Texture analysis of non-small cell lung cancer on unenhanced computed tomography: initial evidence for a relationship with tumour glucose metabolism and stage

Relative low glycometabolism may also occur in invasive lung adenocarcinoma with visceral pleural invasion: case report and comments

Lung cancer is the main cause of cancer deaths in the world and positron emission tomography (PET) is considered as the most accurate diagnosis and staging technique for lung cancer. For human cancers, fluorodeoxyglucose (FDG)-PET imaging of most primary and metastatic tumors will show significantly increased glucose uptake because high metabolic activity of cancer cells. But there still have the question of false negative or positive rates in diagnostic accuracy need to be considered.A 51 year old man was diagnosed a lung tumor in the right middle lobe without enlargement of lymph nodes by computed tomography (CT). The F-FDG-PET/CT presented 1 slight increased metabolism in the tumor region. After resection of the tumor, postoperative pathological examination confirmed that it was invasive lung adenocarcinoma and with visceral pleural invasion while showed relative low glucose absorption in PET/CT.A form of invasive lung adenocarcinoma was diagnosized. The tumor tissues were further confirmed by immunohistochemical assessments, which showed that thyroid transcription factor 1 (TTF-1 or NKX2-1) and Cytokeratin 7 (CK7) were all significant positive.Diagnosis of lung cancer even all other cancers by FDG-PET should be carefully considered the question of accuracy. Our case has added additional literature for us to considering the false-negative of lung cancer diagnosis by F-FDG-PET/CT.

Quantification of heterogeneity as a biomarker in tumor imaging: a systematic review

BACKGROUND

Many techniques are proposed for the quantification of tumor heterogeneity as an imaging biomarker for differentiation between tumor types, tumor grading, response monitoring and outcome prediction. However, in clinical practice these methods are barely used. This study evaluates the reported performance of the described methods and identifies barriers to their implementation in clinical practice.

METHODOLOGY

The Ovid, Embase, and Cochrane Central databases were searched up to 20 September 2013. Heterogeneity analysis methods were classified into four categories, i.e., non-spatial methods (NSM), spatial grey level methods (SGLM), fractal analysis (FA) methods, and filters and transforms (F&T). The performance of the different methods was compared.

PRINCIPAL FINDINGS

Of the 7351 potentially relevant publications, 209 were included. Of these studies, 58% reported the use of NSM, 49% SGLM, 10% FA, and 28% F&T. Differentiation between tumor types, tumor grading and/or outcome prediction was the goal in 87% of the studies. Overall, the reported area under the curve (AUC) ranged from 0.5 to 1 (median 0.87). No relation was found between the performance and the quantification methods used, or between the performance and the imaging modality. A negative correlation was found between the tumor-feature ratio and the AUC, which is presumably caused by overfitting in small datasets. Cross-validation was reported in 63% of the classification studies. Retrospective analyses were conducted in 57% of the studies without a clear description.

CONCLUSIONS

In a research setting, heterogeneity quantification methods can differentiate between tumor types, grade tumors, and predict outcome and monitor treatment effects. To translate these methods to clinical practice, more prospective studies are required that use external datasets for validation: these datasets should be made available to the community to facilitate the development of new and improved methods.

Glioma: application of whole-tumor texture analysis of diffusion-weighted imaging for the evaluation of tumor heterogeneity

BACKGROUND AND PURPOSE

To apply a texture analysis of apparent diffusion coefficient (ADC) maps to evaluate glioma heterogeneity, which was correlated with tumor grade.

MATERIALS AND METHODS

Forty patients with glioma (WHO grade II (n = 8), grade III (n = 10) and grade IV (n = 22)) underwent diffusion-weighted imaging (DWI), and the corresponding ADC maps were obtained. Regions of interest containing the lesions were drawn on every section of the ADC map containing the tumor, and volume-based data of the entire tumor were constructed. Texture and first order features including entropy, skewness and kurtosis were derived from the ADC map using in-house software. A histogram analysis of the ADC map was also performed. The texture and histogram parameters were compared between low-grade and high-grade gliomas using an unpaired student's t-test. Additionally, a one-way analysis of variance analysis with a post-hoc test was performed to compare the parameters of each grade.

RESULTS

Entropy was observed to be significantly higher in high-grade gliomas than low-grade tumors (6.861±0.539 vs. 6.261±0.412, P  = 0.006). The fifth percentiles of the ADC cumulative histogram also showed a significant difference between high and low grade gliomas (836±235 vs. 1030±185, P = 0.037). Only entropy proved to be significantly different between grades III and IV (6.295±0.4963 vs. 7.119±0.3165, P<0.001). The diagnostic accuracy of ADC entropy was significantly higher than that of the fifth percentile of the ADC histogram (P = 0.0034) in distinguishing high- from low-grade glioma.

CONCLUSION

A texture analysis of the ADC map based on the entire tumor volume can be useful for evaluating glioma grade, which provides tumor heterogeneity.

Prognostic value and reproducibility of pretreatment CT texture features in stage III non-small cell lung cancer

PURPOSE

To determine whether pretreatment CT texture features can improve patient risk stratification beyond conventional prognostic factors (CPFs) in stage III non-small cell lung cancer (NSCLC).

METHODS AND MATERIALS

We retrospectively reviewed 91 cases with stage III NSCLC treated with definitive chemoradiation therapy. All patients underwent pretreatment diagnostic contrast enhanced computed tomography (CE-CT) followed by 4-dimensional CT (4D-CT) for treatment simulation. We used the average-CT and expiratory (T50-CT) images from the 4D-CT along with the CE-CT for texture extraction. Histogram, gradient, co-occurrence, gray tone difference, and filtration-based techniques were used for texture feature extraction. Penalized Cox regression implementing cross-validation was used for covariate selection and modeling. Models incorporating texture features from the 33 image types and CPFs were compared to those with models incorporating CPFs alone for overall survival (OS), local-regional control (LRC), and freedom from distant metastases (FFDM). Predictive Kaplan-Meier curves were generated using leave-one-out cross-validation. Patients were stratified based on whether their predicted outcome was above or below the median. Reproducibility of texture features was evaluated using test-retest scans from independent patients and quantified using concordance correlation coefficients (CCC). We compared models incorporating the reproducibility seen on test-retest scans to our original models and determined the classification reproducibility.

RESULTS

Models incorporating both texture features and CPFs demonstrated a significant improvement in risk stratification compared to models using CPFs alone for OS (P=.046), LRC (P=.01), and FFDM (P=.005). The average CCCs were 0.89, 0.91, and 0.67 for texture features extracted from the average-CT, T50-CT, and CE-CT, respectively. Incorporating reproducibility within our models yielded 80.4% (±3.7% SD), 78.3% (±4.0% SD), and 78.8% (±3.9% SD) classification reproducibility in terms of OS, LRC, and FFDM, respectively.

CONCLUSIONS

Pretreatment tumor texture may provide prognostic information beyond that obtained from CPFs. Models incorporating feature reproducibility achieved classification rates of ∼80%. External validation would be required to establish texture as a prognostic factor.

Quantitative CT analysis of pulmonary ground-glass opacity nodules for the distinction of invasive adenocarcinoma from pre-invasive or minimally invasive adenocarcinoma

Objectives

We aimed to analyze the CT findings of ground-glass opacity nodules diagnosed pathologically as adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma in order to investigate whether quantitative CT parameters enable distinction of invasive adenocarcinoma from pre-invasive or minimally invasive adenocarcinoma.

Methods

We reviewed CT images and pathologic specimens from 191 resected ground-glass opacity nodules with little or no solid component at CT. Nodule size, volume, density, mass, skewness/kurtosis, and CT attenuation values at the 2.5^th–97.5^th percentiles on histogram, and texture parameters (uniformity and entropy) were assessed from CT datasets.

Results

Of 191 tumors, 38 were AISs (20%), 61 were MIAs (32%), and 92 (48%) were invasive adenocarcinomas. Multivariate logistic regression analysis helped identify the 75^th percentile CT attenuation value (P = 0.04) and entropy (P<0.01) as independent predictors for invasive adenocarcinoma, with an area under the receiver operating characteristic curve of 0.780.

Conclusion

Quantitative analysis of preoperative CT imaging metrics can help distinguish invasive adenocarcinoma from pre-invasive or minimally invasive adenocarcinoma.

High quality machine-robust image features: identification in nonsmall cell lung cancer computed tomography images

PURPOSE

For nonsmall cell lung cancer (NSCLC) patients, quantitative image features extracted from computed tomography (CT) images can be used to improve tumor diagnosis, staging, and response assessment. For these findings to be clinically applied, image features need to have high intra and intermachine reproducibility. The objective of this study is to identify CT image features that are reproducible, nonredundant, and informative across multiple machines.

METHODS

Noncontrast-enhanced, test-retest CT image pairs were obtained from 56 NSCLC patients imaged on three CT machines from two institutions. Two machines ("M1" and "M2") used cine 4D-CT and one machine ("M3") used breath-hold helical 3D-CT. Gross tumor volumes (GTVs) were semiautonomously segmented then pruned by removing voxels with CT numbers less than a prescribed Hounsfield unit (HU) cutoff. Three hundred and twenty eight quantitative image features were extracted from each pruned GTV based on its geometry, intensity histogram, absolute gradient image, co-occurrence matrix, and run-length matrix. For each machine, features with concordance correlation coefficient values greater than 0.90 were considered reproducible. The Dice similarity coefficient (DSC) and the Jaccard index (JI) were used to quantify reproducible feature set agreement between machines. Multimachine reproducible feature sets were created by taking the intersection of individual machine reproducible feature sets. Redundant features were removed through hierarchical clustering based on the average correlation between features across multiple machines.

RESULTS

For all image types, GTV pruning was found to negatively affect reproducibility (reported results use no HU cutoff). The reproducible feature percentage was highest for average images (M1 = 90.5%, M2 = 94.5%, M1∩M2 = 86.3%), intermediate for end-exhale images (M1 = 75.0%, M2 = 71.0%, M1∩M2 = 52.1%), and lowest for breath-hold images (M3 = 61.0%). Between M1 and M2, the reproducible feature sets generated from end-exhale images were relatively machine-sensitive (DSC = 0.71, JI = 0.55), and the reproducible feature sets generated from average images were relatively machine-insensitive (DSC = 0.90, JI = 0.87). Histograms of feature pair correlation distances indicated that feature redundancy was machine-sensitive and image type sensitive. After hierarchical clustering, 38 features, 28 features, and 33 features were found to be reproducible and nonredundant for M1∩M2 (average images), M1∩M2 (end-exhale images), and M3, respectively. When blinded to the presence of test-retest images, hierarchical clustering showed that the selected features were informative by correctly pairing 55 out of 56 test-retest images using only their reproducible, nonredundant feature set values.

CONCLUSIONS

Image feature reproducibility and redundancy depended on both the CT machine and the CT image type. For each image type, the authors found a set of cross-machine reproducible, nonredundant, and informative image features that would be useful for future image-based models. Compared to end-exhale 4D-CT and breath-hold 3D-CT, average 4D-CT derived image features showed superior multimachine reproducibility and are the best candidates for clinical correlation.

The role of texture analysis in imaging as an outcome predictor and potential tool in radiotherapy treatment planning

Predicting a tumour's response to radiotherapy prior to the start of treatment could enhance clinical care management by enabling the personalization of treatment plans based on predicted outcome. In recent years, there has been accumulating evidence relating tumour texture to patient survival and response to treatment. Tumour texture could be measured from medical images that provide a non-invasive method of capturing intratumoural heterogeneity and hence could potentially enable a prior assessment of a patient's predicted response to treatment. In this article, work presented in the literature regarding texture analysis in radiotherapy in relation to survival and outcome is discussed. Challenges facing integrating texture analysis in radiotherapy planning are highlighted and recommendations for future directions in research are suggested.

Robust Radiomics feature quantification using semiautomatic volumetric segmentation

Due to advances in the acquisition and analysis of medical imaging, it is currently possible to quantify the tumor phenotype. The emerging field of Radiomics addresses this issue by converting medical images into minable data by extracting a large number of quantitative imaging features. One of the main challenges of Radiomics is tumor segmentation. Where manual delineation is time consuming and prone to inter-observer variability, it has been shown that semi-automated approaches are fast and reduce inter-observer variability. In this study, a semiautomatic region growing volumetric segmentation algorithm, implemented in the free and publicly available 3D-Slicer platform, was investigated in terms of its robustness for quantitative imaging feature extraction. Fifty-six 3D-radiomic features, quantifying phenotypic differences based on tumor intensity, shape and texture, were extracted from the computed tomography images of twenty lung cancer patients. These radiomic features were derived from the 3D-tumor volumes defined by three independent observers twice using 3D-Slicer, and compared to manual slice-by-slice delineations of five independent physicians in terms of intra-class correlation coefficient (ICC) and feature range. Radiomic features extracted from 3D-Slicer segmentations had significantly higher reproducibility (ICC = 0.85±0.15, p = 0.0009) compared to the features extracted from the manual segmentations (ICC = 0.77±0.17). Furthermore, we found that features extracted from 3D-Slicer segmentations were more robust, as the range was significantly smaller across observers (p = 3.819e-07), and overlapping with the feature ranges extracted from manual contouring (boundary lower: p = 0.007, higher: p = 5.863e-06). Our results show that 3D-Slicer segmented tumor volumes provide a better alternative to the manual delineation for feature quantification, as they yield more reproducible imaging descriptors. Therefore, 3D-Slicer can be employed for quantitative image feature extraction and image data mining research in large patient cohorts.

Noninvasive image texture analysis differentiates K-ras mutation from pan-wildtype NSCLC and is prognostic

BACKGROUND

Non-invasive characterization of a tumor's molecular features could enhance treatment management. Quantitative computed tomography (CT) based texture analysis (QTA) has been used to derive tumor heterogeneity information, and the appearance of the tumors has been shown to relate to patient outcome in non-small cell lung cancer (NSCLC) and other cancers. In this study, we examined the potential of tumoral QTA to differentiate K-ras mutant from pan-wildtype tumors and its prognostic potential using baseline pre-treatment non-contrast CT imaging in NSCLC.

METHODS

Tumor DNA from patients with early-stage NSCLC was analyzed on the LungCarta Panel. Cases with a K-ras mutation or pan-wildtype for 26 oncogenes and tumor suppressor genes were selected for QTA. QTA was applied to regions of interest in the primary tumor. Non-parametric Mann Whitney test assessed the ability of the QTA, clinical and patient characteristics to differentiate between K-ras mutation from pan-wildtype. A recursive decision tree was developed to determine whether the differentiation of K-ras mutant from pan-wildtype tumors could be improved by sequential application of QTA parameters. Kaplan-Meier survival analysis assessed the ability of these markers to predict survival.

RESULTS

QTA was applied to 48 cases identified, 27 had a K-ras mutation and 21 cases were pan-wildtype. Positive skewness and lower kurtosis were significantly associated with the presence of a K-ras mutation. A five node decision tree had sensitivity, specificity, and accuracy values (95% CI) of 96.3% (78.1-100), 81.0% (50.5-97.4), and 89.6% (72.9-97.0); respectively. Kurtosis was a significant predictor of OS and DFS, with a lower kurtosis value linked with poorer survival.

CONCLUSIONS

Lower kurtosis and positive skewness are significantly associated with K-ras mutations. A QTA feature such as kurtosis is prognostic for OS and DFS. Non-invasive QTA can differentiate the presence of K-ras mutation from pan-wildtype NSCLC and is associated with patient survival.

The use of molecular imaging combined with genomic techniques to understand the heterogeneity in cancer metastasis

Tumour heterogeneity has, in recent times, come to play a vital role in how we understand and treat cancers; however, the clinical translation of this has lagged behind advances in research. Although significant advancements in oncological management have been made, personalized care remains an elusive goal. Inter- and intratumour heterogeneity, particularly in the clinical setting, has been difficult to quantify and therefore to treat. The histological quantification of heterogeneity of tumours can be a logistical and clinical challenge. The ability to examine not just the whole tumour but also all the molecular variations of metastatic disease in a patient is obviously difficult with current histological techniques. Advances in imaging techniques and novel applications, alongside our understanding of tumour heterogeneity, have opened up a plethora of non-invasive biomarker potential to examine tumours, their heterogeneity and the clinical translation. This review will focus on how various imaging methods that allow for quantification of metastatic tumour heterogeneity, along with the potential of developing imaging, integrated with other in vitro diagnostic approaches such as genomics and exosome analyses, have the potential role as a non-invasive biomarker for guiding the treatment algorithm.

Multifunctional imaging signature for V-KI-RAS2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutations in colorectal cancer

This study explores the potential for multifunctional imaging to provide a signature for V-KI-RAS2 Kirsten rat sarcoma viral oncogene homolog (KRAS) gene mutations in colorectal cancer.

METHODS

This prospective study approved by the institutional review board comprised 33 patients undergoing PET/CT before surgery for proven primary colorectal cancer. Tumor tissue was examined histologically for presence of the KRAS mutations and for expression of hypoxia-inducible factor-1 (HIF-1) and minichromosome maintenance protein 2 (mcm2). The following imaging parameters were derived for each tumor: (18)F-FDG uptake ((18)F-FDG maximum standardized uptake value [SUVmax]), CT texture (expressed as mean of positive pixels [MPP]), and blood flow measured by dynamic contrast-enhanced CT. A recursive decision tree was developed in which the imaging investigations were applied sequentially to identify tumors with KRAS mutations. Monte Carlo analysis provided mean values and 95% confidence intervals for sensitivity, specificity, and accuracy.

RESULTS

The final decision tree comprised 4 decision nodes and 5 terminal nodes, 2 of which identified KRAS mutants. The true-positive rate, false-positive rate, and accuracy (95% confidence intervals) of the decision tree were 82.4% (63.9%-93.9%), 0% (0%-10.4%), and 90.1% (79.2%-96.0%), respectively. KRAS mutants with high (18)F-FDG SUVmax and low MPP showed greater frequency of HIF-1 expression (P = 0.032). KRAS mutants with low (18)F-FDG SUV(max), high MPP, and high blood flow expressed mcm2 (P = 0.036).

CONCLUSION

Multifunctional imaging with PET/CT and recursive decision-tree analysis to combine measurements of tumor (18)F-FDG uptake, CT texture, and perfusion has the potential to identify imaging signatures for colorectal cancers with KRAS mutations exhibiting hypoxic or proliferative phenotypes.

Lung texture in serial thoracic CT scans: registration-based methods to compare anatomically matched regions

PURPOSE

The aim of this study was to compare three demons registration-based methods to identify spatially matched regions in serial computed tomography (CT) scans for use in texture analysis.

METHODS

Two thoracic CT scans containing no lung abnormalities and acquired during serial examinations separated by at least one week were retrospectively collected from 27 patients. Over 1000 regions of interest (ROIs) were randomly placed in the lungs of each baseline scan. Anatomically matched ROIs in the corresponding follow-up scan were placed by mapping the baseline scan ROI center pixel to (1) the original follow-up scan, (2) the follow-up scan resampled to match the baseline scan voxel size, and (3) the follow-up scan aligned to the baseline scan through affine registration. Mappings used the vector field obtained through demons deformable registration of each follow-up scan variant to the baseline scan. 140 texture features distributed among five feature classes were calculated in all ROIs. Feature value differences between paired ROIs were evaluated using Bland-Altman 95% limits of agreement. For each feature, (1) the mean feature value change and (2) the difference between the upper and lower limits of agreement were normalized to the mean feature value to obtain, respectively, the normalized bias and normalized range of agreement (nRoA). Nonparametric tests were used to evaluate differences in normalized bias and nRoA across the three methods.

RESULTS

Because patient CT scans contained no pathology, minimal changes in feature values were expected (i.e., low nRoA and normalized bias). Seventy-five features with very large feature value variability (nRoA ≥ 100%) were excluded from further analysis. Across the remaining 65 features, significant differences in normalized bias were observed among the three methods. The lowest normalized bias (median: 0.06%) was achieved when feature values were calculated on original follow-up scans. The affine registration method achieved the lowest nRoA, though nRoA was not significantly increased using original follow-up scans. Features with low nRoA values also had low normalized bias, though the converse was not necessarily true. Using nRoA as a metric, a set of 20 features having both low nRoA and normalized bias were identified.

CONCLUSIONS

Three methods to facilitate texture analysis of serial CT scans using demons registration for ROI placement were evaluated. The bias in feature value change between matched ROIs was minimized when feature values were calculated on original baseline and follow-up scans. A set of features that had both low bias and variability (nRoA) in feature value change using this method were identified. This texture analysis approach could facilitate future measurement of pathologic changes between CT scans without necessitating calculation of feature values on deformed scans.

Density features of screened lung tumors in low-dose computed tomography

RATIONALE AND OBJECTIVES

Using low-dose computed tomography (LDCT), small and heterogeneous lung tumors are detected in screening. The criteria for assessing detected tumors are crucial for determining follow-up or resection strategies. The purpose of this study was to investigate the capacity of density features in differentiating lung tumors.

MATERIALS AND METHODS

From July 2008 to December 2011, 48 surgically confirmed tumors (29 malignancies, comprising 17 cases of adenocarcinoma and 12 cases of adenocarcinoma in situ [AdIs], and 19 benignancies, comprising 11 cases of atypical adenomatous hyperplasia [AAH] and eight cases of benign non-AAH) in 38 patients were retrospectively evaluated, indicating that the positive predictive value (PPV) of physicians is 60.4% (29/48). Three types of density features, tumor disappearance rate (TDR), mean, and entropy, were obtained from the CT values of detected tumors.

RESULTS

Entropy is capable of differentiating malignancy from benignancy but is limited in differentiating AdIs from benign non-AAH. The combination of entropy and TDR is effective for predicting malignancy with an accuracy of 87.5% (42/48) and a PPV of 89.7% (26/29), improving the PPV of physicians by 29.3%. The combination of entropy and mean adequately clarifies the four pathology groups with an accuracy of 72.9% (35/48). For tumors with a mean below -400 Hounsfield units, the criterion of an entropy larger than 5.4 might be appropriate for diagnosing malignancy. For others, the pathology is either benign non-AAH or adenocarcinoma; adenocarcinoma has a higher entropy than benign non-AAH, with the exception of tuberculoma.

CONCLUSIONS

Combining density features enables differentiating heterogeneous lung tumors in LDCT.

Textural features of pretreatment 18F-FDG PET/CT images: prognostic significance in patients with advanced T-stage oropharyngeal squamous cell carcinoma

Previous studies have shown that total lesion glycolysis (TLG) may serve as a prognostic indicator in oropharyngeal squamous cell carcinoma (OPSCC). We sought to investigate whether the textural features of pretreatment (18)F-FDG PET/CT images can provide any additional prognostic information over TLG and clinical staging in patients with advanced T-stage OPSCC.

METHODS

We retrospectively analyzed the pretreatment (18)F-FDG PET/CT images of 70 patients with advanced T-stage OPSCC who had completed concurrent chemoradiotherapy, bioradiotherapy, or radiotherapy with curative intent. All of the patients had data on human papillomavirus (HPV) infection and were followed up for at least 24 mo or until death. A standardized uptake value (SUV) of 2.5 was taken as a cutoff for tumor boundary. The textural features of pretreatment (18)F-FDG PET/CT images were extracted from histogram analysis (SUV variance and SUV entropy), normalized gray-level cooccurrence matrix (uniformity, entropy, dissimilarity, contrast, homogeneity, inverse different moment, and correlation), and neighborhood gray-tone difference matrix (coarseness, contrast, busyness, complexity, and strength). Receiver-operating-characteristic curves were used to identify the optimal cutoff values for the textural features and TLG.

RESULTS

Thirteen patients were HPV-positive. Multivariate Cox regression analysis showed that age, tumor TLG, and uniformity were independently associated with progression-free survival (PFS) and disease-specific survival (DSS). TLG, uniformity, and HPV positivity were significantly associated with overall survival (OS). A prognostic scoring system based on TLG and uniformity was derived. Patients who presented with TLG > 121.9 g and uniformity ≤ 0.138 experienced significantly worse PFS, DSS, and OS rates than those without (P < 0.001, < 0.001, and 0.002, respectively). Patients with TLG > 121.9 g or uniformity ≤ 0.138 were further divided according to age, and different PFS and DSS were observed.

CONCLUSION

Uniformity extracted from the normalized gray-level cooccurrence matrix represents an independent prognostic predictor in patients with advanced T-stage OPSCC. A scoring system was developed and may serve as a risk-stratification strategy for guiding therapy.

Automated localization and segmentation of lung tumor from PET-CT thorax volumes based on image feature analysis

Positron emission tomography - computed tomography (PET-CT) plays an essential role in early tumor detection, diagnosis, staging and treatment. Automated and more accurate lung tumor detection and delineation from PET-CT is challenging. In this paper, on the basis of quantitative analysis of contrast feature of PET volume in SUV (standardized uptake value), our method firstly automatically localized the lung tumor. Then based on analysing the surrounding CT features of the initial tumor definition, our decision strategy determines the tumor segmentation from CT or from PET. The algorithm has been validated on 20 PET-CT studies involving non-small cell lung cancer (NSCLC). Experimental results demonstrated that our method was able to segment the tumor when adjacent to mediastinum or chest wall, and the algorithm outperformed the other five lung segmentation methods in terms of overlapping measure.

Tumor heterogeneity and permeability as measured on the CT component of PET/CT predict survival in patients with non-small cell lung cancer

PURPOSE

We prospectively examined the role of tumor textural heterogeneity on positron emission tomography/computed tomography (PET/CT) in predicting survival compared with other clinical and imaging parameters in patients with non-small cell lung cancer (NSCLC).

EXPERIMENTAL DESIGN

The feasibility study consisted of 56 assessed consecutive patients with NSCLC (32 males, 24 females; mean age 67 ± 9.7 years) who underwent combined fluorodeoxyglucose (FDG) PET/CT. The validation study population consisted of 66 prospectively recruited consecutive consenting patients with NSCLC (37 males, 29 females; mean age, 67.5 ± 7.8 years) who successfully underwent combined FDG PET/CT-dynamic contrast-enhanced (DCE) CT. Images were used to derive tumoral PET/CT textural heterogeneity, DCE CT permeability, and FDG uptake (SUVmax). The mean follow-up periods were 22.6 ± 13.3 months and 28.5± 13.2 months for the feasibility and validation studies, respectively. Optimum threshold was determined for clinical stage and each of the above biomarkers (where available) from the feasibility study population. Kaplan-Meier analysis was used to assess the ability of the biomarkers to predict survival in the validation study. Cox regression determined survival factor independence.

RESULTS

Univariate analysis revealed that tumor CT-derived heterogeneity (P < 0.001), PET-derived heterogeneity (P = 0.003), CT-derived permeability (P = 0.002), and stage (P < 0.001) were all significant survival predictors. The thresholds used in this study were derived from a previously conducted feasibility study. Tumor SUVmax did not predict survival. Using multivariable analysis, tumor CT textural heterogeneity (P = 0.021), stage (P = 0.001), and permeability (P < 0.001) were independent survival predictors. These predictors were independent of patient treatment.

CONCLUSIONS

Tumor stage and CT-derived textural heterogeneity were the best predictors of survival in NSCLC. The use of CT-derived textural heterogeneity should assist the management of many patients with NSCLC.

Quantifying tumour heterogeneity with CT

Heterogeneity is a key feature of malignancy associated with adverse tumour biology. Quantifying heterogeneity could provide a useful non-invasive imaging biomarker. Heterogeneity on computed tomography (CT) can be quantified using texture analysis which extracts spatial information from CT images (unenhanced, contrast-enhanced and derived images such as CT perfusion) that may not be perceptible to the naked eye. The main components of texture analysis can be categorized into image transformation and quantification. Image transformation filters the conventional image into its basic components (spatial, frequency, etc.) to produce derived subimages. Texture quantification techniques include structural-, model- (fractal dimensions), statistical- and frequency-based methods. The underlying tumour biology that CT texture analysis may reflect includes (but is not limited to) tumour hypoxia and angiogenesis. Emerging studies show that CT texture analysis has the potential to be a useful adjunct in clinical oncologic imaging, providing important information about tumour characterization, prognosis and treatment prediction and response.

Automatic Segmentation of Lung Carcinoma Using 3D Texture Features in 18-FDG PET/CT

Target definition is the largest source of geometric uncertainty in radiation therapy. This is partly due to a lack of contrast between tumor and healthy soft tissue for computed tomography (CT) and due to blurriness, lower spatial resolution, and lack of a truly quantitative unit for positron emission tomography (PET). First-, second-, and higher-order statistics, Tamura, and structural features were characterized for PET and CT images of lung carcinoma and organs of the thorax. A combined decision tree (DT) with K-nearest neighbours (KNN) classifiers as nodes containing combinations of 3 features were trained and used for segmentation of the gross tumor volume. This approach was validated for 31 patients from two separate institutions and scanners. The results were compared with thresholding approaches, the fuzzy clustering method, the 3-level fuzzy locally adaptive Bayesian algorithm, the multivalued level set algorithm, and a single KNN using Hounsfield units and standard uptake value. The results showed the DTKNN classifier had the highest sensitivity of 73.9%, second highest average Dice coefficient of 0.607, and a specificity of 99.2% for classifying voxels when using a probabilistic ground truth provided by simultaneous truth and performance level estimation using contours drawn by 3 trained physicians.

Measurements of heterogeneity in gliomas on computed tomography relationship to tumour grade

To undertake a preliminary study that uses CT texture analysis (CTTA) to quantify heterogeneity in gliomas on contrast-enhanced CT and to assess the relationship between tumour heterogeneity and grade. Retrospective analysis of contrast enhanced CT images was performed in 44 patients with histologically proven cerebral glioma between 2007 and 2010. 11 tumours were low grade (Grade I = 3; Grade II, = 8) and 33 high grade (Grade III = 10, Grade IV = 23). CTTA assessment of tumour heterogeneity was performed using a proprietary software algorithm (TexRAD) that selectively filters and extracts textures at different anatomical scales between filter values 1.0 (fine detail) and 2.5 (coarse features). Heterogeneity was quantified as standard deviation (SD) with or without filtration. Tumour heterogeneity, size and attenuation were correlated with tumour grade. For each parameter, receiver operating characteristics characterised the diagnostic performance for discrimination of high grade from low grade glioma and of grade III tumours from grade IV. Further the CTTA was compared to the radiological diagnosis. Tumour heterogeneity correlated significantly with grade (SD without filtration rs = 0.664, p < 0.001, SD with coarse filtration (rs = 0.714, p < 0.001). Tumour size and attenuation showed only moderate correlations with tumour grade (rs = 0.426, p = 0.004 and rs = 0.447, p = 0.002 respectively). Coarse texture was the best discriminator between high and low grade tumours (AUC 0.832, p < 0.0001) and between grade III and grade IV gliomas (AUC = 0.878 p = 0.0001). Compared to the radiological diagnosis, CTTA further characterised the indetermined cases. By quantifying tumour heterogeneity, CTTA has the potential to provide a marker of tumour grade for patients with cerebral glioma. By differentiating between high and low grade tumours, CTTA could possibly assist clinical management.

Are pretreatment 18F-FDG PET tumor textural features in non-small cell lung cancer associated with response and survival after chemoradiotherapy?

There is evidence in some solid tumors that textural features of tumoral uptake in (18)F-FDG PET images are associated with response to chemoradiotherapy and survival. We have investigated whether a similar relationship exists in non-small cell lung cancer (NSCLC).

METHODS

Fifty-three patients (mean age, 65.8 y; 31 men, 22 women) with NSCLC treated with chemoradiotherapy underwent pretreatment (18)F-FDG PET/CT scans. Response was assessed by CT Response Evaluation Criteria in Solid Tumors (RECIST) at 12 wk. Overall survival (OS), progression-free survival (PFS), and local PFS (LPFS) were recorded. Primary tumor texture was measured by the parameters coarseness, contrast, busyness, and complexity. The following parameters were also derived from the PET data: primary tumor standardized uptake values (SUVs) (mean SUV, maximum SUV, and peak SUV), metabolic tumor volume, and total lesion glycolysis.

RESULTS

Compared with nonresponders, RECIST responders showed lower coarseness (mean, 0.012 vs. 0.027; P = 0.004) and higher contrast (mean, 0.11 vs. 0.044; P = 0.002) and busyness (mean, 0.76 vs. 0.37; P = 0.027). Neither complexity nor any of the SUV parameters predicted RECIST response. By Kaplan-Meier analysis, OS, PFS, and LPFS were lower in patients with high primary tumor coarseness (median, 21.1 mo vs. not reached, P = 0.003; 12.6 vs. 25.8 mo, P = 0.002; and 12.9 vs. 20.5 mo, P = 0.016, respectively). Tumor coarseness was an independent predictor of OS on multivariable analysis. Contrast and busyness did not show significant associations with OS (P = 0.075 and 0.059, respectively), but PFS and LPFS were longer in patients with high levels of each (for contrast: median of 20.5 vs. 12.6 mo, P = 0.015, and median not reached vs. 24 mo, P = 0.02; and for busyness: median of 20.5 vs. 12.6 mo, P = 0.01, and median not reached vs. 24 mo, P = 0.006). Neither complexity nor any of the SUV parameters showed significant associations with the survival parameters.

CONCLUSION

In NSCLC, baseline (18)F-FDG PET scan uptake showing abnormal texture as measured by coarseness, contrast, and busyness is associated with nonresponse to chemoradiotherapy by RECIST and with poorer prognosis. Measurement of tumor metabolic heterogeneity with these parameters may provide indices that can be used to stratify patients in clinical trials for lung cancer chemoradiotherapy.

Assessment of tumor heterogeneity by CT texture analysis: can the largest cross-sectional area be used as an alternative to whole tumor analysis?

OBJECTIVE

To determine if there is a difference between contrast enhanced CT texture features from the largest cross-sectional area versus the whole tumor, and its effect on clinical outcome prediction.

METHODS

Entropy (E) and uniformity (U) were derived for different filter values (1.0-2.5: fine to coarse textures) for the largest primary tumor cross-sectional area and the whole tumor of the staging contrast enhanced CT in 55 patients with primary colorectal cancer. Parameters were compared using non-parametric Wilcoxon test. Kaplan-Meier analysis was performed to determine the relationship between CT texture and 5-year overall survival.

RESULTS

E was higher and U lower for the whole tumor indicating greater heterogeneity at all filter levels (1.0-2.5): median (range) for E and U for whole tumor versus largest cross-sectional area of 7.89 (7.43-8.31) versus 7.62 (6.94-8.08) and 0.005 (0.004-0.01) versus 0.006 (0.005-0.01) for filter 1.0; 7.88 (7.22-8.48) versus 7.54 (6.86-8.1) and 0.005 (0.003-0.01) versus 0.007 (0.004-0.01) for filter 1.5; 7.88 (7.17-8.54) versus 7.48 (5.84-8.25) and 0.005 (0.003-0.01) versus 0.007 (0.004-0.02) for filter 2.0; and 7.83 (7.03-8.57) versus 7.42 (5.19-8.26) and 0.005 (0.003-0.01) versus 0.006 (0.004-0.03) for filter 2.5 respectively (p ≤ 0.001). Kaplan-Meier analysis demonstrated better separation of E and U for whole tumor analysis for 5-year overall survival.

CONCLUSION

Whole tumor analysis appears more representative of tumor heterogeneity.

Non-small cell lung cancer: histopathologic correlates for texture parameters at CT

PURPOSE

To correlate computed tomographic (CT) texture in non-small cell lung cancer (NSCLC) with histopathologic markers for angiogenesis and hypoxia.

MATERIALS AND METHODS

The study was institutional review board approved, and informed consent was obtained. Fourteen patients with NSCLC underwent CT prior to intravenous administration of pimonidazole (0.5 g/m(2)), a marker of hypoxia, 24 hours before surgery. Texture was assessed for unenhanced and contrast material-enhanced CT images by using a software algorithm that selectively filters and extracts texture at different anatomic scales (1.0 [fine detail] to 2.5 [coarse features]), with quantification of the standard deviation (SD) of all pixel values and the mean value of positive pixels (MPP) and uniformity of distribution of positive gray-level pixel values (UPP). After surgery, matched tumor sections were stained for angiogenesis (CD34 expression) and for markers of hypoxia (glucose transporter protein 1 [Glut-1] and pimonidazole). The percentage and average intensity of the tumor stained were assessed. A linear mixed-effects model was used to assess the correlations between CT texture and staining intensity.

RESULTS

SD and MPP quantified from medium to coarse texture on contrast-enhanced CT images showed significant associations with the average intensity of tumor staining with pimonidazole (for SD: filter value, 2.5; slope = 0.003; P = .0003). UPP (medium to coarse texture) on unenhanced CT images showed a significant inverse association with tumor Glut-1 expression (filter value, 2.5; slope = -115.13; P = .0008). MPP quantified from medium to coarse texture on both unenhanced and contrast-enhanced CT images showed significant inverse associations with tumor CD34 expression (unenhanced CT: filter value, 1.8; slope = -0.0008; P = .003; contrast-enhanced CT: filter value, 1.8; slope = -0.0006; P = .004).

CONCLUSION

Texture parameters derived from CT images of NSCLC have the potential to act as imaging correlates for tumor hypoxia and angiogenesis.

SUPPLEMENTAL MATERIAL

http://radiology.rsna.org/lookup/suppl/doi:10.1148/radiol.12112428/-/DC1.

Assessment of primary colorectal cancer heterogeneity by using whole-tumor texture analysis: contrast-enhanced CT texture as a biomarker of 5-year survival

PURPOSE

To determine if computed tomographic (CT) texture features of primary colorectal cancer are related to 5-year overall survival rate.

MATERIALS AND METHODS

Institutional review board waiver was obtained for this retrospective analysis. Texture features of the entire primary tumor were assessed with contrast material-enhanced staging CT studies obtained in 57 patients as part of an ethically approved study and by using proprietary software. Entropy, uniformity, kurtosis, skewness, and standard deviation of the pixel distribution histogram were derived from CT images without filtration and with filter values corresponding to fine (1.0), medium (1.5, 2.0), and coarse (2.5) textures. Patients were followed up until death and were censored at 5 years if they were still alive. Kaplan-Meier analysis was performed to determine the relationship, if any, between CT texture and 5-year overall survival rate. The Cox proportional hazards model was used to assess independence of texture parameters from stage.

RESULTS

Follow-up data were available for 55 of 57 patients. There were eight stage I, 19 stage II, 17 stage III, and 11 stage IV cancers. Fine-texture feature Kaplan-Meier survival plots for entropy, uniformity, kurtosis, skewness, and standard deviation of the pixel distribution histogram were significantly different for tumors above and below each respective threshold receiver operating characteristic (ROC) curve optimal cutoff value (P = .001, P = .018, P = .032, P = .008, and P = .001, respectively), with poorer prognosis for ROC optimal values (a) less than 7.89 for entropy, (b) at least 0.01 for uniformity, (c) less than 2.48 for kurtosis, (d) at least -0.38 for skewness, and (e) less than 61.83 for standard deviation. Multivariate Cox proportional hazards regression analysis showed that each parameter was independent from the stage predictor of overall survival rate (P = .001, P = .009, P = .006, P = .02, and P = .001, respectively).

CONCLUSION

Fine-texture features are associated with poorer 5-year overall survival rate in patients with primary colorectal cancer.

SUPPLEMENTAL MATERIAL

http://radiology.rsna.org/lookup/suppl/doi:10.1148/radiol.12120254/-/DC1.

Assessment of tumor heterogeneity: an emerging imaging tool for clinical practice?

Background

Tumor spatial heterogeneity is an important prognostic factor, which may be reflected in medical images

Methods

Image texture analysis is an approach of quantifying heterogeneity that may not be appreciated by the naked eye. Different methods can be applied including statistical-, model-, and transform-based methods.

Results

Early evidence suggests that texture analysis has the potential to augment diagnosis and characterization as well as improve tumor staging and therapy response assessment in oncological practice.

Conclusion

This review provides an overview of the application of texture analysis with different imaging modalities, CT, MRI, and PET, to date and describes the technical challenges that have limited its widespread clinical implementation so far. With further efforts to refine its application, image texture analysis has the potential to develop into a valuable clinical tool for oncologic imaging.

Teaching Points

• Tumor spatial heterogeneity is an important prognostic factor.

• Image texture analysis is an approach of quantifying heterogeneity.

• Different methods can be applied, including statistical-, model-, and transform-based methods.

• Texture analysis could improve the diagnosis, tumor staging, and therapy response assessment.

PET imaging: implications for the future of therapy monitoring with PET/CT in oncology

Among the methods based on molecular imaging, the measure of the tracer uptake variation between a baseline and follow-up scan with the SUV and [(18)F]FDG-PET/CT is a very powerful tool for assessing response to treatment in oncology. However, the development of new targeted therapeutics and tissue pharmacokinetic evaluation of existing ones are increasingly requiring therapy monitoring with alternative tracers and indicators. In parallel, the potential predictive and prognostic value of other image-derived parameters, such as tumour volume and textural features, relating to tumoral heterogeneity, has recently emerged from several works.

Tumour heterogeneity in non-small cell lung carcinoma assessed by CT texture analysis: a potential marker of survival

PURPOSE

To establish the potential for tumour heterogeneity in non-small cell lung cancer (NSCLC) as assessed by CT texture analysis (CTTA) to provide an independent marker of survival for patients with NSCLC.

MATERIALS AND METHODS

Tumour heterogeneity was assessed by CTTA of unenhanced images of primary pulmonary lesions from 54 patients undergoing (18)F-fluorodeoxyglucose (FDG) PET-CT for staging of NSCLC. CTTA comprised image filtration to extract fine, medium and coarse features with quantification of the distribution of pixel values (uniformity) within the filtered images. Receiver operating characteristics identified thresholds for PET and CTTA parameters that were related to patient survival using Kaplan-Meier analysis.

RESULTS

The median (range) survival was 29.5 (1-38) months. 24, 10, 14 and 6 patients had tumour stages I, II, III and IV respectively. PET stage and tumour heterogeneity assessed by CTTA were significant independent predictors of survival (PET stage: Odds ratio 3.85, 95% confidence limits 0.9-8.09, P = 0.002; CTTA: Odds ratio 56.4, 95% confidence limits 4.79-666, p = 0.001). SUV was not a significantly associated with survival.

CONCLUSION

Assessment of tumour heterogeneity by CTTA of non-contrast enhanced images has the potential for to provide a novel, independent predictor of survival for patients with NSCLC.

KEY POINTS

Computed tomography is a routine staging procedure in non-small cell lung cancer. CT texture analysis (CTTA) can quantify heterogeneity within these lung tumours. CTTA seems to offer a novel independent predictor of survival for NSCLC. CTTA could contribute to disease risk-stratification for patients with NSCLC.

Heterogeneity of focal breast lesions and surrounding tissue assessed by mammographic texture analysis: preliminary evidence of an association with tumor invasion and estrogen receptor status

AIM

This pilot study investigates whether heterogeneity in focal breast lesions and surrounding tissue assessed on mammography is potentially related to cancer invasion and hormone receptor status.

MATERIALS AND METHODS

Texture analysis (TA) assessed the heterogeneity of focal lesions and their surrounding tissues in digitized mammograms from 11 patients randomly selected from an imaging archive [ductal carcinoma in situ (DCIS) only, n = 4; invasive carcinoma (IC) with DCIS, n = 3; IC only, n = 4]. TA utilized band-pass image filtration to highlight image features at different spatial frequencies (filter values: 1.0-2.5) from fine to coarse texture. The distribution of features in the derived images was quantified using uniformity.

RESULTS

Significant differences in uniformity were observed between patient groups for all filter values. With medium scale filtration (filter value = 1.5) pure DCIS was more uniform (median = 0.281) than either DCIS with IC (median = 0.246, p = 0.0102) or IC (median = 0.249, p = 0.0021). Lesions with high levels of estrogen receptor expression were more uniform, most notably with coarse filtration (filter values 2.0 and 2.5, r(s) = 0.812, p = 0.002). Comparison of uniformity values in focal lesions and surrounding tissue showed significant differences between DCIS with or without IC versus IC (p = 0.0009).

CONCLUSION

This pilot study shows the potential for computer-based assessments of heterogeneity within focal mammographic lesions and surrounding tissue to identify adverse pathological features in mammographic lesions. The technique warrants further investigation as a possible adjunct to existing computer aided diagnosis systems.

Imaging follow-up of RF ablation of lung tumours

Imaging is important in the decision-making process of how to treat a lung tumour, which ideally should be a multi-disciplinary team decision. Imaging is important during radiofrequency ablation (RFA) treatment with regard to optimal placement of the electrode, the immediate post-treatment criteria and very early detection of complications of the procedure. Imaging is very important in the treatment follow-up. In lung RFA, as in many other interventional procedures, the traditional morphological imaging techniques to evaluate treatment response have difficulties and functional imaging techniques may potentially be more useful. However, larger studies showing this impact have not yet been performed.

Tumour heterogeneity in oesophageal cancer assessed by CT texture analysis: preliminary evidence of an association with tumour metabolism, stage, and survival

AIM

To undertake a pilot study assessing whether tumour heterogeneity evaluated using computed tomography texture analysis (CTTA) has the potential to provide a marker of tumour aggression and prognosis in oesophageal cancer.

MATERIALS AND METHODS

In 21 patients, unenhanced CT images of the primary oesophageal lesion obtained using positron-emission tomography (PET)-CT examinations underwent CTTA. CTTA was carried out using a software algorithm that selectively filters and extracts textures at different anatomical scales between filter values 1.0 (fine detail) and 2.5 (coarse features) with quantification as entropy and uniformity (measures image heterogeneity). Texture parameters were correlated with average tumour 2-[(18)F]-fluoro-2-deoxy-d-glucose (FDG) uptake [standardized uptake values (SUV(mean) and SUV(max))] and clinical staging as determined by endoscopic ultrasound (nodal involvement) and PET-CT (distant metastases). The relationship between tumour stage, FDG uptake, and texture with survival was assessed using Kaplan-Meier analysis.

RESULTS

Tumour heterogeneity correlated with SUV(max) and SUV(mean). The closest correlations were found for SUV(mean) measured as uniformity and entropy with coarse filtration (r=-0.754, p<0.0001; and r=0.748, p=0.0001 respectively). Heterogeneity was also significantly greater in patients with clinical stage III or IV for filter values between 1.0 and 2.0 (maximum difference at filter value 1.5: entropy: p=0.027; uniformity p=0.032). The median (range) survival was 21 (4-34) months. Tumour heterogeneity assessed by CTTA (coarse uniformity) was an independent predictor of survival [odds ratio (OR)=4.45 (95% CI: 1.08, 18.37); p=0.039].

CONCLUSION

CTTA assessment of tumour heterogeneity has the potential to identify oesophageal cancers with adverse biological features and provide a prognostic indicator of survival.