Jiajing Xu

Non–Small Cell Lung Cancer: Identifying Prognostic Imaging Biomarkers by Leveraging Public Gene Expression Microarray Data—Methods and Preliminary Results

PURPOSE To identify prognostic imaging biomarkers in non-small cell lung cancer (NSCLC) by means of a radiogenomics strategy that integrates gene expression and medical images in patients for whom survival outcomes are not available by leveraging survival data in public gene expression data sets. MATERIALS AND METHODS A radiogenomics strategy for associating image features with clusters of coexpressed genes (metagenes) was defined. First, a radiogenomics correlation map is created for a pairwise association between image features and metagenes. Next, predictive models of metagenes are built in terms of image features by using sparse linear regression. Similarly, predictive models of image features are built in terms of metagenes. Finally, the prognostic significance of the predicted image features are evaluated in a public gene expression data set with survival outcomes. This radiogenomics strategy was applied to a cohort of 26 patients with NSCLC for whom gene expression and 180 image features from computed tomography (CT) and positron emission tomography (PET)/CT were available. RESULTS There were 243 statistically significant pairwise correlations between image features and metagenes of NSCLC. Metagenes were predicted in terms of image features with an accuracy of 59%-83%. One hundred fourteen of 180 CT image features and the PET standardized uptake value were predicted in terms of metagenes with an accuracy of 65%-86%. When the predicted image features were mapped to a public gene expression data set with survival outcomes, tumor size, edge shape, and sharpness ranked highest for prognostic significance. CONCLUSION This radiogenomics strategy for identifying imaging biomarkers may enable a more rapid evaluation of novel imaging modalities, thereby accelerating their translation to personalized medicine.

Glioblastoma Multiforme: Exploratory Radiogenomic Analysis by Using Quantitative Image Features

PURPOSE To derive quantitative image features from magnetic resonance (MR) images that characterize the radiographic phenotype of glioblastoma multiforme (GBM) lesions and to create radiogenomic maps associating these features with various molecular data. MATERIALS AND METHODS Clinical, molecular, and MR imaging data for GBMs in 55 patients were obtained from the Cancer Genome Atlas and the Cancer Imaging Archive after local ethics committee and institutional review board approval. Regions of interest (ROIs) corresponding to enhancing necrotic portions of tumor and peritumoral edema were drawn, and quantitative image features were derived from these ROIs. Robust quantitative image features were defined on the basis of an intraclass correlation coefficient of 0.6 for a digital algorithmic modification and a test-retest analysis. The robust features were visualized by using hierarchic clustering and were correlated with survival by using Cox proportional hazards modeling. Next, these robust image features were correlated with manual radiologist annotations from the Visually Accessible Rembrandt Images (VASARI) feature set and GBM molecular subgroups by using nonparametric statistical tests. A bioinformatic algorithm was used to create gene expression modules, defined as a set of coexpressed genes together with a multivariate model of cancer driver genes predictive of the modules expression pattern. Modules were correlated with robust image features by using the Spearman correlation test to create radiogenomic maps and to link robust image features with molecular pathways. RESULTS Eighteen image features passed the robustness analysis and were further analyzed for the three types of ROIs, for a total of 54 image features. Three enhancement features were significantly correlated with survival, 77 significant correlations were found between robust quantitative features and the VASARI feature set, and seven image features were correlated with molecular subgroups (P < .05 for all). A radiogenomics map was created to link image features with gene expression modules and allowed linkage of 56% (30 of 54) of the image features with biologic processes. CONCLUSION Radiogenomic approaches in GBM have the potential to predict clinical and molecular characteristics of tumors noninvasively. Online supplemental material is available for this article.

Automated Retrieval of CT Images of Liver Lesions on the Basis of Image Similarity: Method and Preliminary Results

PURPOSE To develop a system to facilitate the retrieval of radiologic images that contain similar-appearing lesions and to perform a preliminary evaluation of this system with a database of computed tomographic (CT) images of the liver and an external standard of image similarity. MATERIALS AND METHODS Institutional review board approval was obtained for retrospective analysis of deidentified patient images. Thereafter, 30 portal venous phase CT images of the liver exhibiting one of three types of liver lesions (13 cysts, seven hemangiomas, 10 metastases) were selected. A radiologist used a controlled lexicon and a tool developed for complete and standardized description of lesions to identify and annotate each lesion with semantic features. In addition, this software automatically computed image features on the basis of image texture and boundary sharpness. Semantic and computer-generated features were weighted and combined into a feature vector representing each image. An independent reference standard was created for pairwise image similarity. This was used in a leave-one-out cross-validation to train weights that optimized the rankings of images in the database in terms of similarity to query images. Performance was evaluated by using precision-recall curves and normalized discounted cumulative gain (NDCG), a common measure for the usefulness of information retrieval. RESULTS When used individually, groups of semantic, texture, and boundary features resulted in various levels of performance in retrieving relevant lesions. However, combining all features produced the best overall results. Mean precision was greater than 90% at all values of recall, and mean, best, and worst case retrieval accuracy was greater than 95%, 100%, and greater than 78%, respectively, with NDCG. CONCLUSION Preliminary assessment of this approach shows excellent retrieval results for three types of liver lesions visible on portal venous CT images, warranting continued development and validation in a larger and more comprehensive database.

A Comprehensive Descriptor of Shape: Method and Application to Content-Based Retrieval of Similar Appearing Lesions in Medical Images

We have developed a method to quantify the shape of liver lesions in CT images and to evaluate its performance for retrieval of images with similarly-shaped lesions. We employed a machine learning method to combine several shape descriptors and defined similarity measures for a pair of shapes as a weighted combination of distances calculated based on each feature. We created a dataset of 144 simulated shapes and established several reference standards for similarity and computed the optimal weights so that the retrieval result agrees best with the reference standard. Then we evaluated our method on a clinical database consisting of 79 portal-venous-phase CT liver images, where we derived a reference standard of similarity from radiologists’ visual evaluation. Normalized Discounted Cumulative Gain (NDCG) was calculated to compare this ordering with the expected ordering based on the reference standard. For the simulated lesions, the mean NDCG values ranged from 91% to 100%, indicating that our methods for combining features were very accurate in representing true similarity. For the clinical images, the mean NDCG values were still around 90%, suggesting a strong correlation between the computed similarity and the independent similarity reference derived the radiologists.

On the Feasibility of Predicting Radiological Observations from Computational Imaging Features of Liver Lesions in CT Scans

We aim to predict radiological observations using computationally-derived imaging features extracted from CT images. Our dataset consists of 79 portal venous phase liver CT images containing lesions identified and annotated by a radiologist using a controlled vocabulary of 76 semantic terms. Computationally-derived features were extracted describing intensity, texture, shape, and edge sharpness. Linear discriminative analysis, logistic regression and LASSO were explored to predict the radiological observations using computational features. The approach was evaluated by leave one out cross-validation. Informative radiological observations such as lesion enhancement, hyper vascular attenuation, and homogeneous retention were discovered to be well-predicted by computational features. By exploiting relationships between computable and semantic features, this approach could lead to more accurate and efficient radiology reporting.

Snake model-based lymphoma segmentation for sequential CT images

The measurement of the size of lesions in follow-up CT examinations of cancer patients is important to evaluate the success of treatment. This paper presents an automatic algorithm for identifying and segmenting lymph nodes in CT images across longitudinal time points. Firstly, a two-step image registration method is proposed to locate the lymph nodes including coarse registration based on body region detection and fine registration based on a double-template matching algorithm. Then, to make the initial segmentation approximate the boundaries of lymph nodes, the initial image registration result is refined with intensity and edge information. Finally, a snake model is used to evolve the refined initial curve and obtain segmentation results. Our algorithm was tested on 26 lymph nodes at multiple time points from 14 patients. The image at the earlier time point was used as the baseline image to be used in evaluating the follow-up image, resulting in 76 total test cases. Of the 76 test cases, we made a 76 (100%) successful detection and 38/40 (95%) correct clinical assessment according to Response Evaluation Criteria in Solid Tumors (RECIST). The quantitative evaluation based on several metrics, such as average Hausdorff distance, indicates that our algorithm is produces good results. In addition, the proposed algorithm is fast with an average computing time 2.58s. The proposed segmentation algorithm for lymph nodes is fast and can achieve high segmentation accuracy, which may be useful to automate the tracking and evaluation of cancer therapy.

Abstract 5561: Radiogenomic analysis indicates MR images are potentially predictive of EGFR mutation status in glioblastoma multiforme

Proceedings: AACR 103rd Annual Meeting 2012‐‐ Mar 31‐Apr 4, 2012; Chicago, IL Objective: To predict mutations of key genes in glioblastoma multiforme (GBM) from MR image features. Methods: We obtained mutational and MR image data from 35 patients in the Cancer Genome Atlas (TCGA) GBM database. T1-weighted axial images pre and post gadolinium contrast MRI were processed as follows: a board certified neuro-radiologist traced a region of interest (ROI) around the enhanced part of the largest lesion in the T1 post-contrast MRI and confirmed by comparing it with the T1 pre-contrast image. This ROI was then used to compute features that characterized the intensity of the enhanced lesion, the sharpness of lesion boundaries and the boundary shape. We used the resulting data set to build a linear regression model with regularization to predict the presence of a mutation in terms of image features. We focused on predicting mutations in EGFR because several drugs target it and because EGFR mutations are prevalent in the TCGA GBM data. We evaluated the performance of predicting EGFR mutations from MR imaging data using 5-fold cross validation (5F-CV) and the area under the ROC Curve (AUC). The optimal ROC operating point was defined as the point with the maximal sum of sensitivity and specificity when the cost of false positives and false negatives are considered equivalent. Results: We found that computationally-derived MR image features can predict the presence of EGFR mutations with an AUC of 0.80. The optimal operating point had a sensitivity and specificity of 83% and 93% respectively. The top ranked features in the image-based EGFR-predictor model suggest that EGFR-mutated tumors have blurrier edges than EGFR-wild-type tumors. Conclusion: Our preliminary radiogenomic analysis of GBM suggests MR images may be used to non-invasively determine the mutation status of EGFR, an important drug target in glioblastoma. The ability to determine important genomic aberrations based on image data suggests a increasingly important role for imaging in personalized medicine and warrants further investigation. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5561. doi:1538-7445.AM2012-5561

Abstract 4148: Integrating medical images and transcriptomic data in non-small cell lung cancer

Objective: To build an association map between medical images (CT/PET) and gene expression microarrays for Non-Small Cell Lung Cancer (NSCLC) from which to derive relationships between imaging features and gene expression. Methods: We studied 26 cases of NSCLC using CT and PET images and microarray data from excised tumors. An experienced thoracic radiologist annotated the CT image using “semantic features” from a controlled vocabulary, a nuclear medicine physician extracted the Standard Uptake Value (SUV) from the PET scan, and we developed and applied algorithms to extract “computational features” that characterized the lesion9s image texture using Gabor and other texture features, the sharpness of lesion boundaries and the lesion boundary shape, including notions of compactness, roughness, and other shape signatures. We preprocessed the microarray data using log transformation and quantile normalization, obtained 100 co-expressed gene clusters using k-means clustering, and computed a metagene for each cluster using its first principal component. We performed (a) univariate and (b) multivariate analyses to integrate imaging features and metagenes using (a) Significance Analysis of Microarrays (SAM) with False Discovery Rate (FDR) multiple testing correction, and (b) Sparse Canonical Correlation Analysis (SCCA), respectively. Results: Image features included 44 semantic terms, 107 computational features, and SUV for each tumor. In a univariate analysis, 60 CT-features and SUV were significantly associated with at least one metagene, and on average 3.8 metagenes (FDR Conclusion: The integration of medical image features and gene expression promises to reveal molecular characteristics underlying medical image features. For translational purposes, this work highlights the potential use of medical image features as predictive markers for molecularly-targeted therapeutics. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4148. doi:10.1158/1538-7445.AM2011-4148