PET Image Features Research Articles

Prediction of Neoadjuvant Chemoradiotherapy Response in Rectal Cancer Patients Using Harmonized Radiomics of Multcenter 18F-FDG-PET Image

We developed machine and deep learning models to predict chemoradiotherapy in rectal cancer using 18F-FDG PET images and harmonized image features extracted from 18F-FDG PET/CT images. Patients diagnosed with pathologic T-stage III rectal cancer with a tumor size > 2 cm were treated with neoadjuvant chemoradiotherapy. Patients with rectal cancer were divided into an internal dataset (n = 116) and an external dataset obtained from a separate institution (n = 40), which were used in the model. AUC was calculated to select image features associated with radiochemotherapy response. In the external test, the machine-learning signature extracted from 18F-FDG PET image features achieved the highest accuracy and AUC value of 0.875 and 0.896. The harmonized first-order radiomics model had a higher efficiency with accuracy and an AUC of 0.771 than the second-order model in the external test. The deep learning model using the balanced dataset showed an accuracy of 0.867 in the internal test but an accuracy of 0.557 in the external test. Deep-learning models using 18F-FDG PET images must be harmonized to demonstrate reproducibility with external data. Harmonized 18F-FDG PET image features as an element of machine learning could help predict chemoradiotherapy responses in external tests with reproducibility.

RADT-15. 18F-DOPA PET/CT SURVEILLANCE FOR GLIOBLASTOMA: A RADIOMIC MODEL FROM A PROSPECTIVE PHASE II CLINICAL TRIAL PREDICTING SURVIVAL IN IDH-WILDTYPE, MGMT-UNMETHYLATED PATIENTS

Abstract BACKGROUND Interpretation of serial magnetic resonance imaging (MRI) for glioblastoma following radiation therapy (RT) is complicated by difficulty differentiating tumor from treatment-related changes, even using updated RANO criteria. The incorporation of novel imaging such as 3,4-dihydroxy-6-[18F]-fluoro-L-phenylalanine (18F-DOPA) PET/CT to post-treatment serial imaging may improve prognostication and better facilitate future treatment decisions. METHODS The secondary analysis of a recent phase II prospective clinical trial of 18F-DOPA PET/CT-directed dose-escalated RT included patients with IDH-wildtype, MGMT-unmethylated glioblastoma who underwent post-treatment serial 18F-DOPA PET/CT surveillance. Quantitative features were extracted from pre-RT and post-RT serial PET/CT images, and robust prognostic features were selected using an in-house workflow. Both an automated machine learning (ML) algorithm and an interpretable ML algorithm were utilized to correlate surveillance PET image features with subsequent survival of greater than 12 months versus less than 12 months from the surveillance timepoint. Changes from pre-RT to post-RT PET/CT (delta model) were also assessed for association with post-RT survival and validated with a separate cohort. RESULTS Thirty-five patients with IDH-wildtype, MGMT-unmethylated glioblastoma who had at least one available (range: 1-14) post-treatment 18F-DOPA PET/CT were included. Twenty-four were used for model training, while 11 were used for validation. Ultimately, a five-feature post-RT model utilizing two shape, two texture, and one first-order radiomic feature was selected. For the delta model, five texture, two shape, and one first order radiomic feature were selected. The models show 90% accuracy in predicting survival < 12 months post-surveillance on the training set, and 68%-73% accuracy (AUC 0.64-0.73) for the validation cohort. Delta features were significantly associated with overall survival (p < 0.05). CONCLUSIONS Post-RT serial 18F-DOPA PET/CT imaging can help distinguish true tumor progression in patients with glioblastoma using a radiomics model. Tumor response evaluated for changes from pre-RT to post-RT 18F-DOPA PET/CT also predicted subsequent overall survival.

APU-Net: An Attention Mechanism Parallel U-Net for Lung Tumor Segmentation.

Lung cancer is one of the malignant tumors with high morbidity and mortality, and lung nodules are the early stages of lung cancer. The symptoms of pulmonary nodules are not obvious in the clinic, and the optimal treatment time is missed due to the missed diagnosis in the clinic. A parallel U-Net network called APU-Net is proposed. Firstly, two parallel U-Net networks are used to extract the features of different modalities. Among them, the subnetwork UNet_B extracts the CT image features, and the subnetwork UNet_A consists of two encoders to extract the PET/CT and PET image features. Secondly, multimodal feature extraction blocks are used to extract features for PET/CT and PET images in UNet_B network. Thirdly, a hybrid attention mechanism is added to the encoding paths of the UNet_A and UNet_B. Finally, a multiscale feature aggregation block is used for extracting feature maps of different scales of decoding path. On the lung tumor 18FDGPET/CT multimodal medical images dataset, experiments' results show that the DSC, Recall, VOE, and RVD coefficients of APU-Net are 96.86%, 97.53%, 3.18%, and 3.29%, respectively. APU-Net can improve the segmentation accuracy of the adhesion between the lesion of complex shape and the normal tissue. This has positive significance for computer-aided diagnosis.

OD73 - How to identify robust PET image features for their use in quantitative analysis

OD73 - How to identify robust PET image features for their use in quantitative analysis

Predictive Value of Positron Emission Tomography/CT Imaging in Distant Metastasis in Early and Locally Advanced Lung Cancer

The imaging features of advanced non-small cell lung cancer identified by imaging approaches are particularly critical for the clinical prognosis and treatment. Our study explores the predictive value of Positron Emission Tomography–Computed Tomography (PET/CT) imaging in distant metastasis in early and locally advanced lung cancer. From September 2017 to September 2019, 121 patients with PET images were enrolled. 80 patient’s data were used to simulate the cohort and the data on 41 patients were used to validate the cohort. Quantitative PET image features were assessed. A Cox model was used to predict the distant metastasis. A combination of imaging and histology type prognostic models was also evaluated. The best prognostic model includes two features, which can quantify intratumoral heterogeneity. In independent validation cohort, the consistency index of model was 0.71. This prognostic model allows high-risk and low-risk group with distant metastases (hazard ratio 4.8, P < 0.05). The ROC curve revealed a separation of the combined imaging and histology type among high-and low-risk group. The consistency index was also improved to 0.80. PET features related to distant metastases have been identified and these features may help develop appropriate treatment options for early and locally advanced lung cancer patients.

Metabolic radiogenomics in lung cancer: associations between FDG PET image features and oncogenic signaling pathway alterations

This study investigated the associations between image features extracted from tumor 18F-fluorodeoxyglucose (FDG) uptake and genetic alterations in patients with lung cancer. A total of 137 patients (age, 62.7 ± 10.2 years) who underwent FDG positron emission tomography/computed tomography (PET/CT) and targeted deep sequencing analysis for a tumor lesion, comprising 61 adenocarcinoma (ADC), 31 squamous cell carcinoma (SQCC), and 45 small cell lung cancer (SCLC) patients, were enrolled in this study. From the tumor lesions, 86 image features were extracted, and 381 genes were assessed. PET features were associated with genetic mutations: 41 genes with 24 features in ADC; 35 genes with 22 features in SQCC; and 43 genes with 25 features in SCLC (FDR < 0.05). Clusters based on PET features showed an association with alterations in oncogenic signaling pathways: Cell cycle and WNT signaling pathways in ADC (p = 0.023, p = 0.035, respectively); Cell cycle, p53, and WNT in SQCC (p = 0.045, 0.009, and 0.029, respectively); and TGFβ in SCLC (p = 0.030). In addition, SUVpeak and SUVmax were associated with a mutation of the TGFβ signaling pathway in ADC (FDR = 0.001, < 0.001). In this study, PET image features had significant associations with alterations in genes and oncogenic signaling pathways in patients with lung cancer.

Cyclo-oxygenase-2 expression is associated with mean standardised uptake value on 18F-Fluorodeoxyglucose positron emission tomography in oesophageal adenocarcinoma.

This pilot study investigated the association of four PET image features and cyclo-oxygenase-2 (COX-2) expression in patients with oesophageal adenocarcinoma. The prognostic significance of these biomarkers was also assessed. 50 consecutive patients [median age = 68 (range 47 - 84), males = 45) with oesophageal adenocarcinoma had PET/CT staging between January 2011 and July 2015. The maximum and mean standardised uptake values (SUVmax and SUVmean), metabolic tumour volume (MTV) and tumour lesion glycolysis (TLG) were calculated from the primary tumour. Their association with COX-2 status was assessed using Mann-Whitney U tests. Kaplan-Meier and Cox regression analysis tested their prognostic significance. A p-value < 0.05 was considered statistically significant. 32 tumours (64.0%) were COX-2 positive. There was a significant association between SUVmean and COX-2 status (p = 0.019). TLG (hazard ratio (HR) 1.001, 95 % confidence intervals (CI) 1.000 - 1.002, p = 0.018) was significantly associated with overall survival on multivariable analysis. This study investigated the association between PET image features and COX-2 expression in oesophageal adenocarcinoma. The preliminary results signal that a combination of TLG (calculated as product of MTV and SUVmean) and COX-2 status may be a strong and clinically important prognostic biomarker. Our research group are planning a prospective, multi-centre study to validate these findings. Mean standardised uptake value (SUVmean) on PET imaging is associated with COX-2 expression in oesophageal adenocarcinoma.

External validation of a prognostic model incorporating quantitative PET image features in oesophageal cancer

AimEnhanced prognostic models are required to improve risk stratification of patients with oesophageal cancer so treatment decisions can be optimised. The primary aim was to externally validate a published prognostic model incorporating PET image features. Transferability of the model was compared using only clinical variables. MethodsThis was a Transparent Reporting of a multivariate prediction model for Individual Prognosis Or Diagnosis (TRIPOD) type 3 study. The model was validated against patients treated with neoadjuvant chemoradiotherapy according to the Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for oesophageal or junctional cancer (CROSS) trial regimen using pre- and post-harmonised image features. The Kaplan–Meier method with log-rank significance tests assessed risk strata discrimination. A Cox proportional hazards model assessed model calibration. Primary outcome was overall survival (OS). ResultsBetween 2010 and 2015, 449 patients were included in the development (n = 302), internal validation (n = 101) and external validation (n = 46) cohorts. No statistically significant difference in OS between patient quartiles was demonstrated in prognostic models incorporating PET image features (X2 = 1.42, df = 3, p = 0.70) or exclusively clinical variables (age, disease stage and treatment; X2 = 1.19, df = 3, p = 0.75). The calibration slope β of both models was not significantly different from unity (p = 0.29 and 0.29, respectively). Risk groups defined using only clinical variables suggested differences in OS, although these were not statistically significant (X2 = 0.71, df = 2, p = 0.70). ConclusionThe prognostic model did not enable significant discrimination between the validation risk groups, but a second model with exclusively clinical variables suggested some transferable prognostic ability. PET harmonisation did not significantly change the results of model validation.

Early Metabolic Response at Mid-Radiation Therapy FDG-PET Imaging Predicts Patterns of Treatment Failure in Locally Advanced Oropharyngeal Cancer

Imaging with 18-fluorodeoxyglucose-positron emission tomography (FDG-PET) is used for staging, radiation treatment planning, and evaluation of therapy response. The purpose of this work was to assess early metabolic response at mid-radiotherapy (RT) FDG-PET imaging obtained at 30-36 Gy and investigate its ability to predict patterns of treatment failure in locally advanced oropharyngeal cancer. We retrospectively collected data for oropharyngeal squamous cell carcinoma patients who were treated with definitive chemoradiotherapy from 2009 to 2015 at our institution and who have available pre- and mid-RT PET-CT scans. Three types of PET image features were analyzed, including SUV maximum, metabolic tumor volume (defined at a threshold of 50% of SUVmax), and total lesion glycolysis. All three features were calculated on both pre- and mid-RT PET scans. Additionally, the longitudinal change between the two scans was also computed for all image features. Cox proportional hazard regression analysis was used to evaluate the prognostic value of individual PET image features for predicting locoregional recurrence and distant metastasis. We further adjusted for clinicopathologic factors including tumor and nodal stage, HPV and smoking status in multivariate Cox regression analysis A total of 104 patients with a median follow-up of 30 months were analyzed. On univariate analysis, mid-RT MTV was the only PET imaging parameter associated with locoregional recurrence (hazard ratio [HR] = 9.6, 95% confidence interval [CI]: 1.2-76.6, P = 0.0005). On multivariate analysis, mid-RT MTV remained a significant predictor (HR = 1.5; 95% CI: 1.02-2.3, P = 0.039) for locoregional recurrence, in addition to nodal stage (HR = 7.6; 95% CI: 1.2-49.0, P = 0.032), HPV positivity (HR = 0.14; 95% CI: 0.03-0.79, P = 0.025). Both pre-RT SUVmax and the change between SUVmax pre- and mid-RT scans were associated with distant metastasis (HR = 4.1, 95% CI: 1.2-14.4, P = 0.016, HR = 3.4, 95% CI: 1.1-10.5, P = 0.026, respectively) on univariate analysis. However, there were no significant predictors for distant metastasis on multivariate analysis. MTV at mid-RT PET was an independent predictor of locoregional recurrence. If validated, it may aid with individualized risk-adaptive therapy for oropharyngeal cancer.

A multiobjective Bayesian networks approach for joint prediction of tumor local control and radiation pneumonitis in nonsmall-cell lung cancer (NSCLC) for response-adapted radiotherapy.

Individualization of therapeutic outcomes in NSCLC radiotherapy is likely to be compromised by the lack of proper balance of biophysical factors affecting both tumor local control (LC) and side effects such as radiation pneumonitis (RP), which are likely to be intertwined. Here, we compare the performance of separate and joint outcomes predictions for response-adapted personalized treatment planning. A total of 118 NSCLC patients treated on prospective protocols with 32 cases of local progression and 20 cases of RP grade 2 or higher (RP2) were studied. Sixty-eight patients with 297 features before and during radiotherapy were used for discovery and 50 patients were reserved for independent testing. A multiobjective Bayesian network (MO-BN) approach was developed to identify important features for joint LC/RP2 prediction using extended Markov blankets as inputs to develop a BN predictive structure. Cross-validation (CV) was used to guide the MO-BN structure learning. Area under the free-response receiver operating characteristic (AU-FROC) curve was used to evaluate joint prediction performance. Important features including single nucleotide polymorphisms (SNPs), micro RNAs, pretreatment cytokines, pretreatment PET radiomics together with lung and tumor gEUDs were selected and their biophysical inter-relationships with radiation outcomes (LC and RP2) were identified in a pretreatment MO-BN. The joint LC/RP2 prediction yielded an AU-FROC of 0.80 (95% CI: 0.70-0.86) upon internal CV. This improved to 0.85 (0.75-0.91) with additional two SNPs, changes in one cytokine and two radiomics PET image features through the course of radiotherapy in a during-treatment MO-BN. This MO-BN model outperformed combined single-objective Bayesian networks (SO-BNs) during-treatment [0.78 (0.67-0.84)]. AU-FROC values in the evaluation of the MO-BN and individual SO-BNs on the testing dataset were 0.77 and 0.68 for pretreatment, and 0.79 and 0.71 for during-treatment, respectively. MO-BNs can reveal possible biophysical cross-talks between competing radiotherapy clinical endpoints. The prediction is improved by providing additional during-treatment information. The developed MO-BNs can be an important component of decision support systems for personalized response-adapted radiotherapy.

Significance of the impact of motion compensation on the variability of PET image features

In lung cancer, quantification by positron emission tomography/computed tomography (PET/CT) imaging presents challenges due to respiratory movement. Our primary aim was to study the impact of motion compensation implied by retrospectively gated (4D)-PET/CT on the variability of PET quantitative parameters. Its significance was evaluated by comparison with the variability due to (i) the voxel size in image reconstruction and (ii) the voxel size in image post-resampling. The method employed for feature extraction was chosen based on the analysis of (i) the effect of discretization of the standardized uptake value (SUV) on complementarity between texture features (TF) and conventional indices, (ii) the impact of the segmentation method on the variability of image features, and (iii) the variability of image features across the time-frame of 4D-PET. Thirty-one PET-features were involved. Three SUV discretization methods were applied: a constant width (SUV resolution) of the resampling bin (method RW), a constant number of bins (method RN) and RN on the image obtained after histogram equalization (method EqRN). The segmentation approaches evaluated were 40 of SUVmax and the contrast oriented algorithm (COA). Parameters derived from 4D-PET images were compared with values derived from the PET image obtained for (i) the static protocol used in our clinical routine (3D) and (ii) the 3D image post-resampled to the voxel size of the 4D image and PET image derived after modifying the reconstruction of the 3D image to comprise the voxel size of the 4D image. Results showed that TF complementarity with conventional indices was sensitive to the SUV discretization method. In the comparison of COA and 40 contours, despite the values not being interchangeable, all image features showed strong linear correlations (r > 0.91, ). Across the time-frames of 4D-PET, all image features followed a normal distribution in most patients. For our patient cohort, the compensation of tumor motion did not have a significant impact on the quantitative PET parameters. The variability of PET parameters due to voxel size in image reconstruction was more significant than variability due to voxel size in image post-resampling. In conclusion, most of the parameters (apart from the contrast of neighborhood matrix) were robust to the motion compensation implied by 4D-PET/CT. The impact on parameter variability due to the voxel size in image reconstruction and in image post-resampling could not be assumed to be equivalent.

Abstract P5-01-03: Stratifying triple-negative breast cancer prognosis using 18F-FDG-PET/CT imaging

Abstract Introduction: Triple-negative breast cancer (TNBC) is a highly diverse group of cancers, and may benefit from molecular-targeted therapies. This study aims to stratify prognosis of TNBC patients using pre-treatment 18F-FDG PET/CT, alone and with correlation to immunohistochemistry biomarkers. Method: 200 consecutive TNBC breast cancer patients treated between 2008 and 2012 who received lumpectomy or mastectomy as primary treatment were retrieved. Among the full cohort, 79 patients had pre-treatment 18F FDG PET/CT scans. Immunostaining status (percentage and intensity) of basal biomarkers (EGFR, CK5/6), Ki-67, P53, and other clinicopathological variables (age, tumor size, pathological T/N stage, nuclear grade, and lymph node metastasis) were obtained. Three PET image features were evaluated: maximum uptake values (SUVmax), mean uptake (SUVmean) and target volume (SUVvol) defined by SUV&amp;gt;2.5. The relationships among tumor metabolic activities and clinicopathological factors were evaluated. All variables were analyzed versus disease-free survival (DFS) using univariate and multivariate Cox analysis, Kaplan-Meier curves and log-rank tests. The optimal cutoff points of variables were estimated using time-dependent survival receiver operating characteristic (ROC) analysis. Results: All PET features significantly correlated with proliferation marker Ki-67 (all p&amp;lt;0.010). SUVmax stratified the prognosis of TNBC patients with optimal cutoff derived by ROC analysis (≤3.5 vs &amp;gt;3.5, AUC=0.654, p=0.006). Basal biomarkers EGFR and CK5/6 and image features SUVmax, SUVmean, SUVvol were significant associated with DFS in univariate Cox analysis, whereas SUVmax (p=0.001) and EGFR (p=0.001) were also significant in multivariate Cox analysis. To integrate prognosis of biological and imaging markers, patients were first stratified by EGFR into low (≤15%) and high (&amp;gt;15%) risk groups. Further, SUVmax was used as a variable to stratify the two EGFR groups. In the high EGFR group, patients with high FDG uptake (SUVmax&amp;gt;3.5) had worse survival outcome (median DFS=7.6 months) than those patients with low FDG uptake (SUVmax≤3.5, median DFS=11.6 months). In the low EGFR group, high SUVmax also indicated worse survival outcome (17.2 months) than low SUVmax (22.8 months). The risk stratification with integrative EGFR and PET was statistically significant with log-rank p&amp;lt;&amp;lt;0.001. Multivariate Cox analysis for disease-free survivalVariablesHR (95% CI)p-valuePathology, T stage, ≤ 3 vs &amp;gt;32.337(0.428-7.384)0.148EGFR, ≤15% vs &amp;gt; 15%9.109(1.997-41.55)0.004CK5/6, ≤ 50% vs &amp;gt; 50%1.471(0.598-3.614)0.401SUVmax, ≤3.5 vs &amp;gt; 3.53.883(1.13-13.32)0.031 TNBC patient risk groups stratified by EGFR and SUVmax (with the median values of variables)Risk groups (EGFR&amp;gt;15, SUVmax&amp;gt;3.5)patient#DFS monthsEGFR %SUVmaxSUVmeanSUVvolKi-67%1 (-, -)1222.852.00.60.2342 (-, +)1517.258.94.37.2673 (+, -)1311.6502.72.60.9354 (+, +)377.66011.35.210.960 Conclusions: Pre-treatment 18F-FDG PET/CT imaging has significant prognostic value for predicting survival outcome of TNBC patients. Integrated with basal-biomarker EGFR, PET imaging can further stratify patient risks in the pre-treatment stage, and help select appropriate treatment strategies for individual patients. Citation Format: Yue Y, Cui X, Bose S, Audeh W, Zhang X, Fraass B. Stratifying triple-negative breast cancer prognosis using 18F-FDG-PET/CT imaging. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P5-01-03.

Stratifying triple-negative breast cancer prognosis using 18F-FDG-PET/CT imaging.

This study aims to stratify prognosis of triple-negative breast cancer (TNBC) patients using pre-treatment 18F-FDG-PET/CT, alone and with correlation to immunohistochemistry biomarkers. 200 consecutive TNBC breast cancer patients treated between 2008 and 2012 were retrieved. Among the full cohort, 79 patients had pre-treatment 18F-FDG-PET/CT scans. Immunostaining status of basal biomarkers (EGFR, CK5/6) and other clinicopathological variables were obtained. Three PET image features were evaluated: maximum uptake values (SUVmax), mean uptake (SUVmean), and metabolic volume (SUVvol) defined by SUV > 2.5. All variables were analyzed versus disease-free survival (DFS) using univariate and multivariate Cox analysis, Kaplan-Meier curves, and log-rank tests. The optimal cutoff points of variables were estimated using time-dependent survival receiver operating characteristic (ROC) analysis. All PET features significantly correlated with proliferation marker Ki-67 (all p < 0.010). SUVmax stratified the prognosis of TNBC patients with optimal cutoff derived by ROC analysis (≤3.5 vs. >3.5, AUC = 0.654, p = 0.006). SUVmax and EGFR were significant prognostic factors in univariate and multivariate Cox analyses. To integrate prognosis of biological and imaging markers, patients were first stratified by EGFR into low (≤15 %) and high (>15 %) risk groups. Further, SUVmax was used as a variable to stratify the two EGFR groups. In the high EGFR group, patients with high FDG uptake (SUVmax > 3.5) had worse survival outcome (median DFS = 7.6 months) than those patients with low FDG uptake (SUVmax ≤ 3.5, median DFS = 11.6 months). In the low EGFR group, high SUVmax also indicated worse survival outcome (17.2 months) than low SUVmax (22.8 months). The risk stratification with integrative EGFR and PET was statistically significant with log-rank p ≪ 0.001. Pre-treatment 18F-FDG-PET/CT imaging has significant prognostic value for predicting survival outcome of TNBC patients. Integrated with basal-biomarker EGFR, PET imaging can further stratify patient risks in the pre-treatment stage and help select appropriate treatment strategies for individual patients.

MO‐AB‐BRA‐10: Cancer Therapy Outcome Prediction Based On Dempster‐Shafer Theory and PET Imaging

Purpose:In cancer therapy, utilizing FDG‐18 PET image‐based features for accurate outcome prediction is challenging because of 1) limited discriminative information within a small number of PET image sets, and 2) fluctuant feature characteristics caused by the inferior spatial resolution and system noise of PET imaging. In this study, we proposed a new Dempster‐Shafer theory (DST) based approach, evidential low‐dimensional transformation with feature selection (ELT‐FS), to accurately predict cancer therapy outcome with both PET imaging features and clinical characteristics.Methods:First, a specific loss function with sparse penalty was developed to learn an adaptive low‐rank distance metric for representing the dissimilarity between different patients’ feature vectors. By minimizing this loss function, a linear low‐dimensional transformation of input features was achieved. Also, imprecise features were excluded simultaneously by applying a l2,1‐norm regularization of the learnt dissimilarity metric in the loss function. Finally, the learnt dissimilarity metric was applied in an evidential K‐nearest‐neighbor (EK‐ NN) classifier to predict treatment outcome.Results:Twenty‐five patients with stage II–III non‐small‐cell lung cancer and thirty‐six patients with esophageal squamous cell carcinomas treated with chemo‐radiotherapy were collected. For the two groups of patients, 52 and 29 features, respectively, were utilized. The leave‐one‐out cross‐validation (LOOCV) protocol was used for evaluation. Compared to three existing linear transformation methods (PCA, LDA, NCA), the proposed ELT‐FS leads to higher prediction accuracy for the training and testing sets both for lung‐cancer patients (100+/−0.0, 88.0+/−33.17) and for esophageal‐cancer patients (97.46+/−1.64, 83.33+/−37.8). The ELT‐FS also provides superior class separation in both test data sets.Conclusion:A novel DST‐ based approach has been proposed to predict cancer treatment outcome using PET image features and clinical characteristics. A specific loss function has been designed for robust accommodation of feature set incertitude and imprecision, facilitating adaptive learning of the dissimilarity metric for the EK‐NN classifier.

SU‐E‐J‐275: Review ‐ Computerized PET/CT Image Analysis in the Evaluation of Tumor Response to Therapy

Purpose:To review the literature in using computerized PET/CT image analysis for the evaluation of tumor response to therapy.Methods:We reviewed and summarized more than 100 papers that used computerized image analysis techniques for the evaluation of tumor response with PET/CT. This review mainly covered four aspects: image registration, tumor segmentation, image feature extraction, and response evaluation.Results:Although rigid image registration is straightforward, it has been shown to achieve good alignment between baseline and evaluation scans. Deformable image registration has been shown to improve the alignment when complex deformable distortions occur due to tumor shrinkage, weight loss or gain, and motion. Many semi‐automatic tumor segmentation methods have been developed on PET. A comparative study revealed benefits of high levels of user interaction with simultaneous visualization of CT images and PET gradients. On CT, semi‐automatic methods have been developed for only tumors that show marked difference in CT attenuation between the tumor and the surrounding normal tissues. Quite a few multi‐modality segmentation methods have been shown to improve accuracy compared to single‐modality algorithms. Advanced PET image features considering spatial information, such as tumor volume, tumor shape, total glycolytic volume, histogram distance, and texture features have been found more informative than the traditional SUVmax for the prediction of tumor response. Advanced CT features, including volumetric, attenuation, morphologic, structure, and texture descriptors, have also been found advantage over the traditional RECIST and WHO criteria in certain tumor types. Predictive models based on machine learning technique have been constructed for correlating selected image features to response. These models showed improved performance compared to current methods using cutoff value of a single measurement for tumor response.Conclusion:This review showed that computerized PET/CT image analysis holds great potential to improve the accuracy in evaluation of tumor response.This work was supported in part by the National Cancer Institute Grant R01CA172638.

SU‐E‐QI‐20: A Review of Advanced PET and CT Image Features for the Evaluation of Tumor Response

Purpose:To review the literature in using quantitative PET and CT image features for the evaluation of tumor response.Methods:We reviewed and summarized more than fifty papers that use advanced, quantitative PET/CT image features for the evaluation of tumor response. We also discussed future works on extracting disease‐specific features, combining multiple and complementary features in response modeling, delineating tumor in multimodality images, and exploring biological explanations of these advanced features.Results:Advanced PET image features considering spatial information, such as tumor volume, tumor shape, total glycolytic volume, histogram distance, and texture features (characterizing spatial distribution of FDG uptake) have been found more informative than the traditional SUVmax for the prediction of tumor response. Advanced CT features, including volumetric, attenuation, morphologic, structure, and texture descriptors, have also been found advantage over the traditional RECIST and WHO criteria in certain tumor types.Conclusions:Advanced, quantitative FDG PET/CT image features have been shown promising for the evaluation of tumor response. With the emerging multi‐modality imaging performed at multiple time points for each patient, it becomes more important to analyze the serial images quantitatively, select and combine both complementary and contradictory information from various sources, for accurate and personalized evaluation of tumor response to therapy.