A novel iterative method for lesion delineation and volumetric quantification with FDG PET

PET-Based Volumetric Biomarkers for Risk Stratification of Non-Small Cell Lung Cancer Patients

Despite the recent advances in lung cancer biology, molecular pathology, and treatment, this malignancy remains the leading cause of cancer-related death worldwide and non-small cell lung cancer (NSCLC) is the most common form found at diagnosis. Accurate staging of the disease is a fundamental prognostic factor that correctly predicts progression-free (PFS) and overall survival (OS) of NSCLC patients. However, outcome of patients within each TNM staging group can change widely highlighting the need to identify additional prognostic biomarkers to better stratify patients on the basis of risk. 18F-FDG PET/CT plays an essential role in staging, evaluation of treatment response, and tumoral target delineation in NSCLC patients. Moreover, a number of studies showed the prognostic role of imaging parameters derived from PET images, such as metabolic tumor volume (MTV) and total lesion glycolysis (TLG). These parameters represent three-dimensional PET-based measurements providing information on both tumor volume and metabolic activity and previous studies reported their ability to predict OS and PFS of NSCLC patients. This review will primarily focus on the studies that showed the prognostic and predictive role of MTV and TLG in NSCLC patients, addressing also their potential utility in the new era of immunotherapy of NSCLC.

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.

Prognostic value of SUR in patients with trimodality treatment of locally advanced esophageal carcinoma

The prognosis of patients with esophageal carcinoma remains dismal despite ongoing efforts to improve treatment options. For locally advanced tumors, several randomized trials have shown the benefit of neoadjuvant chemoradiation followed by surgery compared to surgery alone. The aim of this exploratory study was to evaluate the prognostic value of different baseline positron emission tomography (PET) parameters and their potentially additional prognostic impact at the end of neoadjuvant radiochemotherapy. Furthermore, the standard uptake ratio (SUR) as a new parameter for quantification of tumor metabolism was compared to the conventional PET parameters metabolic active volume (MTV), total lesion glycolysis (TLG), and standardized uptake value (SUV) taking into account known basic parameters. Methods:18F-FDG-PET/CT was performed in 76 consecutive patients ((60±10) years, 71 males) with newly diagnosed esophageal cancer before and during the last week of neoadjuvant radiochemotherapy. MTV of the primary tumor was delineated with an adaptive threshold method. The blood SUV was determined by manually delineating the aorta in the low dose CT. SUR values were computed as scan time corrected ratio of tumor SUVmax and mean blood SUV. Univariate Cox regression and Kaplan-Meier analysis with respect to locoregional control (LRC), freedom from distant metastases (FFDM), and overall survival (OS) was performed. Additionally, independence of PET parameters from standard clinical factors was analyzed with multivariate Cox regression. Results: In multivariate analysis two parameters showed a significant correlation with all endpoints: restaging MTV and restaging SUR. Furthermore, restaging TLG was prognostic for LCR and FFDM. For all endpoints the largest effect size was found for restaging SUR. The only basic factors remaining significant in multivariate analyses were histology for OS and FFDM and age for LRC. Conclusion: PET provides independent prognostic information for OS, LRC, and FFDM in addition to standard clinical parameters in this patient cohort. Our results suggest that the prognostic value of tracer uptake can be improved when characterized by SUR rather than by SUV. Overall, our investigation revealed a higher prognostic value of restaging parameters compared to baseline PET; therapy-adjustments would still be possible at this point of time. Further investigations are required to confirm these hypothesis-generating results.

Molecular Imaging-Guided Radiotherapy for the Treatment of Head-and-Neck Squamous Cell Carcinoma: Does it Fulfill the Promises?

With the routine use of intensity modulated radiation therapy for the treatment of head-and-neck squamous cell carcinoma allowing highly conformed dose distribution, there is an increasing need for refining both the selection and the delineation of gross tumor volumes (GTV). In this framework, molecular imaging with positron emission tomography and magnetic resonance imaging offers the opportunity to improve diagnostic accuracy and to integrate tumor biology mainly related to the assessment of tumor cell density, tumor hypoxia, and tumor proliferation into the treatment planning equation. Such integration, however, requires a deep comprehension of the technical and methodological issues related to image acquisition, reconstruction, and segmentation. Until now, molecular imaging has had a limited value for the selection of nodal GTV, but there are increasing evidences that both FDG positron emission tomography and diffusion-weighted magnetic resonance imaging has a potential value for the delineation of the primary tumor GTV, effecting on dose distribution. With the apprehension of the heterogeneity in tumor biology through molecular imaging, growing evidences have been collected over the years to support the concept of dose escalation/dose redistribution using a planned heterogeneous dose prescription, the so-called "dose painting" approach. Validation trials are ongoing, and in the coming years, one may expect to position the dose painting approach in the armamentarium for the treatment of patients with head-and-neck squamous cell carcinoma.

Pretherapeutic FDG-PET total metabolic tumor volume predicts response to induction therapy in pediatric Hodgkin's lymphoma

Background

Standardized treatment in pediatric patients with Hodgkin’s lymphoma (HL) follows risk stratification by tumor stage, erythrocyte sedimentation rate and tumor bulk. We aimed to identify quantitative parameters from pretherapeutic FDG-PET to assist prediction of response to induction chemotherapy.

Methods

Retrospective analysis in 50 children with HL (f:18; m:32; median age, 14.8 [4–18] a) consecutively treated according to EuroNet-PHL-C1 (n = 42) or -C2 treatment protocol (n = 8). Total metabolic tumor volume (MTV) in pretherapeutic FDG-PET was defined using a semi-automated, background-adapted threshold. Metabolic (SUVmax, SUVmean, SUVpeak, total lesion glycolysis [MTV*SUVmean]) and heterogeneity parameters (asphericity [ASP], entropy, contrast, local homogeneity, energy, and cumulative SUV-volume histograms) were derived. Early response assessment (ERA) was performed after 2 cycles of induction chemotherapy according to treatment protocol and verified by reference rating. Prediction of inadequate response (IR) in ERA was based on ROC analysis separated by stage I/II (1 and 26 patients) and stage III/IV disease (7 and 16 patients) or treatment group/level (TG/TL) 1 to 3.

Results

IR was seen in 28/50 patients (TG/TL 1, 6/12 patients; TG/TL 2, 10/17; TG/TL 3, 12/21). Among all PET parameters, MTV best predicted IR; ASP was the best heterogeneity parameter. AUC of MTV was 0.84 (95%-confidence interval, 0.69–0.99) in stage I/II and 0.86 (0.7–1.0) in stage III/IV. In patients of TG/TL 1, AUC of MTV was 0.92 (0.74–1.0); in TG/TL 2 0.71 (0.44–0.99), and in TG/TL 3 0.85 (0.69–1.0). Patients with high vs. low MTV had IR in 86 vs. 0% in TG/TL 1, 80 vs. 29% in TG/TL 2, and 90 vs. 27% in TG/TL 3 (cut-off, > 80 ml, > 160 ml, > 410 ml).

Conclusions

In this explorative study, high total MTV best predicted inadequate response to induction therapy in pediatric HL of all pretherapeutic FDG-PET parameters – in both low and high stages as well as the 3 different TG/TL.

Trial registration

Ethics committee number: EA2/151/16 (retrospectively registered).

Electronic supplementary material

The online version of this article (10.1186/s12885-018-4432-4) contains supplementary material, which is available to authorized users.

Joint solution for PET image segmentation, denoising, and partial volume correction

Segmentation, denoising, and partial volume correction (PVC) are three major processes in the quantification of uptake regions in post-reconstruction PET images. These problems are conventionally addressed by independent steps. In this study, we hypothesize that these three processes are dependent; therefore, jointly solving them can provide optimal support for quantification of the PET images. To achieve this, we utilize interactions among these processes when designing solutions for each challenge. We also demonstrate that segmentation can help in denoising and PVC by locally constraining the smoothness and correction criteria. For denoising, we adapt generalized Anscombe transformation to Gaussianize the multiplicative noise followed by a new adaptive smoothing algorithm called regional mean denoising. For PVC, we propose a volume consistency-based iterative voxel-based correction algorithm in which denoised and delineated PET images guide the correction process during each iteration precisely. For PET image segmentation, we use affinity propagation (AP)-based iterative clustering method that helps the integration of PVC and denoising algorithms into the delineation process. Qualitative and quantitative results, obtained from phantoms, clinical, and pre-clinical data, show that the proposed framework provides an improved and joint solution for segmentation, denoising, and partial volume correction.

Radiomics Features Differentiate Between Normal and Tumoral High-Fdg Uptake

Identification of FDGavid- neoplasms may be obscured by high-uptake normal tissues, thus limiting inferences about the natural history of disease. We introduce a FDG-PET radiomics tissue classifier for differentiating FDGavid- normal tissues from tumor. Thirty-three scans from 15 patients with Hodgkin lymphoma and 68 scans from 23 patients with Ewing sarcoma treated on two prospective clinical trials were retrospectively analyzed. Disease volumes were manually segmented on FDG-PET and CT scans. Brain, heart, kidneys and bladder and tumor volumes were automatically segmented on PET images. Standard-uptake-value (SUV) derived shape and first order radiomics features were computed to build a random forest classifier. Manually segmented volumes were compared to automatically segmented tumor volumes. Classifier accuracy for normal tissues was 90%. Classifier performance was varied across normal tissue types (brain, left kidney and bladder, hear and right kidney were 100%, 96%, 97%, 83% and 87% respectively). Automatically segmented tumor volumes showed high concordance with the manually segmented tumor volumes (R² = 0.97). Inclusion of texture-based radiomics features minimally contributed to classifier performance. Accurate normal tissue segmentation and classification facilitates accurate identification of FDGavid tissues and classification of those tissues as either tumor or normal tissue.

FDG PET/CT in cancer therapy monitoring

Objectives: We developed and tested a software tool for computer-assisted analysis of FDG-PET/CT in cancer therapy monitoring. The tool provides automatic semi-quantitative analysis of a baseline scan together with up to two follow-up scans (standardized uptake values, glycolytic volume). The tool also supports visual analysis by local spatial registration which allows display of tumor lesions with the same orientation in all scans. The tool’s stability and accuracy was tested at typical everyday image quality. Patients, methods: Ten unselected cancer patients in whom three FDG PET/CT scans had been performed were included. A total of 18 lesions were analyzed. Results: Automatic lesion tracking worked properly in all lesions but one. In this lesion local coregistration had to be adjusted manually tuwhich, however, is easily performed with the tool. Semi-automatic lesion segmentation and fully automatic semi-quantitative analysis worked properly in all cases. Computer-assisted analysis was significantly less time consuming than manual analysis. Conclusions: The novel software tool appears useful for analysis of FDGPET/ CT in cancer therapy monitoring in clinical routine patient care.

Potential of asphericity as a novel diagnostic parameter in the evaluation of patients with 68Ga-PSMA-HBED-CC PET-positive prostate cancer lesions

BACKGROUND

The aim of this study was to evaluate the diagnostic value of the asphericity (ASP) as a novel quantitative parameter, reflecting the spatial heterogeneity of tracer uptake, in the staging process of patients with ⁶⁸Ga-PSMA-HBED-CC positron emission tomography (PET)-positive prostate cancer (PC). In this study, 37 patients (median age 72 years, range 52-82 years) with newly diagnosed PC, who received a ⁶⁸Ga-PSMA-HBED-CC PET fused with computed tomography (⁶⁸Ga-PSMA-PET/CT), a magnetic resonance imaging (MRI) of the prostate, and a core needle biopsy (within 74.2 ± 80.2 days) with an available Gleason score (GSc) were extracted from the local database. The ASP and the viable tumor volume (VTV) was calculated using the rover software (ABX GmbH, Radeberg, Germany), a segmentation tool for automated tumor volume delineation. Additionally, parameters including total lesion binding rate (TLB), maximum, mean and peak standardized uptake value (SUVmax/mean/peak), prostate-specific antigen (PSA), D'Amico classification, and prostate imaging reporting and data system (PI-RADS) were analyzed.

RESULTS

The ASP mean differed significantly (p ≤ 0.05) between the different GSc groups: GSc 6-7: 11.9 ± 4.8%, GSc 8: 25.5 ± 4.8%, GSc 9-10: 33.3 ± 6.8%. A significant correlation between ASP and GSc (rho = 0.88; CI 0.78-0.94; p < 0.05) was measured. The ASP enabled an independent (p > 0.05) prediction of the GSc. A moderate correlation was measured between ASP and the D'Amico classification (rho = 0.6; CI 0.32-0.78; p < 0.05). The VTV showed a moderate correlation with the SUVmax (rho = 0.58; CI 0.32-0.76; p < 0.05) and the GSc (rho = 0.51; CI 0.23-0.72; p < 0.05).

CONCLUSION

The asphericity in ⁶⁸Ga-PSMA-PET could represent a promising novel quantitative parameter for an improved non-invasive tumor staging of patients with PC.

Adaptive region-growing with maximum curvature strategy for tumor segmentation in 18F-FDG PET

Accurate tumor segmentation in PET is crucial in many oncology applications. We developed an adaptive region-growing (ARG) algorithm with a maximum curvature strategy (ARG_MC) for tumor segmentation in PET. The ARG_MC repeatedly applied a confidence connected region-growing algorithm with increasing relaxing factor f. The optimal relaxing factor (ORF) was then determined at the transition point on the f-volume curve, where the volume just grew from the tumor into the surrounding normal tissues. The ARG_MC along with five widely used algorithms were tested on a phantom with 6 spheres at different signal to background ratios and on two clinic datasets including 20 patients with esophageal cancer and 11 patients with non-Hodgkin lymphoma (NHL). The ARG_MC did not require any phantom calibration or any a priori knowledge of the tumor or PET scanner. The identified ORF varied with tumor types (mean ORF  =  9.61, 3.78 and 2.55 respectively for the phantom, esophageal cancer, and NHL datasets), and varied from one tumor to another. For the phantom, the ARG_MC ranked the second in segmentation accuracy with an average Dice similarity index (DSI) of 0.86, only slightly worse than Daisne's adaptive thresholding method (DSI  =  0.87), which required phantom calibration. For both the esophageal cancer dataset and the NHL dataset, the ARG_MC had the highest accuracy with an average DSI of 0.87 and 0.84, respectively. The ARG_MC was robust to parameter settings and region of interest selection, and it did not depend on scanners, imaging protocols, or tumor types. Furthermore, the ARG_MC made no assumption about the tumor size or tumor uptake distribution, making it suitable for segmenting tumors with heterogeneous FDG uptake. In conclusion, the ARG_MC was accurate, robust and easy to use, it provides a highly potential tool for PET tumor segmentation in clinic.

Classification and evaluation strategies of auto-segmentation approaches for PET: Report of AAPM task group No. 211

PURPOSE

The purpose of this educational report is to provide an overview of the present state-of-the-art PET auto-segmentation (PET-AS) algorithms and their respective validation, with an emphasis on providing the user with help in understanding the challenges and pitfalls associated with selecting and implementing a PET-AS algorithm for a particular application.

APPROACH

A brief description of the different types of PET-AS algorithms is provided using a classification based on method complexity and type. The advantages and the limitations of the current PET-AS algorithms are highlighted based on current publications and existing comparison studies. A review of the available image datasets and contour evaluation metrics in terms of their applicability for establishing a standardized evaluation of PET-AS algorithms is provided. The performance requirements for the algorithms and their dependence on the application, the radiotracer used and the evaluation criteria are described and discussed. Finally, a procedure for algorithm acceptance and implementation, as well as the complementary role of manual and auto-segmentation are addressed.

FINDINGS

A large number of PET-AS algorithms have been developed within the last 20 years. Many of the proposed algorithms are based on either fixed or adaptively selected thresholds. More recently, numerous papers have proposed the use of more advanced image analysis paradigms to perform semi-automated delineation of the PET images. However, the level of algorithm validation is variable and for most published algorithms is either insufficient or inconsistent which prevents recommending a single algorithm. This is compounded by the fact that realistic image configurations with low signal-to-noise ratios (SNR) and heterogeneous tracer distributions have rarely been used. Large variations in the evaluation methods used in the literature point to the need for a standardized evaluation protocol.

CONCLUSIONS

Available comparison studies suggest that PET-AS algorithms relying on advanced image analysis paradigms provide generally more accurate segmentation than approaches based on PET activity thresholds, particularly for realistic configurations. However, this may not be the case for simple shape lesions in situations with a narrower range of parameters, where simpler methods may also perform well. Recent algorithms which employ some type of consensus or automatic selection between several PET-AS methods have potential to overcome the limitations of the individual methods when appropriately trained. In either case, accuracy evaluation is required for each different PET scanner and scanning and image reconstruction protocol. For the simpler, less robust approaches, adaptation to scanning conditions, tumor type, and tumor location by optimization of parameters is necessary. The results from the method evaluation stage can be used to estimate the contouring uncertainty. All PET-AS contours should be critically verified by a physician. A standard test, i.e., a benchmark dedicated to evaluating both existing and future PET-AS algorithms needs to be designed, to aid clinicians in evaluating and selecting PET-AS algorithms and to establish performance limits for their acceptance for clinical use. The initial steps toward designing and building such a standard are undertaken by the task group members.

Tumor Delineation and Quantitative Assessment of Glucose Metabolic Rate within Histologic Subtypes of Non-Small Cell Lung Cancer by Using Dynamic 18F Fluorodeoxyglucose PET

Purpose To assess whether dynamic fluorine 18 (¹⁸F) fluorodeoxyglucose (FDG) positron emission tomography (PET) has added value over static ¹⁸F-FDG PET for tumor delineation in non-small cell lung cancer (NSCLC) radiation therapy planning by using pathology volumes as the reference standard and to compare pharmacokinetic rate constants of ¹⁸F-FDG metabolism, including regional variation, between NSCLC histologic subtypes. Materials and Methods The study was approved by the institutional review board. Patients gave written informed consent. In this prospective observational study, 1-hour dynamic ¹⁸F-FDG PET/computed tomographic examinations were performed in 35 patients (36 resectable NSCLCs) between 2009 and 2014. Static and parametric images of glucose metabolic rate were obtained to determine lesion volumes by using three delineation strategies. Pathology volume was calculated from three orthogonal dimensions (n = 32). Whole tumor and regional rate constants and blood volume fraction (V_B) were computed by using compartment modeling. Results Pathology volumes were larger than PET volumes (median difference, 8.7-25.2 cm³; Wilcoxon signed rank test, P < .001). Static fuzzy locally adaptive Bayesian (FLAB) volumes corresponded best with pathology volumes (intraclass correlation coefficient, 0.72; P < .001). Bland-Altman analyses showed the highest precision and accuracy for static FLAB volumes. Glucose metabolic rate and ¹⁸F-FDG phosphorylation rate were higher in squamous cell carcinoma (SCC) than in adenocarcinoma (AC), whereas V_B was lower (Mann-Whitney U test or t test, P = .003, P = .036, and P = .019, respectively). Glucose metabolic rate, ¹⁸F-FDG phosphorylation rate, and V_B were less heterogeneous in AC than in SCC (Friedman analysis of variance). Conclusion Parametric images are not superior to static images for NSCLC delineation. FLAB-based segmentation on static ¹⁸F-FDG PET images is in best agreement with pathology volume and could be useful for NSCLC autocontouring. Differences in glycolytic rate and V_B between SCC and AC are relevant for research in targeting agents and radiation therapy dose escalation. ^© RSNA, 2016 Online supplemental material is available for this article.

Performance of automatic image segmentation algorithms for calculating total lesion glycolysis for early response monitoring in non-small cell lung cancer patients during concomitant chemoradiotherapy

BACKGROUND AND PURPOSE

This study evaluated the use of total lesion glycolysis (TLG) determined by different automatic segmentation algorithms, for early response monitoring in non-small cell lung cancer (NSCLC) patients during concomitant chemoradiotherapy.

MATERIALS AND METHODS

Twenty-seven patients with locally advanced NSCLC treated with concomitant chemoradiotherapy underwent (18)F-fluorodeoxyglucose (FDG) PET/CT imaging before and in the second week of treatment. Segmentation of the primary tumours and lymph nodes was performed using fixed threshold segmentation at (i) 40% SUVmax (T40), (ii) 50% SUVmax (T50), (iii) relative-threshold-level (RTL), (iv) signal-to-background ratio (SBR), and (v) fuzzy locally adaptive Bayesian (FLAB) segmentation. Association of primary tumour TLG (TLGT), lymph node TLG (TLGLN), summed TLG (TLGS=TLGT+TLGLN), and relative TLG decrease (ΔTLG) with overall-survival (OS) and progression-free survival (PFS) was determined using univariate Cox regression models.

RESULTS

Pretreatment TLGT was predictive for PFS and OS, irrespective of the segmentation method used. Inclusion of TLGLN improved disease and early response assessment, with pretreatment TLGS more strongly associated with PFS and OS than TLGT for all segmentation algorithms. This was also the case for ΔTLGS, which was significantly associated with PFS and OS, with the exception of RTL and T40.

CONCLUSIONS

ΔTLGS was significantly associated with PFS and OS, except for RTL and T40. Inclusion of TLGLN improves early treatment response monitoring during concomitant chemoradiotherapy with FDG-PET.

FDG-PET-based differential uptake volume histograms: a possible approach towards definition of biological target volumes

OBJECTIVE

Integration of fluorine-18 fludeoxyglucose ((18)F-FDG)-positron emission tomography (PET) functional data into conventional anatomically based gross tumour volume delineation may lead to optimization of dose to biological target volumes (BTV) in radiotherapy. We describe a method for defining tumour subvolumes using (18)F-FDG-PET data, based on the decomposition of differential uptake volume histograms (dUVHs).

METHODS

For 27 patients with histopathologically proven non-small-cell lung carcinoma (NSCLC), background uptake values were sampled within the healthy lung contralateral to a tumour in those image slices containing tumour and then scaled by the ratio of mass densities between the healthy lung and tumour. Signal-to-background (S/B) uptake values within volumes of interest encompassing the tumour were used to reconstruct the dUVHs. These were subsequently decomposed into the minimum number of analytical functions (in the form of differential uptake values as a function of S/B) that yielded acceptable net fits, as assessed by χ(2) values.

RESULTS

Six subvolumes consistently emerged from the fitted dUVHs over the sampled volume of interest on PET images. Based on the assumption that each function used to decompose the dUVH may correspond to a single subvolume, the intersection between the two adjacent functions could be interpreted as a threshold value that differentiates them. Assuming that the first two subvolumes spread over the tumour boundary, we concentrated on four subvolumes with the highest uptake values, and their S/B thresholds [mean ± standard deviation (SD)] were 2.88 ± 0.98, 4.05 ± 1.55, 5.48 ± 2.06 and 7.34 ± 2.89 for adenocarcinoma, 3.01 ± 0.71, 4.40 ± 0.91, 5.99 ± 1.31 and 8.17 ± 2.42 for large-cell carcinoma and 4.54 ± 2.11, 6.46 ± 2.43, 8.87 ± 5.37 and 12.11 ± 7.28 for squamous cell carcinoma, respectively.

CONCLUSION

(18)F-FDG-based PET data may potentially be used to identify BTV within the tumour in patients with NSCLC. Using the one-way analysis of variance statistical tests, we found a significant difference among all threshold levels among adenocarcinomas, large-cell carcinoma and squamous cell carcinomas. On the other hand, the observed significant variability in threshold values throughout the patient cohort (expressed as large SDs) can be explained as a consequence of differences in the physiological status of the tumour volume for each patient at the time of the PET/CT scan. This further suggests that patient-specific threshold values for the definition of BTVs could be determined by creation and curve fitting of dUVHs on a patient-by-patient basis.

ADVANCES IN KNOWLEDGE

The method of (18)F-FDG-PET-based dUVH decomposition described in this work may lead to BTV segmentation in tumours.

Update on F-18-fluoro-deoxy-glucose-PET/computed tomography in nonsmall cell lung cancer

PURPOSE OF REVIEW

The aim of this review is to provide an outline of current evidence for the use of F-18-fluoro-deoxy-glucose PET computed tomography (FDG-PET/CT) in nonsmall cell lung cancer (NSCLC) for diagnosis, staging, radiotherapy planning, response assessment and response monitoring.

RECENT FINDINGS

Management of patients with NSCLC requires a multimodality approach to accurately diagnose and stage patients. In this approach, FDG-PET/CT has become a standard staging instrument in lung cancer. FDG-PET/CT is, in addition to staging, also valuable for the characterization of the solitary pulmonary nodule. An increased uptake in the nodule as compared with mediastinal blood pool is suspected for malignancy. In radiotherapy planning, FDG-PET/CT can assist the radiation oncologist for optimal dose delivery to the tumour, while sparing healthy tissues. Evidence of the prognostic and predictive implications of FDG-PET/CT is accumulating. Volumetric parameters of PET, such as metabolic active tumour volume and total lesion glycolysis, are promising predictive and prognostic biomarkers. However, for implementation of metabolic response parameters in clinical practice, more randomized, PET-based, multicentre trials are necessary. The introduction of integrated PET and MRI scanners did not change the pivotal role of standard FDG-PET/CT yet, as with current technology, PET/MRI did not show superior performance in thoracic staging.

SUMMARY

The role of PET is described for diagnosis, staging and response assessment.

A new method for segmentation of FDG PET metabolic tumour volume using the peritumoural halo layer and a 10-step colour scale. A study in patients with papillary thyroid carcinoma

AIM

We observed a layer between tumour activity and background on FDG PET/CT with the 10-step colour scale and the window level set properly. We named the layer peritumoral halo layer (PHL). We performed this study to establish the reliability of metabolic tumor volume (MTV) segmentation using PHL (MTV(PHL)) in patients with papillary thyroid carcinoma.

PATIENTS, METHODS

Of a total of 140 papillary thyroid carcinoma (PTC) patients, 70 (50.0%) had FDG-avid PTC. In these patients, MTV(PHL), MTV segmented according to fixed 50% SUVmax (MTV(50%)), and fixed SUV with 2.5 to 4.0 (MTV(2.5) to MTV(4.0)) were compared with pathologic tumour volume (PTV). The absolute percentage difference between MTV(PHL) and PTV was compared in micropapillary carcinoma (MPTC) and non-micropapillary carcinoma (non-MPTC) subgroups. The % SUVmax and SUV thresholds of MTV(PHL) were compared with tumour SUVmax.

RESULTS

Among the MTVs, MTV(50%) was not correlated with PTV (r = -0.16, p = 0.182) and was not reliable according to the Bland-Altman plot. Although MTV(2.5), MTV(3.0), MTV(3.5), and MTV(4.0) correlated with PTV (r = 0.85, 0.86, 0.87, and 0.87, respectively; p < 0.001), these MTVs were not reliable on Bland-Altman analyses. MTV(PHL) was significantly correlated with PTV (r = 0.80, p < 0.001), and the Bland-Altman plot did not show systemic error. The MTV(PHL) was more accurate in non-MPTC than in MPTC (p < 0.001), and the absolute % difference was smaller as PTV became larger (σ = -0.65, p < 0.001). The MTV(PHL) thresholds had correlations with SUVmax (% SUVmax threshold: σ = -0.87, p < 0.001; SUV threshold: r = 0.88, p < 0.001).

CONCLUSIONS

MTV(PHL) was more reliable than MTV(%SUVmax) or MTV(SUV). The reliability of MTV(PHL) improved with larger PTVs. The threshold of the MTV(PHL) was naturally altered by PHL according to SUVmax.

Glucose metabolism in NSCLC is histology-specific and diverges the prognostic potential of 18FDG-PET for adenocarcinoma and squamous cell carcinoma

INTRODUCTION

Biological features of non-small-cell lung carcinomas (NSCLCs) are important determinants for prognosis. In this study, differences in glucose metabolism between adeno- and squamous cell NSCLCs were quantified using the hypoxia and glycolysis-related markers glucose transporter 1 (GLUT1), carbonic anhydrase IX (CAIX), monocarboxylate transporter 1 (MCT1) and 4 (MCT4) vasculature, and 18-fluoro-2-deoxyglucose (FDG)-uptake. Relevance of these markers for disease-free survival (DFS) was analyzed.

METHODS

Patients with curatively resected stage I to II and resectable stage IIIA, cN0-1 adeno- or squamous cell NSCLC, of whom fresh-frozen lung resection biopsies and pretreatment FDG-positron emission tomography (PET) scans were available, were included in this study (n = 108). FDG-uptake was quantified by calculating total lesion glycolysis (TLG). Metabolic marker expression was measured by immunofluorescent staining (protein) and quantitative polymerase chain reaction (messenger ribonucleic acid [mRNA]). Patients were retrospectively evaluated for DFS.

RESULTS

mRNA and protein expression of metabolic markers, with the exception of MCT4, and TLG were higher in squamous cell carcinomas than in adenocarcinomas, whereas adenocarcinomas were better vascularized. Adenocarcinomas had a worse DFS compared with squamous cell carcinomas (p = 0.016) based on the potential to metastasize. High TLG was associated with a worse DFS only in adenocarcinomas.

CONCLUSION

Our findings suggest that the adenocarcinomas exhibit glycolysis under normoxic conditions, whereas squamous cell carcinomas are exposed to diffusion-limited hypoxia resulting in a very high anaerobic glycolytic rate. Although squamous cell carcinomas have a higher FDG-uptake, in general regarded as a poor prognostic factor, adenocarcinomas have a higher metastatic potential and a worse DFS. These findings show that FDG-PET should be interpreted in relation to histology. This may improve the prognostic potential of FDG-PET and may aid in exploiting FDG-PET in treatment strategies allied to histology.

A Segmentation Algorithm for Quantitative Analysis of Heterogeneous Tumors of the Cervix With ¹⁸F-FDG PET/CT

As positron-emission tomography (PET) images have low spatial resolution and much noise, accurate image segmentation is one of the most challenging issues in tumor quantification. Tumors of the uterine cervix present a particular challenge because of urine activity in the adjacent bladder. Here, we propose and validate an automatic segmentation method adapted to cervical tumors. Our proposed methodology combined the gradient field information of both the filtered PET image and the level set function into a level set framework by constructing a new evolution equation. Furthermore, we also constructed a new hyperimage to recognize a rough tumor region using the fuzzy c-means algorithm according to the tissue specificity as defined by both PET (uptake) and computed tomography (attenuation) to provide the initial zero level set, which could make the segmentation process fully automatic. The proposed method was verified based on simulation and clinical studies. For simulation studies, seven different phantoms, representing tumors with homogenous/heterogeneous-low/high uptake patterns and different volumes, were simulated with five different noise levels. Twenty-seven cervical cancer patients at different stages were enrolled for clinical evaluation of the method. Dice similarity coefficients (DSC) and Hausdorff distance (HD) were used to evaluate the accuracy of the segmentation method, while a Bland-Altman analysis of the mean standardized uptake value (SUVmean) and metabolic tumor volume (MTV) was used to evaluate the accuracy of the quantification. Using this method, the DSCs and HDs of the homogenous and heterogeneous phantoms under clinical noise level were 93.39 ±1.09% and 6.02 ±1.09 mm, 93.59 ±1.63% and 8.92 ±2.57 mm, respectively. The DSCs and HDs in patients measured 91.80 ±2.46% and 7.79 ±2.18 mm. Through Bland-Altman analysis, the SUVmean and the MTV using our method showed high correlation with the clinical gold standard. The results of both simulation and clinical studies demonstrated the accuracy, effectiveness, and robustness of the proposed method. Further assessment of the quantitative indices indicates the feasibility of this algorithm in accurate quantitative analysis of cervical tumors in clinical practice.

FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0

The purpose of these guidelines is to assist physicians in recommending, performing, interpreting and reporting the results of FDG PET/CT for oncological imaging of adult patients. PET is a quantitative imaging technique and therefore requires a common quality control (QC)/quality assurance (QA) procedure to maintain the accuracy and precision of quantitation. Repeatability and reproducibility are two essential requirements for any quantitative measurement and/or imaging biomarker. Repeatability relates to the uncertainty in obtaining the same result in the same patient when he or she is examined more than once on the same system. However, imaging biomarkers should also have adequate reproducibility, i.e. the ability to yield the same result in the same patient when that patient is examined on different systems and at different imaging sites. Adequate repeatability and reproducibility are essential for the clinical management of patients and the use of FDG PET/CT within multicentre trials. A common standardised imaging procedure will help promote the appropriate use of FDG PET/CT imaging and increase the value of publications and, therefore, their contribution to evidence-based medicine. Moreover, consistency in numerical values between platforms and institutes that acquire the data will potentially enhance the role of semiquantitative and quantitative image interpretation. Precision and accuracy are additionally important as FDG PET/CT is used to evaluate tumour response as well as for diagnosis, prognosis and staging. Therefore both the previous and these new guidelines specifically aim to achieve standardised uptake value harmonisation in multicentre settings.

Insight on automated lesion delineation methods for PET data

BACKGROUND

Defining tumour volume for treatment response and radiotherapy planning is challenging and prone to inter- and intra-observer variability. Various automated tumour delineation methods have been proposed in the literature, each having abilities and limitations. Therefore, there is a need to provide clinicians with practical information on delineation method selection.

METHODS

Six different automated positron emission tomography (PET) delineation methods were evaluated and compared using National Electrical Manufacturer Association image quality (NEMA IQ) phantom data and three in-house synthetic phantoms with clinically relevant lesion shapes including spheres with necrotic core and irregular shapes. The impact of different contrast ratios, emission counts, realisations and reconstruction algorithms on delineation performance was also studied using similarity index (SI) and percentage volume error (%VE) as performance measures.

RESULTS

With the NEMA IQ phantom, contrast thresholding (CT) performed best on average for all sphere sizes and parameter settings (SI = 0.83; %VE = 5.65% ± 24.34%). Adaptive thresholding at 40% (AT40) was the next best method and required no prior parameter tuning (SI = 0.78; %VE = 23.22% ± 70.83%). When using SUV harmonisation filtering prior to delineation (EQ.PET), AT40 remains the best method without prior parameter tuning (SI = 0.81; %VE = 11.39% ± 85.28%). For necrotic core spheres and irregular shapes of the synthetic phantoms, CT remained the best performing method (SI = 0.83; %VE = 26.31% ± 38.26% and SI = 0.62; %VE = 24.52% ± 46.89%, respectively). The second best method was fuzzy locally adaptive Bayesian (FLAB) (SI = 0.83; %VE = 29.51% ± 81.79%) for necrotic core sphere and AT40 (SI = 0.58; %VE = 25.11% ± 32.41%) for irregular shapes. When using EQ.PET prior to delineation, AT40 was the best performing method without prior parameter tuning for both necrotic core (SI = 0.83; %VE = 27.98% ± 59.58%) and complex shapes phantoms (SI = 0.61; %VE = 14.83% ± 49.39%).

CONCLUSIONS

CT and AT40/AT50 are recommended for all lesion sizes and contrasts. Overall, considering background uptake information improves PET delineation accuracy. Applying EQ.PET prior to delineation improves accuracy and reduces coefficient of variation (CV) across different reconstructions and acquisitions.

Increased evidence for the prognostic value of primary tumor asphericity in pretherapeutic FDG PET for risk stratification in patients with head and neck cancer

PURPOSE

In a previous study, we demonstrated the first evidence that the asphericity (ASP) of pretherapeutic FDG uptake in the primary tumor provides independent prognostic information in patients with head and neck cancer. The aim of this work was to confirm these results in an independent patient group examined at a different site.

METHODS

FDG-PET/CT was performed in 37 patients. The primary tumor was delineated by an automatic algorithm based on adaptive thresholding. For the resulting ROIs, the metabolically active part of the tumor (MTV), SUVmax, SUVmean, total lesion glycolysis (TLG) and ASP were computed. Univariate Cox regression with respect to progression free survival (PFS) and overall survival (OS) was performed. For survival analysis, patients were divided in groups of high and low risk according to the parameter cut-offs defined in our previous work. In a second step, the cut-offs were adjusted to the present data. Univariate and multivariate Cox regression was performed for the pooled data consisting of the current and the previously described patient group (N = 68). In multivariate Cox regression, clinically relevant parameters were included.

RESULTS

Univariate Cox regression using the previously published cut-off values revealed TLG (hazard ratio (HR) = 3) and ASP (HR = 3) as significant predictors for PFS. For OS MTV (HR = 2.7) and ASP (HR = 5.9) were significant predictors. Using the adjusted cutoffs MTV (HR = 2.9/3.3), TLG (HR = 3.1/3.3) and ASP (HR = 3.1/5.9) were prognostic for PFS/OS. In the pooled data, multivariate Cox regression revealed a significant prognostic value with respect to PFS/OS for MTV (HR = 2.3/2.1), SUVmax (HR = 2.1/2.5), TLG (HR = 3.5/3.6), and ASP (HR = 3.4/4.4).

CONCLUSIONS

Our results confirm the independent prognostic value of ASP of the pretherapeutic FDG uptake in the primary tumor in patients with head and neck cancer. Moreover, these results demonstrate that ASP can be determined unambiguously across different sites.

Adaptive threshold segmentation of pituitary adenomas from FDG PET images for radiosurgery

In this study we have attempted to optimize a PET based adaptive threshold seg- mentation method for delineating small tumors, particularly in a background of high tracer activity. The metabolic nature of pituitary adenomas and the constraints of MRI imaging in the postoperative setting to delineate these tumors during radio- surgical procedures motivated us to develop this method. Phantom experiments were done to establish a relationship between the threshold required for segmenting the PET images and the target size and the activity concentration within the target in relation to its background. The threshold was developed from multiple linear regression of the experimental data optimized for tumor sizes less than 4 cm3. We validated our method against the phantom target volumes with measured target to background ratios ranging from 1.6 to 14.58. The method was tested on ten retro- spective patients with residual growth hormone-secreting pituitary adenomas that underwent radiosurgery and compared against the volumes delineated by manual method. The predicted volumes against the true volume of the phantom inserts gave a correlation coefficient of 99% (p &lt; 0.01). In the ten retrospective patients, the automatically segmented tumor volumes against volumes manually delineated by the clinicians had a correlation of 94% (p &lt; 0.01). This adaptive threshold segmentation showed promising results in delineating tumor volumes in pituitary adenomas planned for stereotactic radiosurgery, particularly in the postoperative setting where MR and CT images may be associated with artifacts, provided opti- mization experiment is carried out. 

Positron Emission Tomography (PET) in Oncology

Since its introduction in the early nineties as a promising functional imaging technique in the management of neoplastic disorders, FDG-PET, and subsequently FDG-PET/CT, has become a cornerstone in several oncologic procedures such as tumor staging and restaging, treatment efficacy assessment during or after treatment end and radiotherapy planning. Moreover, the continuous technological progress of image generation and the introduction of sophisticated software to use PET scan as a biomarker paved the way to calculate new prognostic markers such as the metabolic tumor volume (MTV) and the total amount of tumor glycolysis (TLG). FDG-PET/CT proved more sensitive than contrast-enhanced CT scan in staging of several type of lymphoma or in detecting widespread tumor dissemination in several solid cancers, such as breast, lung, colon, ovary and head and neck carcinoma. As a consequence the stage of patients was upgraded, with a change of treatment in 10%-15% of them. One of the most evident advantages of FDG-PET was its ability to detect, very early during treatment, significant changes in glucose metabolism or even complete shutoff of the neoplastic cell metabolism as a surrogate of tumor chemosensitivity assessment. This could enable clinicians to detect much earlier the effectiveness of a given antineoplastic treatment, as compared to the traditional radiological detection of tumor shrinkage, which usually takes time and occurs much later.

Delineation gross tumor volume based on positron emission tomography images by a numerical approximation method

OBJECTIVE

A scheme, named SUV_Shape, for the gross tumor volume (GTV) delineation on positron emission tomography (PET) images was designed by a numerical approximation method, and evaluated during this study.

METHODS

Twenty-one vacuous plastic balls of different shapes and sizes, their volumes ranged from 0.56 to 179.50 mL and were confirmed by a BL610 balance (Sartorius, Canada), consisted of four group models. Every group model was filled with a specific activity [18F]-FDG solution (55.1, 38.2, 23.7, and 36.3 kBq/mL) represented tumor, and fixed at the bottom of a barrel which was filled with unlike [18F]-FDG solution (5.4, 6.8, 8.1, and 4.0 kBq/mL, correspondingly) represented the background. The PET data of them were acquired by two-dimensional and three-dimensional mode in a PET/CT scanner (Discovery ST8, GE Healthcare, USA). The volume of each ball was measured by SUV_Shape, and the BL610 balance, labeled as GTVs and GTVt, respectively. Five rabbits implanted VX2 squamous carcinomas were acquired by [18F]-FDG PET/CT. These rabbits were mercy killed within 24 h after PET/CT acquisition. VX2 tumors were surgically removed, and their volumes were measured by SUV_Shape, and caliper, labeled as GTVs and GTVt. The Spearman's ρ between GTVs and GTVt were done.

RESULTS

The tumor-background ratios in four groups of phantom models were 10.3, 5.6, 2.9, and 9.0, respectively. The relationship between GTVt and GTVs for phantom models was significant (Spearman's ρ > 0.95, P < 0.01), regardless of the different acquisition modes. Twelve VX2 tumor nodes or masses were measured; their GTVt ranged from 0.11 to 29.26 mL. The relationship between GTVt and GTVs was significant (Spearman's ρ = 0.893, P < 0.01) for animal tumor models.

CONCLUSIONS

The SUV_Shape scheme could delineate tumors based on their radiopharmaceutical-avid PET images.

Repeatability of metabolically active tumor volume measurements with FDG PET/CT in advanced gastrointestinal malignancies: a multicenter study

PURPOSE

To evaluate the feasibility and repeatability of various metabolically active tumor volume ( MATV metabolically active tumor volume ) quantification methods in fluorine 18 fluorodeoxyglucose ( FDG fluorine 18 fluorodeoxyglucose ) positron emission tomography (PET)/computed tomography (CT) in a multicenter setting and propose the optimal MATV metabolically active tumor volume method together with the minimal threshold for future response evaluation studies.

MATERIALS AND METHODS

The study was approved by the institutional review board of all four participating centers, and patients provided written informed consent. Thirty-four patients with advanced gastrointestinal malignancies underwent two FDG fluorine 18 fluorodeoxyglucose PET/CT examinations within 1 week. MATV metabolically active tumor volume s were defined semiautomatically with 27 variations of tumor delineation methods with different reference values. Feasibility was determined as the percentage of successful tumor segmentations per MATV metabolically active tumor volume method. Repeatability was determined with intraclass correlation coefficients, Bland-Altman plots, and limits of agreement ( LOA limit of agreement s) of the percentage difference between the test and repeat test measurements. In addition, LOA limit of agreement variability per center was investigated.

RESULTS

In total, 136 lesions were identified. Feasibility of tumor segmentation ranged from 54% to 100% (74-136 of 136 lesions); repeatability was evaluated for 19 MATV metabolically active tumor volume methods with feasibility of greater than 95%. The median MATV metabolically active tumor volume derived with 50% threshold of mean standardized uptake value ( SUV standardized uptake value ) of a sphere of 12-mm diameter with highest local intensity ( SUVhp mean SUV of a sphere of 12-mm diameter with highest local intensity ), which may not include the voxel with highest SUV standardized uptake value corrected for local background, was 5.7 and 6.1 mL for test and retest scans, respectively, with a relative LOA limit of agreement of 36.1%. Comparable repeatability was found between centers. A difference in uptake time between scan 1 and 2 of 15 minutes or longer had a minor negative influence on repeatability.

CONCLUSION

MATV metabolically active tumor volume measured with 50% of SUVhp mean SUV of a sphere of 12-mm diameter with highest local intensity corrected for local background is recommended in multicenter FDG fluorine 18 fluorodeoxyglucose PET/CT studies on the basis of a high feasibility (96%) and repeatability ( LOA limit of agreement of 36.1%).

A review on segmentation of positron emission tomography images

Positron Emission Tomography (PET), a non-invasive functional imaging method at the molecular level, images the distribution of biologically targeted radiotracers with high sensitivity. PET imaging provides detailed quantitative information about many diseases and is often used to evaluate inflammation, infection, and cancer by detecting emitted photons from a radiotracer localized to abnormal cells. In order to differentiate abnormal tissue from surrounding areas in PET images, image segmentation methods play a vital role; therefore, accurate image segmentation is often necessary for proper disease detection, diagnosis, treatment planning, and follow-ups. In this review paper, we present state-of-the-art PET image segmentation methods, as well as the recent advances in image segmentation techniques. In order to make this manuscript self-contained, we also briefly explain the fundamentals of PET imaging, the challenges of diagnostic PET image analysis, and the effects of these challenges on the segmentation results.

Influence of cold walls on PET image quantification and volume segmentation: a phantom study

PURPOSE

Commercially available fillable plastic inserts used in positron emission tomography phantoms usually have thick plastic walls, separating their content from the background activity. These "cold" walls can modify the intensity values of neighboring active regions due to the partial volume effect, resulting in errors in the estimation of standardized uptake values. Numerous papers suggest that this is an issue for phantom work simulating tumor tissue, quality control, and calibration work. This study aims to investigate the influence of the cold plastic wall thickness on the quantification of 18F-fluorodeoxyglucose on the image activity recovery and on the performance of advanced automatic segmentation algorithms for the delineation of active regions delimited by plastic walls.

METHODS

A commercial set of six spheres of different diameters was replicated using a manufacturing technique which achieves a reduction in plastic walls thickness of up to 90%, while keeping the same internal volume. Both sets of thin- and thick-wall inserts were imaged simultaneously in a custom phantom for six different tumor-to-background ratios. Intensity values were compared in terms of mean and maximum standardized uptake values (SUVs) in the spheres and mean SUV of the hottest 1 ml region (SUVmax, SUVmean, and SUVpeak). The recovery coefficient (RC) was also derived for each sphere. The results were compared against the values predicted by a theoretical model of the PET-intensity profiles for the same tumor-to-background ratios (TBRs), sphere sizes, and wall thicknesses. In addition, ten automatic segmentation methods, written in house, were applied to both thin- and thick-wall inserts. The contours obtained were compared to computed tomography derived gold standard ("ground truth"), using five different accuracy metrics.

RESULTS

The authors' results showed that thin-wall inserts achieved significantly higher SUVmean, SUVmax, and RC values (up to 25%, 16%, and 25% higher, respectively) compared to thick-wall inserts, which was in agreement with the theory. This effect decreased with increasing sphere size and TBR, and resulted in substantial (>5%) differences between thin- and thick-wall inserts for spheres up to 30 mm diameter and TBR up to 4. Thinner plastic walls were also shown to significantly improve the delineation accuracy for the majority of the segmentation methods tested, by increasing the proportion of lesion voxels detected, although the errors in image quantification remained non-negligible.

CONCLUSIONS

This study quantified the significant effect of a 90% reduction in the thickness of insert walls on SUV quantification and PET-based boundary detection. Mean SUVs inside the inserts and recovery coefficients were particularly affected by the presence of thick cold walls, as predicted by a theoretical approach. The accuracy of some delineation algorithms was also significantly improved by the introduction of thin wall inserts instead of thick wall inserts. This study demonstrates the risk of errors deriving from the use of cold wall inserts to assess and compare the performance of PET segmentation methods.

An automatic method for accurate volume delineation of heterogeneous tumors in PET

PURPOSE

Accurate volumetric tumor delineation is of increasing importance in radiation treatment planning. Many tumors exhibit only moderate tracer uptake heterogeneity and delineation methods using an adaptive threshold lead to robust results. These methods use a tumor reference value R (e.g., ROI maximum) and the tumor background Bg to compute the volume reproducing threshold. This threshold corresponds to an isocontour which defines the tumor boundary. However, the boundaries of strongly heterogeneous tumors can not be described by an isocontour anymore and therefore conventional threshold methods are not suitable for accurate delineation. The aim of this work is the development and validation of a delineation method for heterogeneous tumors.

METHODS

The new method (voxel-specific threshold method, VTM) can be considered as an extension of an adaptive threshold method (lesion-specific threshold method, LTM), where instead of a lesion-specific threshold for the whole ROI, a voxel-specific threshold is computed by determining for each voxel Bg and R in the close vicinity of the voxel. The absolute threshold for the considered voxel is then given by Tabs=T×(R-Bg)+Bg, where T=0.39 was determined with phantom measurements.

VALIDATION

30 clinical datasets from patients with non-small-cell lung cancer were used to generate 30 realistic anthropomorphic software phantoms of tumors with different heterogeneities and well-known volumes and boundaries. Volume delineation was performed with VTM and LTM and compared with the known lesion volumes and boundaries.

RESULTS

In contrast to LTM, VTM was able to reproduce the true tumor boundaries accurately, independent of the heterogeneity. The deviation of the determined volume from the true volume was (0.8±4.2)% for VTM and (11.0±16.4)% for LTM.

CONCLUSIONS

In anthropomorphic software phantoms, the new method leads to promising results and to a clear improvement of volume delineation in comparison to conventional background-corrected thresholding. In the next step, the suitability for clinical routine will be further investigated.

Contourlet-based active contour model for PET image segmentation

PURPOSE

PET-guided radiation therapy treatment planning, clinical diagnosis, assessment of tumor growth, and therapy response rely on the accurate delineation of the tumor volume and quantification of tracer uptake. Most PET image segmentation techniques proposed thus far are suboptimal in the presence of heterogeneity of tracer uptake within the lesion. This work presents an active contour model approach based on the method of Chan and Vese ["Active contours without edges," IEEE Trans. Image Process. 10, 266-277 (2001)] designed to take into account the high level of statistical uncertainty (noise) and to handle the heterogeneity of tumor uptake typically present in PET images.

METHODS

In the proposed method, the fitting terms in the Chan-Vese formulation are modified by introducing new input images, including the smoothed version of the original image using anisotropic diffusion filtering (ADF) and the contourlet transform of the image. The advantage of utilizing ADF for image smoothing is that it avoids blurring the object's edges and preserves the average activity within a region, which is important for accurate PET quantification. Moreover, incorporating the contourlet transform of the image into the fitting terms makes the energy functional more effective in directing the evolving curve toward the object boundaries due to the enhancement of the tumor-to-background ratio (TBR). The proper choice of the energy functional parameters has been formulated by making a clear consensus based on tumor heterogeneity and TBR levels. This cautious parameter selection leads to proper handling of heterogeneous lesions. The algorithm was evaluated using simulated phantom and clinical studies, where the ground truth and histology, respectively, were available for accurate quantitative analysis of the segmentation results. The proposed technique was also compared to a number of previously reported image segmentation techniques.

RESULTS

The results were quantitatively analyzed using three evaluation metrics, including the spatial overlap index (SOI), the mean relative error (MRE), and the mean classification error (MCE). Although the performance of the proposed method was analogous to other methods for some datasets, overall the proposed algorithm outperforms all other techniques. In the largest clinical group comprising nine datasets, the proposed approach improved the SOI from 0.41±0.14 obtained using the best-performing algorithm to 0.54±0.12 and reduced the MRE from 54.23±103.29 to 0.19±16.63 and the MCE from 112.86±69.07 to 60.58±18.43.

CONCLUSIONS

The proposed segmentation technique is superior to other representative segmentation techniques in terms of highest overlap between the segmented volume and the ground truth∕histology and minimum relative and classification errors. Therefore, the proposed active contour model can result in more accurate tumor volume delineation from PET images.

Use of wall-less ¹⁸F-doped gelatin phantoms for improved volume delineation and quantification in PET/CT

Positron emission tomography (PET) with (18)F-FDG is a valuable tool for staging, planning treatment, and evaluating the treatment response for many different types of tumours. The correct volume estimation is of utmost importance in these situations. To date, the most common types of phantoms used in volume quantification in PET utilize fillable, hollow spheres placed in a circular or elliptical cylinder made of polymethyl methacrylate. However, the presence of a non-radioactive sphere wall between the hotspot and the background activity in images of this type of phantom could cause inaccuracies. To investigate the influence of the non-active walls, we developed a phantom without non-active sphere walls for volume delineation and quantification in PET. Three sizes of gelatin hotspots were moulded and placed in a Jaszczak phantom together with hollow plastic spheres of the same sizes containing the same activity concentration. (18)F PET measurements were made with zero background activity and with tumour-to-background ratios of 12.5, 10, 7.5, and 5. The background-corrected volume reproducing threshold, Tvol, was calculated for both the gelatin and the plastic spheres. It was experimentally verified that the apparent background dependence of Tvol, i.e., a decreasing Tvol with increasing background fraction, was not present for wall-less spheres; the opposite results were seen in plastic, hollow spheres in commercially-available phantoms. For the types of phantoms commonly used in activity quantification, the estimation of Tvol using fillable, hollow, plastic spheres with non-active walls would lead to an overestimate of the tumour volume, especially for small volumes in a high activity background.

Semiautomatic methods for segmentation of the proliferative tumour volume on sequential FLT PET/CT images in head and neck carcinomas and their relation to clinical outcome

PURPOSE

Radiotherapy of head and neck cancer induces changes in tumour cell proliferation during treatment, which can be depicted by the PET tracer (18)F-fluorothymidine (FLT). In this study, three advanced semiautomatic PET segmentation methods for delineation of the proliferative tumour volume (PV) before and during (chemo)radiotherapy were compared and related to clinical outcome.

METHODS

The study group comprised 46 patients with 48 squamous cell carcinomas of the head and neck, treated with accelerated (chemo)radiotherapy, who underwent FLT PET/CT prior to treatment and in the 2nd and 4th week of therapy. Primary gross tumour volumes were visually delineated on CT images (GTV CT). PVs were visually determined on all PET scans (PV VIS). The following semiautomatic segmentation methods were applied to sequential PET scans: background-subtracted relative-threshold level (PV RTL), a gradient-based method using the watershed transform algorithm and hierarchical clustering analysis (PV W&C), and a fuzzy locally adaptive Bayesian algorithm (PV FLAB).

RESULTS

Pretreatment PV VIS correlated best with PV FLAB and GTV CT. Correlations with PV RTL and PV W&C were weaker although statistically significant. During treatment, the PV VIS, PV W&C and PV FLAB significant decreased over time with the steepest decline over time for PV FLAB. Among these advanced segmentation methods, PV FLAB was the most robust in segmenting volumes in the third scan (67 % of tumours as compared to 40 % for PV W&C and 27 % for PV RTL). A decrease in PV FLAB above the median between the pretreatment scan and the scan obtained in the 4th week was associated with better disease-free survival (4 years 90 % versus 53 %).

CONCLUSION

In patients with head and neck cancer, FLAB proved to be the best performing method for segmentation of the PV on repeat FLT PET/CT scans during (chemo)radiotherapy. This may potentially facilitate radiation dose adaptation to changing PV.

Functional imaging in radiation therapy planning for head and neck cancer

Functional imaging and its application to radiotherapy (RT) is a rapidly expanding field with new modalities and techniques constantly developing and evolving. As technologies improve, it will be important to pay attention to their implementation. This review describes the main achievements in the field of head and neck cancer (HNC) with particular remarks on the unsolved problems.

Molecular PET imaging for biology-guided adaptive radiotherapy of head and neck cancer

Integration of molecular imaging PET techniques into therapy selection strategies and radiation treatment planning for head and neck squamous cell carcinoma (HNSCC) can serve several purposes. First, pre-treatment assessments can steer decisions about radiotherapy modifications or combinations with other modalities. Second, biology-based objective functions can be introduced to the radiation treatment planning process by co-registration of molecular imaging with planning computed tomography (CT) scans. Thus, customized heterogeneous dose distributions can be generated with escalated doses to tumor areas where radiotherapy resistance mechanisms are most prevalent. Third, monitoring of temporal and spatial variations in these radiotherapy resistance mechanisms early during the course of treatment can discriminate responders from non-responders. With such information available shortly after the start of treatment, modifications can be implemented or the radiation treatment plan can be adapted tailing the biological response pattern. Currently, these strategies are in various phases of clinical testing, mostly in single-center studies. Further validation in multicenter set-up is needed. Ultimately, this should result in availability for routine clinical practice requiring stable production and accessibility of tracers, reproducibility and standardization of imaging and analysis methods, as well as general availability of knowledge and expertise. Small studies employing adaptive radiotherapy based on functional dynamics and early response mechanisms demonstrate promising results. In this context, we focus this review on the widely used PET tracer (18)F-FDG and PET tracers depicting hypoxia and proliferation; two well-known radiation resistance mechanisms.

[Validation of segmentation techniques for positron emission tomography using ex-vivo images of oncological surgical specimens]

OBJECTIVE

To design a novel ex-vivo acquisition technique to establish a common framework to validate different segmentation techniques for oncological PET images. To evaluate several automatic segmentation algorithms on this set of images.

MATERIAL AND METHODS

In 15 patients with cancer, ex-vivo PET studies of surgical specimens removed during surgery were performed after injection of (18)F-FDG. Images were acquired in two scanners: a clinical PET/CT and a high-resolution PET scanner. Real tumor volume was determined in each patient, and a reference image was generated for segmentation of each tumor. Images were segmented with 12 automatic algorithms and with a standard method for PET (relative threshold at 42%) and results were evaluated by quantitative parameters.

RESULTS

It has been possible to demonstrate by segmentation of PET images of surgical specimens that on high resolution PET images, 8 out of 12 evaluated segmentation techniques outperformed the standard method, whose value is 42%. However, none of the algorithms outperformed the standard method when applied on images from the clinical PET/CT. Due to the great interest of this set of PET images, all studies have been published on the Internet in order to provide a common framework for validation and comparison of different segmentation techniques.

CONCLUSIONS

We have proposed a novel technique to validate segmentation techniques for oncological PET images, acquiring ex-vivo PET studies of surgical specimens. We have demonstrated the usefulness of this set of PET images by evaluating several automatic segmentation algorithms.

Pathology-based validation of FDG PET segmentation tools for volume assessment of lymph node metastases from head and neck cancer

PURPOSE

FDG PET is increasingly incorporated into radiation treatment planning of head and neck cancer. However, there are only limited data on the accuracy of radiotherapy target volume delineation by FDG PET. The purpose of this study was to validate FDG PET segmentation tools for volume assessment of lymph node metastases from head and neck cancer against the pathological method as the standard.

METHODS

Twelve patients with head and neck cancer and 28 metastatic lymph nodes eligible for therapeutic neck dissection underwent preoperative FDG PET/CT. The metastatic lymph nodes were delineated on CT (NodeCT) and ten PET segmentation tools were used to assess FDG PET-based nodal volumes: interpreting FDG PET visually (PETVIS), applying an isocontour at a standardized uptake value (SUV) of 2.5 (PETSUV), two segmentation tools with a fixed threshold of 40% and 50%, and two adaptive threshold based methods. The latter four tools were applied with the primary tumour as reference and also with the lymph node itself as reference. Nodal volumes were compared with the true volume as determined by pathological examination.

RESULTS

Both NodeCT and PETVIS showed good correlations with the pathological volume. PET segmentation tools using the metastatic node as reference all performed well but not better than PETVIS. The tools using the primary tumour as reference correlated poorly with pathology. PETSUV was unsatisfactory in 35% of the patients due to merging of the contours of adjacent nodes.

CONCLUSION

FDG PET accurately estimates metastatic lymph node volume, but beyond the detection of lymph node metastases (staging), it has no added value over CT alone for the delineation of routine radiotherapy target volumes. If FDG PET is used in radiotherapy planning, treatment adaptation or response assessment, we recommend an automated segmentation method for purposes of reproducibility and interinstitutional comparison.

Unsupervised tumour segmentation in PET using local and global intensity-fitting active surface and alpha matting

This paper proposes an unsupervised tumour segmentation approach for PET data. The method computes the volumes of interest (VOIs) with sub-voxel precision by considering the limited image resolution and partial volume effects. First, an improved anisotropic diffusion filter is used to remove image noise. A hierarchical local and global intensity active surface modelling scheme is then applied to segment VOIs, followed by an alpha matting step to further refine the segmentation boundary. The proposed method is validated on real PET images of head-and-neck cancer patients with ground truth provided by human experts, as well as custom-designed phantom PET images with objective ground truth. Experimental results show that our method outperforms previous automatic approaches in terms of segmentation accuracy.

The accuracy and reproducibility of SPECT target volumes and activities estimated using an iterative adaptive thresholding technique

OBJECTIVE

Our aim was to design a practical and reproducible image segmentation method for calculations of total absorbed doses in organs and tumours for internally delivered radioisotopes. We have built upon our previously proposed use of two separate thresholds and employed an iterative technique for semiautomatic selection of background regions for segmenting an object of interest using thresholds that depend on the source-to-background ratio of activity concentrations.

METHODS

The parameters of curves relating volume and activity thresholds to source-to-background ratio were established using phantoms with 20 different inserts. The accuracy of our technique was validated using a second phantom experiment, whereas the reproducibility of volume, activity and dose estimates of organs and tumours was investigated using 13 patient studies. The accuracy and reproducibility of segmentations achieved were assessed using images reconstructed with three different methods that ranged from a standard clinical reconstruction to an advanced quantitative reconstruction approach.

RESULTS

In the validation phantom experiment, bottle volumes and activities measured using iterative adaptive thresholding agreed on average with the true values to within 4%, regardless of the reconstruction method used. In the patient studies, volumes and activities estimated from the single-photon emission computed tomography images reconstructed with clinical software agreed with the volumes and activities estimated using the advanced reconstruction approach to within 6%, whereas the corresponding doses agreed to within 4%.

CONCLUSION

The proposed iterative adaptive thresholding technique can accurately determine object volume and activity, which allows standard clinical reconstructions to generate absorbed dose estimates that are similar to those values obtained using more advanced reconstruction methods.

Joint segmentation of anatomical and functional images: applications in quantification of lesions from PET, PET-CT, MRI-PET, and MRI-PET-CT images

We present a novel method for the joint segmentation of anatomical and functional images. Our proposed methodology unifies the domains of anatomical and functional images, represents them in a product lattice, and performs simultaneous delineation of regions based on random walk image segmentation. Furthermore, we also propose a simple yet effective object/background seed localization method to make the proposed segmentation process fully automatic. Our study uses PET, PET-CT, MRI-PET, and fused MRI-PET-CT scans (77 studies in all) from 56 patients who had various lesions in different body regions. We validated the effectiveness of the proposed method on different PET phantoms as well as on clinical images with respect to the ground truth segmentation provided by clinicians. Experimental results indicate that the presented method is superior to threshold and Bayesian methods commonly used in PET image segmentation, is more accurate and robust compared to the other PET-CT segmentation methods recently published in the literature, and also it is general in the sense of simultaneously segmenting multiple scans in real-time with high accuracy needed in routine clinical use.

PET-based primary tumor volumetric parameters and survival of patients with non-small cell lung carcinoma

OBJECTIVE

The purpose of the study was to assess metabolic tumor volume and total glycolytic activity of the primary tumor as prognostic parameters for outcome in patients with non-small cell lung carcinoma (NSCLC).

MATERIALS AND METHODS

Thirty-nine patients who had undergone a baseline staging PET/CT examination at our institution for the diagnosis of NSCLC were retrospectively identified. The maximum standardized uptake value (SUV(max)), metabolic tumor volume, and total glycolytic activity were segmented from PET using the gradient method; 12-month survival and overall survival at the end of follow-up were used as outcome measures. Multivariate logistic regression, receiver operating characteristic curve analysis, and Kaplan-Meier curves for survival analysis were generated and compared using the Mantel-Cox log-rank test.

RESULTS

The mean gradient-based metabolic tumor volume and gradient-based total glycolytic activity were significantly greater in the patients who died (93.3 mL and 597.5 g) than in those who survived (19.3 mL and 193.9 g, respectively) (p < 0.003 and p < 0.031). There was no statistically significant difference in the mean SUV(max) between the patients who survived (12.7) at 12 months and those who had died (13.1) (p = 0.85). On multivariate analysis, gradient-based metabolic tumor volume was the only variable associated with 12-month mortality when adjusted for all other factors.(.) The area under the curve (AUC) for gradient-based metabolic tumor volume was 0.77 (p < 0.006). A significant difference in the time to survival was observed between high and low gradient-based metabolic tumor volume (log-rank p < 0.05) cohorts using the median gradient-based metabolic tumor volume (9.7 mL) as the cut point.

CONCLUSION

PET-based volumetric imaging parameters are potential prognostic markers of outcome in patients with NSCLC.

Computer-aided diagnosis systems for lung cancer: challenges and methodologies

This paper overviews one of the most important, interesting, and challenging problems in oncology, the problem of lung cancer diagnosis. Developing an effective computer-aided diagnosis (CAD) system for lung cancer is of great clinical importance and can increase the patient's chance of survival. For this reason, CAD systems for lung cancer have been investigated in a huge number of research studies. A typical CAD system for lung cancer diagnosis is composed of four main processing steps: segmentation of the lung fields, detection of nodules inside the lung fields, segmentation of the detected nodules, and diagnosis of the nodules as benign or malignant. This paper overviews the current state-of-the-art techniques that have been developed to implement each of these CAD processing steps. For each technique, various aspects of technical issues, implemented methodologies, training and testing databases, and validation methods, as well as achieved performances, are described. In addition, the paper addresses several challenges that researchers face in each implementation step and outlines the strengths and drawbacks of the existing approaches for lung cancer CAD systems.

Prognostic significance of volume-based PET parameters in cancer patients

Accurate prediction of cancer prognosis before the start of treatment is important since these predictions often affect the choice of treatment. Prognosis is usually based on anatomical staging and other clinical factors. However, the conventional system is not sufficient to accurately and reliably determine prognosis. Metabolic parameters measured by ¹⁸F-fluorodeoxyglucose (FDG) positron emission tomography (PET) have the potential to provide valuable information regarding prognosis and treatment response evaluation in cancer patients. Among these parameters, volume-based PET parameters such as metabolic tumor volume and total lesion glycolysis are especially promising. However, the measurement of these parameters is significantly affected by the imaging methodology and specific image characteristics, and a standard method for these parameters has not been established. This review introduces volume-based PET parameters as potential prognostic indicators, and highlights methodological considerations for measurement, potential implications, and prospects for further studies.