistical model was expanded to include images of reduced metabolic activity and the ATLAAS algorithm was re-trained on the new reference dataset. Estimated GTV_PET contours were derived using sixteen PET-AS methods and compared to the GTV_PET using the Dice Similarity Coefficient (DSC). The mean DSC for ATLAAS, 60% Peak Thresholding (PT60), Adaptive Thresholding (AT) and Watershed Thresholding (WT) was 0.72, 0.61, 0.63 and 0.60 respectively. The GTV_PET generated by ATLAAS compared favourably with manually delineated volumes and in comparison, to other PET-AS methods, was more accurate for GTV_PET delineation after ICT. ATLAAS would be a feasible method to reduce inter-observer variability in multi-centre trials.