Segmentation Of Tumor Region Research Articles

Performance analysis of brain tissues and tumor detection and grading system using ANFIS classifier

ABSTRACTAbnormal growth of cells in brain leads to the formation of tumors, which are categorized into benign and malignant. In this article, Co‐Active Adaptive Neuro Fuzzy Inference System (CANFIS) classification based brain tumor detection and its grading system is proposed. It has two phases as brain tumor segmentation and brain tissue segmentation. In brain tumor segmentation, CANFIS classifier is used to classify the test brain image into benign or malignant. Then, morphological operations are applied over the malignant image in order to segment the tumor regions in brain image. The K‐means classifier is used to classify the brain tissues into Grey Matter (GM), White Matter (WM) and Cerebro Spinal Fluid (CSF) regions as three different classes. Next, the segmented tumor is graded as mild, moderate or severe based on the presence of segmented tumor region in brain tissues.

Reproducibility and relative stability in magnetic resonance imaging indices of tumor vascular physiology over a period of 24 h in a rat 9L gliosarcoma model

PurposeThe objective was to study temporal changes in tumor vascular physiological indices in a period of 24h in a 9L gliosarcoma rat model. MethodsFischer-344 rats (N=14) were orthotopically implanted with 9L cells. At 2weeks post-implantation, they were imaged twice in a 24h interval using dynamic contrast enhanced magnetic resonance imaging (DCE-MRI). Data-driven model-selection-based analysis was used to segment tumor regions with varying vascular permeability characteristics. The region with the maximum number of estimable parameters of vascular kinetics was chosen for comparison across the two time points. It provided estimates of three parameters for an MR contrast agent (MRCA): i) plasma volume (vp), ii) forward volumetric transfer constant (Ktrans) and interstitial volume fraction (ve, ratio of Ktrans to reverse transfer constant, kep). In addition, MRCA extracellular distribution volume (VD) was estimated in the tumor and its borders, along with tumor blood flow (TBF) and peritumoral MRCA flux. Descriptors of parametric distributions were compared between the two times. Tumor extent was examined by hematoxylin and eosin (H&E) staining. Picrosirus red staining of secreted collagen was performed as an additional index for 9L cells. ResultsTest-retest differences between population summaries for any parameter were not significant (paired t and Wilcoxon signed rank tests). Bland-Altman plots showed no apparent trends between the differences and averages of the test-retest measures for all indices. The intraclass correlation coefficients showed moderate to almost perfect reproducibility for all of the parameters, except vp. H&E staining showed tumor infiltration in parenchyma, perivascular space and white matter tracts. Collagen staining was observed along the outer edges of main tumor mass. ConclusionThe data suggest the relative stability of these MR indices of tumor microenvironment over a 24h duration in this gliosarcoma model.

An efficient and automatic glioblastoma brain tumor detection using shift‐invariant shearlet transform and neural networks

AbstractThe detection and segmentation of tumor region in brain image is a critical task due to the similarity between abnormal and normal region. In this article, a computer‐aided automatic detection and segmentation of brain tumor is proposed. The proposed system consists of enhancement, transformation, feature extraction, and classification. The shift‐invariant shearlet transform (SIST) is used to enhance the brain image. Further, nonsubsampled contourlet transform (NSCT) is used as multiresolution transform which transforms the spatial domain enhanced image into multiresolution image. The texture features from grey level co‐occurrence matrix (GLCM), Gabor, and discrete wavelet transform (DWT) are extracted with the approximate subband of the NSCT transformed image. These extracted features are trained and classified into either normal or glioblastoma brain image using feed forward back propagation neural networks. Further, K‐means clustering algorithm is used to segment the tumor region in classified glioblastoma brain image. The proposed method achieves 89.7% of sensitivity, 99.9% of specificity, and 99.8% of accuracy.

Detection and diagnosis of meningioma brain tumor using ANFIS classifier

AbstractTumors are formed in brain due to the uncontrolled development of cells. These tumors can be cured if it is timely detected and by proper medication. This article proposes a computer‐aided automatic detection and diagnosis of meningioma brain tumors in brain images using Adaptive Neuro Fuzzy Inference System (ANFIS) classifier. The proposed system consists of feature extraction, classification, and segmentation and diagnosis sections. In this article, Grey level Co‐occurrence Matric (GLCM) and Grid features are extracted from the brain image and these features are classified using ANFIS classifier into normal or abnormal. Then, morphological operations are used to segment the abnormal regions in brain image. Based on the location of these abnormal regions in brain tissues, the segmented tumor regions are diagnosed.

Computer-aided prediction of axillary lymph node status in breast cancer using tumor surrounding tissue features in ultrasound images

Background and objectiveThe presence or absence of axillary lymph node (ALN) metastasis is the most important prognostic factor for patients with early-stage breast cancer. In this study, a computer-aided prediction (CAP) system using the tumor surrounding tissue features in ultrasound (US) images was proposed to determine the ALN status in breast cancer. MethodsThe US imaging database used in this study contained 114 cases of invasive breast cancer and 49 of them were ALN metastasis. After the tumor region segmentation by the level set method, image matting method was used to extract surrounding abnormal tissue of tumor from the acquired images. Then, 21 features composed of 2 intensity, 3 morphology, and 16 textural features are extracted from the surrounding tissue and processed by a logistic regression model. Finally, the prediction model is trained and tested from the selected features. ResultsIn the experiments, the textural feature set extracted from surrounding tissue showed higher performance than intensity and morphology feature sets (Az, 0.7756 vs 0.7071 and 0.6431). The accuracy, sensitivity, specificity and the area index Az under the receiver operating characteristic (ROC) curve for the CAP system were 81.58% (93/114), 81.63% (40/49), 81.54% (53/65), and 0.8269 for using combined feature set. ConclusionsThese results indicated that the proposed CAP system can be helpful to determine the ALN status in patients with breast cancer.

CANFIS based glioma brain tumor classification and retrieval system for tumor diagnosis

AbstractBrain tumor classification and retrieval system plays an important role in medical field. In this paper, an efficient Glioma Brain Tumor detection and its retrieval system is proposed. The proposed methodology consists of two modules as classification and retrieval. The classification modules are designed using preprocessing, feature extraction and tumor detection techniques using Co‐Active Adaptive Neuro Fuzzy Inference System (CANFIS) classifier. The image enhancement can be achieved using Heuristic histogram equalization technique as preprocessing and further texture features as Local Ternary Pattern (LTP) features and Grey Level Co‐occurrence Matrix (GLCM) features are extracted from the enhanced image. These features are used to classify the brain image into normal and abnormal using CANFIS classifier. The tumor region in abnormal brain image is segmented using normalized graph cut segmentation algorithm. The retrieval module is used to retrieve the similar segmented tumor regions from the dataset for diagnosing the tumor region using Euclidean algorithm. The proposed Glioma Brain tumor classification methodology achieves 97.28% sensitivity, 98.16% specificity and 99.14% accuracy. The proposed retrieval system achieves 97.29% precision and 98.16% recall rate with respect to ground truth images.

Volumetric analysis of MR images for glioma classification and their effect on brain tissues

This work aims to identify non-invasive quantitative parameters from three-dimensional brain magnetic resonance images in order: (1) to classify brain tumor (glioma) as low grade (LG) or high grade (HG) and (2) to analyze effect of tumor on brain gray matter (GM) and white matter (WM). In proposed model features were extracted from segmented tumor region based on its volume and shape for distinguishing the tumor grade. Statistical analysis revealed good correlation between segmented tumor volume and tumor grade. Various morphological parameters extracted from segmented tumor region were also significantly different for LG and HG cases ( $$p<0.05$$ ). Also, for brain tumor patients a considerable variation in normalized GM (%GM) volume was obtained compared to normalized WM (%WM) volume for LG and HG cases. We also found that, relative to controls, there was higher effect on %GM and %WM volumes for HG glioma patients as compared to LG glioma patients. The experimental results show that proposed feature set achieves LG/HG classification with high accuracy using support vector machines classifier.

A professional estimate on the computed tomography brain tumor images using SVM-SMO for classification and MRG-GWO for segmentation

In many applications of image processing and computer vision, it is image segmentation that is widely used. The study of the Computed Tomography (CT) images considers image segmentation a very important and vital part in identifying the different kinds of tumor. The classification of the tumor and the non-tumor images followed by the segmentation of tumor region in CT images is done by the proposed methodology. The process of classifying is carried out by Support Vector Machine (SVM) with different kernel functions and optimization techniques. This SVM classifier with Sequential Minimal Optimization (SMO) is predominant over the other methodologies in the analysis of classification process. The segmentation process after the classification process is performed by the Modified Region Growing (MRG) with threshold optimization. As regards the threshold optimization, certain algorithms such as Harmony Search (HS), Evolutionary Programming (EP) and Grey Wolf Optimization (GWO) are made use of. It is with the aid of a wide set of performance measures that the results are demonstrated. The comparative analysis in terms of sensitivity, specificity and accuracy is done for the optimization techniques said supra. An accuracy rate of 99.05% in the analysis of segmentation process is obtained using the GWO technique. It is in the working platform of MATLAB that this proposed methodology is implemented. The experimental results obtained depict that the proposed methodology (MRG-GWO) enjoys high accuracy with enhanced tumor detection in total contrast to the other two techniques (HS and EP) in comparison.

Brain tumor segmentation based on a new threshold approach

Brain cancer is an abnormal cell population that occurs in the brain. Nowadays, medical imaging techniques play an important role in cancer diagnosis. Magnetic resonance imaging (MRI) is one of the most used techniques to identify and locate the tumor in the brain. Images obtained by medical imaging techniques may become a better quality image thru applying image processing techniques. In this study, we aim to develop a method for clearly distinguishing the tissues affected by the cancer. The proposed approach is used to obtain a segmented tumor region clear enough to be observed by the medical practitioner and give them more detail about the tumor in their diagnosis. In the proposed approach, morphological operations, pixel subtraction, threshold based segmentation and image filtering techniques are used. The proposed approach is based on obtaining clear images of the skull, brain and the tumor. When compared, the proposed approach gave a better result than the other approach.

SU‐F‐J‐207: Non‐Small Cell Lung Cancer Patient Survival Prediction with Quantitative Tumor Textures Analysis in Baseline CT

Purpose:Chemo‐radiation therapy (CRT) is widely used in treating patients with locally advanced non‐small cell lung cancer (NSCLC). Determination of the likelihood of patient response to treatment and optimization of treatment regime is of clinical significance. Up to date, no imaging biomarker has reliably correlated to NSCLC patient survival rate. This pilot study is to extract CT texture information from tumor regions for patient survival prediction.Methods:Thirteen patients with stage II‐III NSCLC were treated using CRT with a median dose of 6210 cGy. Non‐contrast‐enhanced CT images were acquired for treatment planning and retrospectively collected for this study. Texture analysis was applied in segmented tumor regions using the Local Binary Pattern method (LBP). By comparing its HU with neighboring voxels, the LBPs of a voxel were measured in multiple scales with different group radiuses and numbers of neighbors. The LBP histograms formed a multi‐dimensional texture vector for each patient, which was then used to establish and test a Support Vector Machine (SVM) model to predict patients’ one year survival. The leave‐one‐out cross validation strategy was used recursively to enlarge the training set and derive a reliable predictor. The predictions were compared with the true clinical outcomes.Results:A 10‐dimensional LBP histogram was extracted from 3D segmented tumor region for each of the 13 patients. Using the SVM model with the leave‐one‐out strategy, only 1 out of 13 patients was misclassified. The experiments showed an accuracy of 93%, sensitivity of 100%, and specificity of 86%.Conclusion:Within the framework of a Support Vector Machine based model, the Local Binary Pattern method is able to extract a quantitative imaging biomarker in the prediction of NSCLC patient survival. More patients are to be included in the study.

Brain tumor detection and diagnosis using ANFIS classifier

In this article, the segmented brain tumor region is diagnosed into mild, moderate, and severe case based on the presence of tumor cells in the brain components such as Gray Matter (GM), White Matter (WM), and cerebrospinal fluid (CSF). The modified spatial fuzzy c mean algorithm is used to segment brain tissues. The feature Local binary pattern is extracted from segmented tissues, which is trained and classified by ANFIS Classifier. The performance of the proposed brain tissues segmentation system is analyzed in terms of sensitivity, specificity, and accuracy with respect to manually segmented ground truth images. The severity of brain tumor is diagnosed into mild case if the segmented brain tumor is present in the grey matter. The severity of brain tumor is diagnosed into moderate case if the segmented brain tumor is present in the WM. The severity of brain tumor is diagnosed into severe case if the segmented brain tumor is present in the CSF region. The immediate surgery is required for severe case and medical treatment is preferred for mild and moderate case.

TU‐AB‐202‐11: Tumor Segmentation by Fusion of Multi‐Tracer PET Images Using Copula Based Statistical Methods

Purpose:Multi‐tracer PET imaging is getting more attention in radiotherapy by providing additional tumor volume information such as glucose and oxygenation. However, automatic PET‐based tumor segmentation is still a very challenging problem. We propose a statistical fusion approach to joint segment the sub‐area of tumors from the two tracers FDG and FMISO PET images.Methods:Non‐standardized Gamma distributions are convenient to model intensity distributions in PET. As a serious correlation exists in multi‐tracer PET images, we proposed a new fusion method based on copula which is capable to represent dependency between different tracers. The Hidden Markov Field (HMF) model is used to represent spatial relationship between PET image voxels and statistical dynamics of intensities for each modality. Real PET images of five patients with FDG and FMISO are used to evaluate quantitatively and qualitatively our method. A comparison between individual and multi‐tracer segmentations was conducted to show advantages of the proposed fusion method.Results:The segmentation results show that fusion with Gaussian copula can receive high Dice coefficient of 0.84 compared to that of 0.54 and 0.3 of monomodal segmentation results based on individual segmentation of FDG and FMISO PET images. In addition, high correlation coefficients (0.75 to 0.91) for the Gaussian copula for all five testing patients indicates the dependency between tumor regions in the multi‐tracer PET images.Conclusion:This study shows that using multi‐tracer PET imaging can efficiently improve the segmentation of tumor region where hypoxia and glucidic consumption are present at the same time. Introduction of copulas for modeling the dependency between two tracers can simultaneously take into account information from both tracers and deal with two pathological phenomena. Future work will be to consider other families of copula such as spherical and archimedian copulas, and to eliminate partial volume effect by considering dependency between neighboring voxels.

Segmentation of Tumor Region from Brain Mri Images Using Fuzzy C-Means Clustering And Seeded Region Growing

The detection of brain tumor is one of the most challenging tasks in the field of medical image processing, since brain images are very complicated and tumors can be analyzed efficiently only by the expert radiologists. Therefore, there is a significant need to automate this process. In this paper, a method for the automatic detection of the tumor from the brain magnetic resonance imaging (MRI) images has been proposed. For this, the region-based segmentation of the input MRI image is done. The wavelet-based decomposition of the input image is done and the input image is reconstructed on the basis of soft thresholding for the enhancement of the image. After that, fuzzy c-means clustering (FCM) followed by seeded region growing is applied to detect and segment the tumor from the brain MRI image and finally comparison with Sobel operator is done on the basis of Relative Ultimate Measurement Accuracy (RUMA) and Standard Deviation (SD). The results for the proposed technique are better than the results obtained by using Sobel operator.

Geo-Cutting Liver Tumor

The issue of intuitive frontal area/foundation division in still pictures is of awesome down to earth significance in picture altering. They maintain a strategic distance from the limit length predisposition of chart cut strategies and results in expanded affectability to seed situation. Another proposed technique for completely programmed handling structures is given taking into account Graph-cut and Geodesic Graph cut calculations. This paper addresses the issue of dividing liver and tumor locales from the stomach CT pictures. The absence of edge displaying in geodesic or comparable methodologies confines their capacity to exactly restrict object limits, something at which chart cut strategies by and large exceed expectations. A predicate is characterized for measuring the confirmation for a limit between two locales utilizing Geodesic Graph-based representation of the picture. The calculation is connected to picture division utilizing two various types of nearby neighborhoods in building the chart. Liver and hepatic tumor division can be naturally prepared by the Geodesic chart cut based strategy. This framework has focused on finding a quick and intuitive division strategy for liver and tumor division. In the pre-handling stage, Mean movement channel is connected to CT picture process and factual thresholding technique is connected for diminishing preparing zone with enhancing discoveries rate. In the Second stage, the liver area has been divided utilizing the calculation of the proposed strategy. Next, the tumor district has been portioned utilizing Geodesic Graph cut strategy. Results demonstrate that the proposed strategy is less inclined to shortcutting than run of the mill diagram cut techniques while being less delicate to seed position and preferable at edge restriction over geodesic strategies. This prompts expanded division exactness and decreased exertion with respect to the client. At long last Segmented Liver and Tumor Regions were appeared from the stomach Computed Tomographic picture.

Quantification of breast tumor heterogeneity for ER status, HER2 status, and TN molecular subtype evaluation on DCE-MRI

PurposeRecognizing molecular markers is helpful for guiding treatment plans for breast cancer. This study correlated estrogen receptor (ER), human epidermal growth factor receptor 2 (HER2), and triple-negative breast cancer (TNBC) statuses to the degree of heterogeneity on breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Materials and methodsA total of 102 biopsy-proven cancers from 102 patients between October 2010 and December 2012 were used in this study, including ER (59 positive, 43 negative), HER2 (47 positive, 55 negative), and TNBC (22 TNBC, 80 non-TNBC). At first, the tumor region was segmented by using a region growing method. Then, the region-based features were extracted by the proposed regionalization method to quantify intra-tumoral heterogeneity on breast DCE-MRI. The three-dimensional morphological features (texture features and shape feature) and the pharmacokinetic model were also extracted from the segmented tumor region. After feature extraction, a logistic regression was used to classify ER, HER2, and TNBC statuses respectively. The performances were evaluated by using receiver operating characteristic (ROC) curve analysis. ResultsThe proposed region-based features achieved the accuracy of 73.53%, 82.35%, and 77.45% for ER, HER2, and TNBC classifications. The corresponding area under the ROC curves (Az) achieves 0.7320, 0.8458, and 0.8328 that were better than those of texture features, shape features, and Tofts pharmacokinetic model. ConclusionThe intra-tumoral heterogeneity quantified by the region-based features can be used to reflect the vasculature complexity of different molecular markers and to provide prediction information of cell surface receptors on clinical examination.

Implementation of Biography Based Neural Clustering (BBNC) with Genetic Processing for tumor detection from medical images

Segmentation is a best method to divide the required region from the medical images. This research is based on segmentation of medical images (MRI, CT scans) based on the previous method known as pre-operative and post-recurrence tumor registration (PORTR) and proposed method biography based neural clustering (BBNC) with genetic processing for tumor segmentation. By using the new technique the extracted part can be view in 3D model and also can get the actual segmented tumor region. This new method will be helpful for diagnostics to find the tumor area as well as pixel difference in segmented part to define the tumor area accurately. While in the previous approach all the parameters have been used likewise, in which the registration method is used to transform the different sets of data into one coordinate system for segmentation of medical images. Registration basically is used to improve the signals to reduce the noise from the images. These techniques are better to find the tumor area from the MRI and CT scans, but after comparing them better results have been obtained in proposed technique. The proposed technique (BBNC) reduces the extracted region again into required and actual region of tumor with accuracy of area, time and pixel difference.

Implementation of Biography Based Neural Clustering (BBNC) with Genetic Processing for Tumor Detection from Medical Images

Implementation of Biography Based Neural Clustering (BBNC) with Genetic Processing for Tumor Detection from Medical Images This research paper presenting the segmentation of medical images (MRI, CT scans) based on the previous method pre operative and post recurrence brain tumor registration (PORTR) and proposed method biography based neural clustering (BBNC) with genetic processing for tumor segmentation. By using the new technique the extracted part can be view in 3D model and also can get the actual segmented tumor region. This new method will be helpful for diagnostics to find the tumor area as well as pixel difference in segmented part to define the tumor area accurately. While in the previous approach all the parameters have been used likewise, in which the registration method is used to transform the different sets of data into one coordinate system for segmentation of medical images. Registration basically is used to improve the signals to reduce the noise from the images. These techniques are better to find the tumor area from the MRI and CT scans, but after comparing them better results have been obtained in proposed technique. The proposed technique (BBNC) reduces the extracted region again into required and actual region of tumor with accuracy of area, time and pixel difference.

Radiomic Machine-Learning Classifiers for Prognostic Biomarkers of Head and Neck Cancer.

"Radiomics" extracts and mines a large number of medical imaging features in a non-invasive and cost-effective way. The underlying assumption of radiomics is that these imaging features quantify phenotypic characteristics of an entire tumor. In order to enhance applicability of radiomics in clinical oncology, highly accurate and reliable machine-learning approaches are required. In this radiomic study, 13 feature selection methods and 11 machine-learning classification methods were evaluated in terms of their performance and stability for predicting overall survival in head and neck cancer patients. Two independent head and neck cancer cohorts were investigated. Training cohort HN1 consisted of 101 head and neck cancer patients. Cohort HN2 (n = 95) was used for validation. A total of 440 radiomic features were extracted from the segmented tumor regions in CT images. Feature selection and classification methods were compared using an unbiased evaluation framework. We observed that the three feature selection methods minimum redundancy maximum relevance (AUC = 0.69, Stability = 0.66), mutual information feature selection (AUC = 0.66, Stability = 0.69), and conditional infomax feature extraction (AUC = 0.68, Stability = 0.7) had high prognostic performance and stability. The three classifiers BY (AUC = 0.67, RSD = 11.28), RF (AUC = 0.61, RSD = 7.36), and NN (AUC = 0.62, RSD = 10.52) also showed high prognostic performance and stability. Analysis investigating performance variability indicated that the choice of classification method is the major factor driving the performance variation (29.02% of total variance). Our study identified prognostic and reliable machine-learning methods for the prediction of overall survival of head and neck cancer patients. Identification of optimal machine-learning methods for radiomics-based prognostic analyses could broaden the scope of radiomics in precision oncology and cancer care.

Abstract LB-B18: Epithelial to mesenchymal transition in human tumor biopsies: Quantitative, histopathological proof of the existence of EMT in vivo by immunofluorescence microscopy

Abstract Epithelial-mesenchymal transition (EMT) is a dynamic process whereby epithelial cells acquire mesenchymal properties. Despite the clinical significance of the acquired mesenchymal properties for metastasis and drug resistance, histopathological evidence of transitional cells in patient samples is lacking and EMT remains an unproven clinical hypothesis. We previously developed and validated a multiplex immunofluorescence assay (IFA) that quantifies the levels of EMT biomarkers (E-Cadherin, Vimentin, β-catenin) in snap-frozen, formalin-fixed, paraffin-embedded (FFPE) tumor tissue (Navas et al, AACR 2013). Building upon that method, we now report a precise, quantitative and unbiased IFA method of tissue analysis (EMT-IFA) using Definiens® software to quantify co-expression of the epithelial marker E-cadherin (E) and mesenchymal marker Vimentin (V) at the cellular level in FFPE clinical biopsies. FFPE human tumor xenografts and cell lines serve as calibrators and reference materials for establishing initial image acquisition parameters for segmented tumor regions of interest. Flanking H&amp;E sections of the clinical biopsies are initially annotated by an anatomic pathologist to identify regions of tumor tissue, non-tumor related areas (including normal or stromal tissues), inflammatory and/or necrotic regions, and freeze artifacts. Adjacent sections are used for the EMT-IFA, and further tumor segmentation from stroma is applied to extracted regions of interests (fields) using the β-catenin layer. Within segmented tumor regions, E-cadherin (E) and Vimentin (V) sub-cellular pixel areas are measured using predetermined thresholds for each biomarker and reported as Log10 (V:E). High resolution images confirmed the co-localization of plasma membrane E-cadherin and cytoplasmic Vimentin to individual transitional tumor cells, and the co-localization of E and V within individual tumor cells of the segmented regions of interest was key for distinguishing EMT from tumor heterogeneity in which adjacent tumor cells can be exclusively E+ or V+. Quantitative analysis of a clinical biopsy series of various histologies revealed all possible phenotypes: epithelial (E+V- colorectal carcinoma), mesenchymal (E-V+ sarcomas), heterogeneous mixtures of E+V- and E-V+ subpopulations, and EMT. Importantly, using the EMT-IFA, we discovered that pharmacological targeting of VEGFR/FGFR/PDGFR signaling with the multi-kinase inhibitor pazopanib stimulated EMT in a preclinical xenograft model of gastric carcinoma (MKN45), as revealed by the significant increase in the V:E ratio (P = 0.0023) coupled with an increase in the EMT phenotype (co-localized V+E+ at the individual cell level) (P = 0.0070) in biopsy specimens. Thus, the EMT-IFA has potential value for investigating EMT in the clinic and its ramifications for drug response and other clinical endpoints. Funded by NCI Contract No. HHSN261200800001E. Citation Format: Tony Navas, Robert J. Kinders, Scott M. Lawrence, Katherine Ferry-Galow, Thomas D. Pfister, Apurva K. Srivastava, Sergio Y. Alcoser, Melinda G. Hollingshead, Lindsay M. Dutko, Brad A. Gouker, Donna O. Butcher, Hala Makhlouf, Rodrigo Chuaqui, Donald P. Bottaro, Shivaani Kummar, Alice Chen, James H. Doroshow, Ralph E. Parchment. Epithelial to mesenchymal transition in human tumor biopsies: Quantitative, histopathological proof of the existence of EMT in vivo by immunofluorescence microscopy. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr LB-B18.

Quantitative computed tomographic descriptors associate tumor shape complexity and intratumor heterogeneity with prognosis in lung adenocarcinoma.

Two CT features were developed to quantitatively describe lung adenocarcinomas by scoring tumor shape complexity (feature 1: convexity) and intratumor density variation (feature 2: entropy ratio) in routinely obtained diagnostic CT scans. The developed quantitative features were analyzed in two independent cohorts (cohort 1: n = 61; cohort 2: n = 47) of patients diagnosed with primary lung adenocarcinoma, retrospectively curated to include imaging and clinical data. Preoperative chest CTs were segmented semi-automatically. Segmented tumor regions were further subdivided into core and boundary sub-regions, to quantify intensity variations across the tumor. Reproducibility of the features was evaluated in an independent test-retest dataset of 32 patients. The proposed metrics showed high degree of reproducibility in a repeated experiment (concordance, CCC≥0.897; dynamic range, DR≥0.92). Association with overall survival was evaluated by Cox proportional hazard regression, Kaplan-Meier survival curves, and the log-rank test. Both features were associated with overall survival (convexity: p = 0.008; entropy ratio: p = 0.04) in Cohort 1 but not in Cohort 2 (convexity: p = 0.7; entropy ratio: p = 0.8). In both cohorts, these features were found to be descriptive and demonstrated the link between imaging characteristics and patient survival in lung adenocarcinoma.