Spatial-temporal [¹⁸F]FDG-PET features for predicting pathologic response of esophageal cancer to neoadjuvant chemoradiation therapy

The effect of harmonization on the variability of PET radiomic features extracted using various segmentation methods

PURPOSE

This study aimed to examine the robustness of positron emission tomography (PET) radiomic features extracted via different segmentation methods before and after ComBat harmonization in patients with non-small cell lung cancer (NSCLC).

METHODS

We included 120 patients (positive recurrence = 46 and negative recurrence = 74) referred for PET scanning as a routine part of their care. All patients had a biopsy-proven NSCLC. Nine segmentation methods were applied to each image, including manual delineation, K-means (KM), watershed, fuzzy-C-mean, region-growing, local active contour (LAC), and iterative thresholding (IT) with 40, 45, and 50% thresholds. Diverse image discretizations, both without a filter and with different wavelet decompositions, were applied to PET images. Overall, 6741 radiomic features were extracted from each image (749 radiomic features from each segmented area). Non-parametric empirical Bayes (NPEB) ComBat harmonization was used to harmonize the features. Linear Support Vector Classifier (LinearSVC) with L1 regularization For feature selection and Support Vector Machine classifier (SVM) with fivefold nested cross-validation was performed using StratifiedKFold with 'n_splits' set to 5 to predict recurrence in NSCLC patients and assess the impact of ComBat harmonization on the outcome.

RESULTS

From 749 extracted radiomic features, 206 (27%) and 389 (51%) features showed excellent reliability (ICC ≥ 0.90) against segmentation method variation before and after NPEB ComBat harmonization, respectively. Among all, 39 features demonstrated poor reliability, which declined to 10 after ComBat harmonization. The 64 fixed bin widths (without any filter) and wavelets (LLL)-based radiomic features set achieved the best performance in terms of robustness against diverse segmentation techniques before and after ComBat harmonization. The first-order and GLRLM and also first-order and NGTDM feature families showed the largest number of robust features before and after ComBat harmonization, respectively. In terms of predicting recurrence in NSCLC, our findings indicate that using ComBat harmonization can significantly enhance machine learning outcomes, particularly improving the accuracy of watershed segmentation, which initially had fewer reliable features than manual contouring. Following the application of ComBat harmonization, the majority of cases saw substantial increase in sensitivity and specificity.

CONCLUSION

Radiomic features are vulnerable to different segmentation methods. ComBat harmonization might be considered a solution to overcome the poor reliability of radiomic features.

The role of baseline 2-[18 F]-FDG-PET/CT metrics and radiomics features in predicting primary gastric lymphoma diagnosis

Diffuse Large B-Cell Lymphomas (DLCBL) and mucosa-associated lymphoid tissue (MALT) are the two most common primary gastric lymphomas (PGLs), but have strongly different features. DLBCL is more aggressive, is frequently diagnosed at an advanced stage and has a poorer prognosis. The aim of this retrospective study was to explore the role of fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography (2-[18 F]-FDG-PET/CT) and radiomics features (RFs) in predicting the final diagnosis of patients with PGLs. Ninety-one patients with newly diagnosed PGLs who underwent pre-treatment 2-[18 F]-FDG-PET/CT were included. PET images were qualitatively and semi-quantitatively analyzed by deriving maximum standardized uptake value body weight (SUVbw), maximum standardized uptake value lean body mass (SUVlbm), maximum standardized uptake value body surface area (SUVbsa), lesion to liver SUVmax ratio (L-L SUV R), lesion to blood-pool SUVmax ratio (L-BP SUV R), metabolic tumor volume (gMTV) and total lesion glycolysis of gastric lesion (gTLG), total MTV (tMTV), TLG, and first-order RFs (histogram-related and shape related). Receiver-operating characteristic (ROC) curve analyses were performed to determine the differential diagnostic values of PET parameters. The final diagnosis was DLBCL in 54 (59%) cases and MALT in 37 cases (41%). PGLs showed FDG avidity in 83 cases (90%), 54/54 of DLBCL and 29/37 of MALT. All PET/CT metabolic features, such as stage of disease and tumor size, were significantly higher in DLBCL than MALT; while the presence of H. Pylori infection was more common in MALT. At univariate analysis, all PET/CT metrics were significantly higher in DLBCL than MALT lymphomas, while among RFs only Shape volume_vx and Shape sphericity showed a significant difference between the two groups. In conclusion we demonstrated that 2-[18 F]-FDG-PET/CT parameters can potentially discriminate between DLBCL and MALT lymphomas with high accuracy. Among first-order RFs, only Shape volume_vx and Shape sphericity helped in the differential diagnosis.

Feasibility analysis of arterial CT radiomics model to predict the risk of local and metastatic recurrence after radical cystectomy for bladder cancer

PURPOSE

To construct a radiomics-clinical nomogram model for predicting the risk of local and metastatic recurrence within 3 years after radical cystectomy (RC) of bladder cancer (BCa) based on the radiomics features and important clinical risk factors for arterial computed tomography (CT) images and to evaluate its efficacy.

METHODS

Preoperative CT datasets of 134 BCa patients (24 recurrent) who underwent RC were collected and divided into training (n = 93) and validation sets (n = 41). Radiomics features were extracted from a 1.5 mm CT layer thickness image in the arterial phase. A radiomics score (Rad-Score) model was constructed using the feature dimension reduction method and a logistic regression model. Combined with important clinical factors, including gender, age, tumor size, tumor number and grade, pathologic T stage, lymph node stage and histology type of the archived lesion, and CT image signs, a radiomics-clinical nomogram was developed, and its performance was evaluated in the training and validation sets. Decision curve analyses (DCA) the potential clinical usefulness.

RESULTS

The radiomics model is finally linear combined by 8 features screened by LASSO regression, and after coefficient weighting, achieved good predictive results. The radiomics nomogram developed by combining two independent predictors, Rad-Score and pathologic T stage, was developed in the training set [AUC, 0.840; 95% confidence interval (CI) 0.743-0.937] and validation set (AUC, 0.883; 95% CI 0.777-0.989). The calibration curve showed good agreement between the predicted probability of the radiomics-clinical model and the actual recurrence rate within 3 years after RC for BCa. DCA show the clinical application value of the radiomics-clinical model.

CONCLUSION

The radiomics-clinical nomogram model constructed based on the radiomics features of arterial CT images and important clinical risk factors is potentially feasible for predicting the risk of recurrence within 3 years after RC for BCa.

Radiomics and dosiomics for predicting complete response to definitive chemoradiotherapy patients with oesophageal squamous cell cancer using the hybrid institution model

OBJECTIVES

To develop a multi-institutional prediction model to estimate the local response to oesophageal squamous cell carcinoma (ESCC) treated with definitive radiotherapy based on radiomics and dosiomics features.

METHODS

The local responses were categorised into two groups (incomplete and complete). An external validation model and a hybrid model that the patients from two institutions were mixed randomly were proposed. The ESCC patients at stages I-IV who underwent chemoradiotherapy from 2012 to 2017 and had follow-up duration of more than 5 years were included. The patients who received palliative or pre-operable radiotherapy and had no FDG PET images were excluded. The segmentations included the GTV, CTV, and PTV which are used in treatment planning. In addition, shrinkage, expansion, and shell regions were created. Radiomic and dosiomic features were extracted from CT, FDG PET images, and dose distribution. Machine learning-based prediction models were developed using decision tree, support vector machine, k-nearest neighbour (kNN) algorithm, and neural network (NN) classifiers.

RESULTS

A total of 116 and 26 patients enrolled at Centre 1 and Centre 2, respectively. The external validation model exhibited the highest accuracy with 65.4% for CT-based radiomics, 77.9% for PET-based radiomics, and 72.1% for dosiomics based on the NN classifiers. The hybrid model exhibited the highest accuracy of 84.4% for CT-based radiomics based on the kNN classifier, 86.0% for PET-based radiomics, and 79.0% for dosiomics based on the NN classifiers.

CONCLUSION

The proposed hybrid model exhibited promising predictive performance for the local response to definitive radiotherapy in ESCC patients.

CLINICAL RELEVANCE STATEMENT

The prediction of the complete response for oesophageal cancer patients may contribute to improving overall survival. The hybrid model has the potential to improve prediction performance than the external validation model that was conventionally proposed.

KEY POINTS

• Radiomics and dosiomics used to predict response in patients with oesophageal cancer receiving definitive radiotherapy. • Hybrid model with neural network classifier of PET-based radiomics improved prediction accuracy by 8.1%. • The hybrid model has the potential to improve prediction performance.

Preoperative 18F-FDG PET/CT and CT radiomics for identifying aggressive histopathological subtypes in early stage lung adenocarcinoma

Lung adenocarcinoma (ADC) is the most common non-small cell lung cancer. Surgical resection is the primary treatment for early-stage lung ADC while lung-sparing surgery is an alternative for non-aggressive cases. Identifying histopathologic subtypes before surgery helps determine the optimal surgical approach. Predominantly solid or micropapillary (MIP) subtypes are aggressive and associated with a higher likelihood of recurrence and metastasis and lower survival rates. This study aims to non-invasively identify these aggressive subtypes using preoperative 18F-FDG PET/CT and diagnostic CT radiomics analysis. We retrospectively studied 119 patients with stage I lung ADC and tumors ≤ 2 cm, where 23 had aggressive subtypes (18 solid and 5 MIPs). Out of 214 radiomic features from the PET/CT and CT scans and 14 clinical parameters, 78 significant features (3 CT and 75 PET features) were identified through univariate analysis and hierarchical clustering with minimized feature collinearity. A combination of Support Vector Machine classifier and Least Absolute Shrinkage and Selection Operator built predictive models. Ten iterations of 10-fold cross-validation (10 ×10-fold CV) evaluated the model. A pair of texture feature (PET GLCM Correlation) and shape feature (CT Sphericity) emerged as the best predictor. The radiomics model significantly outperformed the conventional predictor SUVmax (accuracy: 83.5% vs. 74.7%, p = 9e-9) and identified aggressive subtypes by evaluating FDG uptake in the tumor and tumor shape. It also demonstrated a high negative predictive value of 95.6% compared to SUVmax (88.2%, p = 2e-10). The proposed radiomics approach could reduce unnecessary extensive surgeries for non-aggressive subtype patients, improving surgical decision-making for early-stage lung ADC patients.

The gap before real clinical application of imaging-based machine-learning and radiomic models for chemoradiation outcome prediction in esophageal cancer: a systematic review and meta-analysis

BACKGROUND

Due to tumoral heterogeneity and the lack of robust biomarkers, the prediction of chemoradiotherapy response and prognosis in patients with esophageal cancer (EC) is challenging. The goal of this study was to assess the study quality and clinical value of machine learning and radiomic-based quantitative imaging studies for predicting the outcomes of EC patients after chemoradiotherapy.

MATERIALS AND METHODS

PubMed, Embase, and Cochrane were searched for eligible articles. The methodological quality and risk of bias were evaluated using the Radiomics Quality Score (RQS), Image Biomarkers Standardization Initiative (IBSI) Guideline, and Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD) statement, as well as the modified Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. A meta-analysis of the evidence focusing on predicting chemoradiotherapy response and outcome in EC patients was implemented.

RESULTS

Forty-six studies were eligible for qualitative synthesis. The mean RQS score was 9.07, with an adherence rate of 42.52%. The adherence rates of the TRIPOD and IBSI were 61.70 and 43.17%, respectively. Ultimately, 24 studies were included in the meta-analysis, of which 16 studies had a pooled sensitivity, specificity, and area under the curve (AUC) of 0.83 (0.76-0.89), 0.83 (0.79-0.86), and 0.84 (0.81-0.87) in neoadjuvant chemoradiotherapy datasets, as well as 0.84 (0.75-0.93), 0.89 (0.83-0.93), and 0.93 (0.90-0.95) in definitive chemoradiotherapy datasets, respectively. Moreover, radiomics could distinguish patients from the low-risk and high-risk groups with different disease-free survival (DFS) (pooled hazard ratio: 3.43, 95% CI 2.39-4.92) and overall survival (pooled hazard ratio: 2.49, 95% CI 1.91-3.25). The results of subgroup and regression analyses showed that some of the heterogeneity was explained by the combination with clinical factors, sample size, and usage of the deep learning (DL) signature.

CONCLUSIONS

Noninvasive radiomics offers promising potential for optimizing treatment decision-making in EC patients. However, it is necessary to make scientific advancements in EC radiomics regarding reproducibility, clinical usefulness analysis, and open science categories. Improved model reporting of study objectives, blind assessment, and image processing steps are required to help promote real clinical applications of radiomics in EC research.

Performance of radiomics-based artificial intelligence systems in the diagnosis and prediction of treatment response and survival in esophageal cancer: a systematic review and meta-analysis of diagnostic accuracy

Radiomics can interpret radiological images with more detail and in less time compared to the human eye. Some challenges in managing esophageal cancer can be addressed by incorporating radiomics into image interpretation, treatment planning, and predicting response and survival. This systematic review and meta-analysis provides a summary of the evidence of radiomics in esophageal cancer. The systematic review was carried out using Pubmed, MEDLINE, and Ovid EMBASE databases-articles describing radiomics in esophageal cancer were included. A meta-analysis was also performed; 50 studies were included. For the assessment of treatment response using 18F-FDG PET/computed tomography (CT) scans, seven studies (443 patients) were included in the meta-analysis. The pooled sensitivity and specificity were 86.5% (81.1-90.6) and 87.1% (78.0-92.8). For the assessment of treatment response using CT scans, five studies (625 patients) were included in the meta-analysis, with a pooled sensitivity and specificity of 86.7% (81.4-90.7) and 76.1% (69.9-81.4). The remaining 37 studies formed the qualitative review, discussing radiomics in diagnosis, radiotherapy planning, and survival prediction. This review explores the wide-ranging possibilities of radiomics in esophageal cancer management. The sensitivities of 18F-FDG PET/CT scans and CT scans are comparable, but 18F-FDG PET/CT scans have improved specificity for AI-based prediction of treatment response. Models integrating clinical and radiomic features facilitate diagnosis and survival prediction. More research is required into comparing models and conducting large-scale studies to build a robust evidence base.

Characterization of Mediastinal Bulky Lymphomas with FDG-PET-Based Radiomics and Machine Learning Techniques

BACKGROUND

This study tested the diagnostic value of 18F-FDG PET/CT (FDG-PET) volumetric and texture parameters in the histological differentiation of mediastinal bulky disease due to classical Hodgkin lymphoma (cHL), primary mediastinal B-cell lymphoma (PMBCL) and grey zone lymphoma (GZL), using machine learning techniques.

METHODS

We reviewed 80 cHL, 29 PMBCL and 8 GZL adult patients with mediastinal bulky disease and histopathological diagnoses who underwent FDG-PET pre-treatment. Volumetric and radiomic parameters were measured using FDG-PET both for bulky lesions (BL) and for all lesions (AL) using LIFEx software (threshold SUV ≥ 2.5). Binary and multiclass classifications were performed with various machine learning techniques fed by a relevant subset of radiomic features.

RESULTS

The analysis showed significant differences between the lymphoma groups in terms of SUVmax, SUVmean, MTV, TLG and several textural features of both first- and second-order grey level. Among machine learning classifiers, the tree-based ensembles achieved the best performance both for binary and multiclass classifications in histological differentiation.

CONCLUSIONS

Our results support the value of metabolic heterogeneity as an imaging biomarker, and the use of radiomic features for early characterization of mediastinal bulky lymphoma.

Prognostic Value of [18F]-FDG PET/CT Radiomics Combined with Sarcopenia Status among Patients with Advanced Gastroesophageal Cancer

We investigated, whether 18[18F]-FDG PET/CT-derived radiomics combined with sarcopenia measurements improves survival prognostication among patients with advanced, metastatic gastroesophageal cancer. In our study, 128 consecutive patients with advanced, metastatic esophageal and gastroesophageal cancer (n = 128; 26 females; 102 males; mean age 63.5 ± 11.7 years; age range: 29−91 years) undergoing 18[18F]-FDG PET/CT for staging between November 2008 and December 2019 were included. Segmentation of the primary tumor and radiomics analysis derived from PET and CT images was performed semi-automatically with a commonly used open-source software platform (LIFEX, Version 6.30, lifexsoft.org). Patients’ nutritional status was determined by measuring the skeletal muscle index (SMI) at the level of L3 on the CT component. Univariable and multivariable analyses were performed to establish a survival prediction model including radiomics, clinical data, and SMI score. Univariable Cox proportional hazards model revealed ECOG (<0.001) and bone metastasis (p = 0.028) to be significant clinical parameters for overall survival (OS) and progression free survival (PFS). Age (p = 0.017) was an additional prognostic factor for OS. Multivariable analysis showed improved prognostication for overall and progression free survival when adding sarcopenic status, PET and CT radiomics to the model with clinical parameters only. PET and CT radiomics derived from hybrid 18[18F]-FDG PET/CT combined with sarcopenia measurements and clinical parameters may improve survival prediction among patients with advanced, metastatic gastroesophageal cancer.

Impact of image quality on radiomics applications

Radiomics features extracted from medical images have been widely reported to be useful in the patient specific outcome modeling for variety of assessment and prediction purposes. Successful application of radiomics features as imaging biomarkers, however, is dependent on the robustness of the approach to the variation in each step of the modeling workflow. Variation in the input image quality is one of the main sources that impacts the reproducibility of radiomics analysis when a model is applied to broader range of medical imaging data. The quality of medical image is generally affected by both the scanner related factors such as image acquisition/reconstruction settings and the patient related factors such as patient motion. This article aimed to review the published literatures in this field that reported the impact of various imaging factors on the radiomics features through the change in image quality. The literatures were categorized by different imaging modalities and also tabulated based on the imaging parameters and the class of radiomics features included in the study. Strategies for image quality standardization were discussed based on the relevant literatures and recommendations for reducing the impact of image quality variation on the radiomics in multi-institutional clinical trial were summarized at the end of this article.

Combined 18 F-FDG PET/CT Radiomics and Sarcopenia Score in Predicting Relapse-Free Survival and Overall Survival in Patients With Esophagogastric Cancer

PURPOSE

The aim of this study was to determine if radiomic features combined with sarcopenia measurements on pretreatment 18 F-FDG PET/CT can improve outcome prediction in surgically treated adenocarcinoma esophagogastric cancer patients.

PATIENTS AND METHODS

One hundred forty-five esophageal adenocarcinoma patients with curative therapeutic intent and available pretreatment 18 F-FDG PET/CT were included. Textural features from PET and CT images were evaluated using LIFEx software ( lifexsoft.org ). Sarcopenia measurements were done by measuring the Skeletal Muscle Index at L3 level on the CT component. Univariable and multivariable analyses were conducted to create a model including the radiomic parameters, clinical features, and Skeletal Muscle Index score to predict patients' outcome.

RESULTS

In multivariable analysis, we combined clinicopathological parameters including ECOG, surgical T, and N staging along with imaging derived sarcopenia measurements and radiomic features to build a predictor model for relapse-free survival and overall survival. Overall, adding sarcopenic status to the model with clinical features only (likelihood ratio test P = 0.03) and CT feature ( P = 0.0037) improved the model fit for overall survival. Similarly, adding sarcopenic status ( P = 0.051), CT feature ( P = 0.042), and PET feature ( P = 0.011) improved the model fit for relapse-free survival.

CONCLUSIONS

PET and CT radiomics derived from combined PET/CT integrated with clinicopathological parameters and sarcopenia measurement might improve outcome prediction in patients with nonmetastatic esophagogastric adenocarcinoma.

Radiomic assessment of oesophageal adenocarcinoma: a critical review of 18F-FDG PET/CT, PET/MRI and CT

Objectives

Radiomic models present an avenue to improve oesophageal adenocarcinoma assessment through quantitative medical image analysis. However, model selection is complicated by the abundance of available predictors and the uncertainty of their relevance and reproducibility. This analysis reviews recent research to facilitate precedent-based model selection for prospective validation studies.

Methods

This analysis reviews research on 18F-FDG PET/CT, PET/MRI and CT radiomics in oesophageal adenocarcinoma between 2016 and 2021. Model design, testing and reporting are evaluated according to the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) score and Radiomics Quality Score (RQS). Key results and limitations are analysed to identify opportunities for future research in the area.

Results

Radiomic models of stage and therapeutic response demonstrated discriminative capacity, though clinical applications require greater sensitivity. Although radiomic models predict survival within institutions, generalisability is limited. Few radiomic features have been recommended independently by multiple studies.

Conclusions

Future research must prioritise prospective validation of previously proposed models to further clinical translation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13244-022-01245-0.

Prediction of Non-Response to Neoadjuvant Chemoradiotherapy in Esophageal Cancer Patients with 18F-FDG PET Radiomics Based Machine Learning Classification

Background: Approximately 26% of esophageal cancer (EC) patients do not respond to neoadjuvant chemoradiotherapy (nCRT), emphasizing the need for pre-treatment selection. The aim of this study was to predict non-response using a radiomic model on baseline ¹⁸F-FDG PET. Methods: Retrospectively, 143 ¹⁸F-FDG PET radiomic features were extracted from 199 EC patients (T1N1-3M0/T2–4aN0-3M0) treated between 2009 and 2019. Non-response (n = 57; 29%) was defined as Mandard Tumor Regression Grade 4–5 (n = 44; 22%) or interval progression (n = 13; 7%). Randomly, 139 patients (70%) were allocated to explore all combinations of 24 feature selection strategies and 6 classification methods towards the cross-validated average precision (AP). The predictive value of the best-performing model, i.e AP and area under the ROC curve analysis (AUC), was evaluated on an independent test subset of 60 patients (30%). Results: The best performing model had an AP (mean ± SD) of 0.47 ± 0.06 on the training subset, achieved by a support vector machine classifier trained on five principal components of relevant clinical and radiomic features. The model was externally validated with an AP of 0.66 and an AUC of 0.67. Conclusion: In the present study, the best-performing model on pre-treatment ¹⁸F-FDG PET radiomics and clinical features had a small clinical benefit to identify non-responders to nCRT in EC.

Synergistic impact of motion and acquisition/reconstruction parameters on 18 F-FDG PET radiomic features in non-small cell lung cancer: phantom and clinical studies

Objectives

This study is aimed at examining the synergistic impact of motion and acquisition/reconstruction parameters on ¹⁸F‐FDG PET image radiomic features in non‐small cell lung cancer (NSCLC) patients, and investigating the robustness of features performance in differentiating NSCLC histopathology subtypes.

Methods

An in‐house developed thoracic phantom incorporating lesions with different sizes was used with different reconstruction settings, including various reconstruction algorithms, number of subsets and iterations, full‐width at half‐maximum of post‐reconstruction smoothing filter and acquisition parameters, including injected activity and test–retest with and without motion simulation. To simulate motion, a special motor was manufactured to simulate respiratory motion based on a normal patient in two directions. The lesions were delineated semi‐automatically to extract 174 radiomic features. All radiomic features were categorized according to the coefficient of variation (COV) to select robust features. A cohort consisting of 40 NSCLC patients with adenocarcinoma (n = 20) and squamous cell carcinoma (n = 20) was retrospectively analyzed. Statistical analysis was performed to discriminate robust features in differentiating histopathology subtypes of NSCLC lesions.

Results

Overall, 29% of radiomic features showed a COV ≤5% against motion. Forty‐five percent and 76% of the features showed a COV ≤ 5% against the test–retest with and without motion in large lesions, respectively. Thirty‐three percent and 45% of the features showed a COV ≤ 5% against different reconstruction parameters with and without motion, respectively. For NSCLC histopathological subtype differentiation, statistical analysis showed that 31 features were significant (p‐value < 0.05). Two out of the 31 significant features, namely, the joint entropy of GLCM (AUC = 0.71, COV = 0.019) and median absolute deviation of intensity histogram (AUC = 0.7, COV = 0.046), were robust against the motion (same reconstruction setting).

Conclusions

Motion, acquisition, and reconstruction parameters significantly impact radiomic features, just as their synergies. Radiomic features with high predictive performance (statistically significant) in differentiating histopathological subtype of NSCLC may be eliminated due to non‐reproducibility.

Histogram analysis of 18F-FDG PET imaging SUVs may predict the histologic grade of oral squamous cell carcinoma

OBJECTIVE

We tested the hypothesis that histogram analysis parameters of standardized uptake values (SUVs) obtained preoperatively using ¹⁸F-fluorodeoxyglucose positron emission tomography (¹⁸F-FDG PET) are significantly influenced by differences in metabolic capacity due to the histologic grade of oral squamous cell carcinoma (OSCC).

STUDY DESIGN

The study included 62 patients who were clinically diagnosed with OSCC and received surgical treatment after an ¹⁸F-FDG PET examination. Histogram analysis was performed using all voxels contained in the tumor area of each patient with an SUV ≥2.5. The histogram parameters calculated were the mean and standard deviation of SUVs, maximum SUV, metabolic tumor volume, skewness, and kurtosis. Statistical analyses were performed using a Mann-Whitney U test to calculate the significance of differences in these parameters between groups with well- and moderately- or poorly-differentiated tumors. Statistical significance was assumed at P < .05.

RESULTS

Only a comparison of kurtosis in the histogram showed a significant difference between the well- and moderately/poorly-differentiated tumors (P = .0294).

CONCLUSIONS

The distribution of metabolic capacity in oral squamous cell carcinoma tissues revealed on an ¹⁸F-FDG PET examination may help identify the histologic grade. This finding may provide valuable information for determining the subsequent treatment plan and predicting disease prognosis.

Predicting Response to Neoadjuvant Therapy in Oesophageal Adenocarcinoma

(1) Background: Oesophageal cancers are often late-presenting and have a poor 5-year survival rate. The standard treatment of oesophageal adenocarcinomas involves neoadjuvant chemotherapy with or without radiotherapy followed by surgery. However, less than one third of patients respond to neoadjuvant therapy, thereby unnecessarily exposing patients to toxicity and deconditioning. Hence, there is an urgent need for biomarkers to predict response to neoadjuvant therapy. This review explores the current biomarker landscape. (2) Methods: MEDLINE, EMBASE and ClinicalTrial databases were searched with key words relating to “predictive biomarker”, “neoadjuvant therapy” and “oesophageal adenocarcinoma” and screened as per the inclusion and exclusion criteria. All peer-reviewed full-text articles and conference abstracts were included. (3) Results: The search yielded 548 results of which 71 full-texts, conference abstracts and clinical trials were eligible for review. A total of 242 duplicates were removed, 191 articles were screened out, and 44 articles were excluded. (4) Discussion: Biomarkers were discussed in seven categories including imaging, epigenetic, genetic, protein, immunologic, blood and serum-based with remaining studies grouped in a miscellaneous category. (5) Conclusion: Although promising markers and novel methods have emerged, current biomarkers lack sufficient evidence to support clinical application. Novel approaches have been recommended to assess predictive potential more efficiently.

Methodological quality of machine learning-based quantitative imaging analysis studies in esophageal cancer: a systematic review of clinical outcome prediction after concurrent chemoradiotherapy

PURPOSE

Studies based on machine learning-based quantitative imaging techniques have gained much interest in cancer research. The aim of this review is to critically appraise the existing machine learning-based quantitative imaging analysis studies predicting outcomes of esophageal cancer after concurrent chemoradiotherapy in accordance with PRISMA guidelines.

METHODS

A systematic review was conducted in accordance with PRISMA guidelines. The citation search was performed via PubMed and Embase Ovid databases for literature published before April 2021. From each full-text article, study characteristics and model information were summarized. We proposed an appraisal matrix with 13 items to assess the methodological quality of each study based on recommended best-practices pertaining to quality.

RESULTS

Out of 244 identified records, 37 studies met the inclusion criteria. Study endpoints included prognosis, treatment response, and toxicity after concurrent chemoradiotherapy with reported discrimination metrics in validation datasets between 0.6 and 0.9, with wide variation in quality. A total of 30 studies published within the last 5 years were evaluated for methodological quality and we found 11 studies with at least 6 "good" item ratings.

CONCLUSION

A substantial number of studies lacked prospective registration, external validation, model calibration, and support for use in clinic. To further improve the predictive power of machine learning-based models and translate into real clinical applications in cancer research, appropriate methodologies, prospective registration, and multi-institution validation are recommended.

Immunotherapy treatment outcome prediction in metastatic melanoma through an automated multi-objective delta-radiomics model

Based on recent studies, immunotherapy led by immune checkpoint inhibitors has significantly improved the patient survival rate and effectively reduced the recurrence risk. However, immunotherapy has different therapeutic effects for different patients, leading to difficulties in predicting the treatment response. Conversely, delta-radiomic features, which measure the difference between pre- and post-treatment through quantitative image features, have proven to be promising descriptors for treatment outcome prediction. Consequently, we developed an effective model termed as the automated multi-objective delta-radiomics (Auto-MODR) model for the prediction of immunotherapy response in metastatic melanoma. In Auto-MODR, delta-radiomic features and traditional radiomic features were used as inputs. Furthermore, a novel automated multi-objective model was developed to obtain more reliable and balanced results between sensitivity and specificity. We conducted extensive comparisons with existing studies on treatment outcome prediction. Our method achieved an area under the curve (AUC) of 0.86 in a cross-validation study and an AUC of 0.73 in an independent study. Compared with the model using conventional radiomic features (pre- and post-treatment) only, better performance can be obtained when conventional radiomic and delta-radiomic features are combined. Furthermore, Auto-MODR outperformed the currently available radiomic strategies.

PET Imaging and Rate of Pathologic Complete Response in Esophageal Squamous Cell Carcinoma

BACKGROUND

For locally advanced esophageal squamous cell carcinoma (ESCC), chemoradiation (ChemoRT) followed by surgery offers the best chance of cure, with a 35-50% pathologic complete response (pCR) rate. Given the morbidity of esophagectomy and the possibility of pCR with ChemoRT, a 'watch and wait' strategy has been proposed, particularly for squamous cell carcinoma. The ability to accurately predict which patients will have pCR from ChemoRT is critical in treatment decision making. This study assessed positron emission tomography (PET) in predicting pCR after neoadjuvant ChemoRT for ESCC.

METHODS

ESCC patients treated with ChemoRT followed by surgery were identified. Maximum standard uptake value (SUV), metabolic tumor volume, total lesion glycolysis, and first-order textual features of standard deviation, kurtosis and skewness were measured from PET. Univariable and multivariable generalized linear method analyses were performed. A metabolic complete response (mCR) was defined as a post-therapy PET scan with maximum SUV < 4.0.

RESULTS

Twenty-seven patients underwent ChemoRT followed by surgery, with overall pCR seen in 11 (41%) patients and radiographic mCR seen in 12 (44%) patients. Final pathology for these 12 patients revealed pCR (ypT0N0M0) in 5 (42%) patients and persistent disease in 7 (58%) patients. Univariate analysis did not reveal PET parameters predictive of pCR.

CONCLUSION

Treatment of ESCC with ChemoRT often results in a robust clinical response. Among patients with an mCR after ChemoRT, disease persistence was found in 58%. The inability of PET to predict pCR is important in the context of a 'watch and wait' strategy for ESCC treated with ChemoRT.

A different overview of staging PET/CT images in patients with esophageal cancer: the role of textural analysis with machine learning methods

OBJECTIVE

This study evaluates the ability of several machine learning (ML) algorithms, developed using volumetric and texture data extracted from baseline ¹⁸F-FDG PET/CT studies performed initial staging of patient with esophageal cancer (EC), to predict survival and histopathology.

METHODS

The initial staging ¹⁸F-FDG PET/CT images obtained on newly diagnosed EC patients between January 2008 and June 2019 were evaluated using LIFEx software. A region of interest (ROI) of the primary tumor was created and volumetric and textural features were obtained. A significant relationship between these features and pathological subtypes, 1-year, and 5-year survival was investigated. Due to the nonhomogeneity of the data, nonparametric test (The Mann-Whitney U test) was used for each feature, in pairwise comparisons of independent variables. A p value of < 0.05 was considered significant. Receiver operating curve (ROC) analysis was performed for features with p < 0.05. Correlation between the significant features was evaluated with Spearman correlation test; features with correlation coefficient < 0.8 were evaluated with several ML algorithms.

RESULTS

In predicting survival in a 1-year follow-up J48 was obtained as the most successful algorithm (AUC: 0.581, PRC: 0.565, MCC: 0.258, acc: 64.29%). 5-year survival results were more promising than 1-year survival results with (AUC: 0.820, PRC: 0.860, MCC: 271, acc: 81.36%) by logistic regression. It is revealed that the most successful algorithm was naive bayes (AUC: 0.680 PRC: 0.776, MCC: 0.298, acc: 82.66%) in the histopathological discrimination.

CONCLUSION

Texture analysis with ML algorithms could be predictive of overall survival and discriminating histopathological subtypes of EC.

Evaluation and optimization of radiomics features stability to respiratory motion in 18 F-FDG 3D PET imaging

PURPOSE

To evaluate the impact of respiratory motion on radiomics features in ¹⁸ F-fluoro-2-deoxy-D-glucose three dimensional positron emission tomography (¹⁸ F-FDG 3D PET) imaging and optimize feature stability by combining preprocessing configurations and aggregation strategies.

METHODS

An in-house developed respiratory motion phantom was imaged in 3D PET scanner under nine respiratory patterns including one reference pattern. In total, 487 radiomics features were extracted for each respiratory pattern. Feature stability to respiratory motion was first evaluated by metrics of coefficient of variation (COV) and relative difference (RD) in a fixed preprocessing configuration. Further, one-way ANOVA and trend analysis were performed to evaluate the impact of preprocessing configuration (voxel size, discretization scheme) and aggregation strategy on feature stability. Finally, an optimization framework was proposed by selected feature-specific configurations with minimum COV value, and the diagnostic performance was validated in stable versus unstable features and fixed versus optimal features by a set of 46 patients with lung disease.

RESULTS

PET radiomics features were sensitive to respiratory motion, only 79/487 (16%) features were identified to be very stable in the fixed configuration. Preprocessing configuration and aggregation strategy had an impact on feature stability. For different voxel size, bin number, bin size and aggregation strategy, 188/487 (39%), 70/487 (15%), 148/487 (30%), and 38/95 (29%) features appeared significant changes in feature stability. The optimized configuration had the potential to improve feature stability compared to fixed configuration, with the COV decreased from 22% ±24% to 16% ±20%. Regarding the diagnostic performance, the stable and optimal configuration features outperformed the unstable and fixed configuration features, respectively (AUC 0.88, 0.87 vs. 0.83, 0.85).

CONCLUSIONS

Respiratory motion shows considerable impact on feature stability in 3D PET imaging, while optimizing preprocessing configuration may improve feature stability and diagnostic performance.

Using Genomics Feature Selection Method in Radiomics Pipeline Improves Prognostication Performance in Locally Advanced Esophageal Squamous Cell Carcinoma-A Pilot Study

PURPOSE

To evaluate the prognostic value of baseline and restaging CT-based radiomics with features associated with gene expression in esophageal squamous cell carcinoma (ESCC) patients receiving neoadjuvant chemoradiation (nCRT) plus surgery.

METHODS

We enrolled 106 ESCC patients receiving nCRT from two institutions. Gene expression profiles of 28 patients in the training set were used to detect differentially expressed (DE) genes between patients with and without relapse. Radiomic features that were correlated to DE genes were selected, followed by additional machine learning selection. A radiomic nomogram for disease-free survival (DFS) prediction incorporating the radiomic signature and prognostic clinical characteristics was established for DFS estimation and validated.

RESULTS

The radiomic signature with DE genes feature selection achieved better performance for DFS prediction than without. The nomogram incorporating the radiomic signature and lymph nodal status significantly stratified patients into high and low-risk groups for DFS (p < 0.001). The areas under the curve (AUCs) for predicting 5-year DFS were 0.912 in the training set, 0.852 in the internal test set, 0.769 in the external test set.

CONCLUSIONS

Genomics association was useful for radiomic feature selection. The established radiomic signature was prognostic for DFS. The radiomic nomogram could provide a valuable prediction for individualized long-term survival.

A Systematic Review of PET Textural Analysis and Radiomics in Cancer

Background: Although many works have supported the utility of PET radiomics, several authors have raised concerns over the robustness and replicability of the results. This study aimed to perform a systematic review on the topic of PET radiomics and the used methodologies. Methods: PubMed was searched up to 15 October 2020. Original research articles based on human data specifying at least one tumor type and PET image were included, excluding those that apply only first-order statistics and those including fewer than 20 patients. Each publication, cancer type, objective and several methodological parameters (number of patients and features, validation approach, among other things) were extracted. Results: A total of 290 studies were included. Lung (28%) and head and neck (24%) were the most studied cancers. The most common objective was prognosis/treatment response (46%), followed by diagnosis/staging (21%), tumor characterization (18%) and technical evaluations (15%). The average number of patients included was 114 (median = 71; range 20–1419), and the average number of high-order features calculated per study was 31 (median = 26, range 1–286). Conclusions: PET radiomics is a promising field, but the number of patients in most publications is insufficient, and very few papers perform in-depth validations. The role of standardization initiatives will be crucial in the upcoming years.

A priori prediction of response in multicentre locally advanced breast cancer (LABC) patients using quantitative ultrasound and derivative texture methods

PURPOSE

We develop a multi-centric response predictive model using QUS spectral parametric imaging and novel texture-derivate methods for determining tumour responses to neoadjuvant chemotherapy (NAC) prior to therapy initiation.

MATERIALS AND METHODS

QUS Spectroscopy provided parametric images of mid-band-fit (MBF), spectral-slope (SS), spectral-intercept (SI), average-scatterer-diameter (ASD), and average-acoustic-concentration (AAC) in 78 patients with locally advanced breast cancer (LABC) undergoing NAC. Ultrasound radiofrequency data were collected from Sunnybrook Health Sciences Center (SHSC), University of Texas MD Anderson Cancer Center (MD-ACC), and St. Michaels Hospital (SMH) using two different systems. Texture analysis was used to quantify heterogeneities of QUS parametric images. Further, a second-pass texture analysis was applied to obtain texture-derivate features. QUS, texture- and texture-derivate parameters were determined from both tumour core and a 5-mm tumour margin and were used in comparison to histopathological analysis for developing a response predictive model to classify responders versus non-responders. Model performance was assessed using leave-one-out cross-validation. Three standard classification algorithms including a linear discriminant analysis (LDA), k-nearest-neighbors (KNN), and support vector machines-radial basis function (SVM-RBF) were evaluated.

RESULTS

A combination of tumour core and margin classification resulted in a peak response prediction performance of 88% sensitivity, 78% specificity, 84% accuracy, 0.86 AUC, 84% PPV, and 83% NPV, achieved using the SVM-RBF classification algorithm. Other parameters and classifiers performed less well running from 66% to 80% accuracy.

CONCLUSIONS

A QUS-based framework and novel texture-derivative method enabled accurate prediction of responses to NAC. Multi-centric response predictive model provides indications of the robustness of the approach to variations due to different ultrasound systems and acquisition parameters.

Comparison of methods for texture analysis of QUS parametric images in the characterization of breast lesions

Purpose

Accurate and timely diagnosis of breast carcinoma is very crucial because of its high incidence and high morbidity. Screening can improve overall prognosis by detecting the disease early. Biopsy remains as the gold standard for pathological confirmation of malignancy and tumour grading. The development of diagnostic imaging techniques as an alternative for the rapid and accurate characterization of breast masses is necessitated. Quantitative ultrasound (QUS) spectroscopy is a modality well suited for this purpose. This study was carried out to evaluate different texture analysis methods applied on QUS spectral parametric images for the characterization of breast lesions.

Methods

Parametric images of mid-band-fit (MBF), spectral-slope (SS), spectral-intercept (SI), average scatterer diameter (ASD), and average acoustic concentration (AAC) were determined using QUS spectroscopy from 193 patients with breast lesions. Texture methods were used to quantify heterogeneities of the parametric images. Three statistical-based approaches for texture analysis that include Gray Level Co-occurrence Matrix (GLCM), Gray Level Run-length Matrix (GRLM), and Gray Level Size Zone Matrix (GLSZM) methods were evaluated. QUS and texture-parameters were determined from both tumour core and a 5-mm tumour margin and were used in comparison to histopathological analysis in order to classify breast lesions as either benign or malignant. We developed a diagnostic model using different classification algorithms including linear discriminant analysis (LDA), k-nearest neighbours (KNN), support vector machine with radial basis function kernel (SVM-RBF), and an artificial neural network (ANN). Model performance was evaluated using leave-one-out cross-validation (LOOCV) and hold-out validation.

Results

Classifier performances ranged from 73% to 91% in terms of accuracy dependent on tumour margin inclusion and classifier methodology. Utilizing information from tumour core alone, the ANN achieved the best classification performance of 93% sensitivity, 88% specificity, 91% accuracy, 0.95 AUC using QUS parameters and their GLSZM texture features.

Conclusions

A QUS-based framework and texture analysis methods enabled classification of breast lesions with >90% accuracy. The results suggest that optimizing method for extracting discriminative textural features from QUS spectral parametric images can improve classification performance. Evaluation of the proposed technique on a larger cohort of patients with proper validation technique demonstrated the robustness and generalization of the approach.

Pretreatment CT and 18 F-FDG PET-based radiomic model predicting pathological complete response and loco-regional control following neoadjuvant chemoradiation in oesophageal cancer

INTRODUCTION

To develop a radiomic-based model to predict pathological complete response (pCR) and outcome following neoadjuvant chemoradiotherapy (NACRT) in oesophageal cancer.

METHODS

We analysed 68 patients with oesophageal cancer treated with NACRT followed by esophagectomy, who had staging 18F-fluorodeoxyglucose (¹⁸ F-FDG) positron emission tomography (PET) and computed tomography (CT) scans performed at our institution. An in-house data-chjmirocterization algorithm was used to extract 3D-radiomic features from the segmented primary disease. Prediction models were constructed and internally validated. Composite feature, F_c  = α * F_PET  + (1 - α) * F_CT , 0 ≤ α ≤ 1, was constructed for each corresponding CT and PET feature. Loco-regional control (LRC), recurrence-free survival (RFS), metastasis-free survival (MFS) and overall survival (OS) were estimated by Kaplan-Meier analysis, and compared using log-rank test.

RESULTS

Median follow-up was 59 months. pCR was achieved in 34 (50%) patients. Five-year RFS, LRC, MFS and OS were 67.1%, 88.5%, 75.6% and 57.6%, respectively. Tumour Regression Grade (TRG) 0-1 indicative of complete response or minimal residual disease was significantly associated with improved 5-year LRC [93.7% vs 71.8%; P = 0.020; HR 0.19, 95% CI 0.04-0.85]. Four sepjmirote pCR predictive models were built for CT alone, PET alone, CT+PET and composite. CT, PET and CT+PET models had AUC 0.73 ± 0.08, 0.66 ± 0.08 and 0.77 ± 0.07, respectively. The composite model resulted in an improvement of pCR predicting power with AUC 0.87 ± 0.06. Stratifying patients with a low versus high radiomic score showed clinically relevant improvement in 5-year LRC favouring low-score group (91.1% vs. 80%, 95% CI 0.09-1.77, P = 0.2).

CONCLUSION

The composite CT/PET radiomics model was highly predictive of pCR following NACRT. Validation in larger data sets is warranted to determine whether the model can predict clinical outcomes.

18F-FDG PET/CT Metrics Are Correlated to the Pathological Response in Esophageal Cancer Patients Treated With Induction Chemotherapy Followed by Neoadjuvant Chemo-Radiotherapy

Background and Objective

The aim of this study was to assess the ability of Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography (¹⁸F-FDG PET/CT) to provide functional information useful in predicting pathological response to an intensive neoadjuvant chemo-radiotherapy (nCRT) protocol for both esophageal squamous cell carcinoma (SCC) and adenocarcinoma (ADC) patients.

Material and Methods

Esophageal carcinoma (EC) patients, treated in our Center between 2014 and 2018, were retrospectively reviewed. The nCRT protocol schedule consisted of an induction phase of weekly administered docetaxel, cisplatin, and 5-fluorouracil (TCF) for 3 weeks, followed by a concomitant phase of weekly TCF for 5 weeks with concurrent radiotherapy (50–50.4 Gy in 25–28 fractions). Three ¹⁸F-FDG PET/CT scans were performed: before (PET₁) and after (PET₂) induction chemotherapy (IC), and prior to surgery (PET₃). Correlation between PET parameters [maximum and mean standardized uptake value (SUV_max and SUV_mean), metabolic tumor volume (MTV), and total lesion glycolysis (TLG)], radiomic features and tumor regression grade (TGR) was investigated.

Results

Fifty-four patients (35 ADC, 19 SCC; 48 cT3/4; 52 cN+) were eligible for the analysis. Pathological response to nCRT was classified as major (TRG1-2, 41/54, 75.9%) or non-response (TRG3-4, 13/54, 24.1%). A major response was statistically correlated with SCC subtype (p = 0.02) and smaller tumor length (p = 0.03). MTV and TLG measured prior to IC (PET₁) were correlated to TRG1-2 response (p = 0.02 and p = 0.02, respectively). After IC (PET₂), SUV_mean and TLG correlated with major response (p = 0.03 and p = 0.04, respectively). No significance was detected when relative changes of metabolic parameters between PET₁ and PET₂ were evaluated. At textural quantitative analysis, three independent radiomic features extracted from PET₁ images ([JointEnergy and InverseDifferenceNormalized of GLCM and LowGrayLevelZoneEmphasis of GLSZM) were statistically correlated with major response (p < 0.0002).

Conclusions

¹⁸F-FDG PET/CT traditional metrics and textural features seem to predict pathologic response (TRG) in EC patients treated with induction chemotherapy followed by neoadjuvant chemo-radiotherapy. Further investigations are necessary in order to obtain a reliable predictive model to be used in the clinical practice.

Differentiating gastric cancer and gastric lymphoma using texture analysis (TA) of positron emission tomography (PET)

Background:

Texture analysis (TA) can quantify intra-tumor heterogeneity using standard medical images. The present study aimed to assess the application of positron emission tomography (PET) TA in the differential diagnosis of gastric cancer and gastric lymphoma.

Methods:

The pre-treatment PET images of 79 patients (45 gastric cancer, 34 gastric lymphoma) between January 2013 and February 2018 were retrospectively reviewed. Standard uptake values (SUVs), first-order texture features, and second-order texture features of the grey-level co-occurrence matrix (GLCM) were analyzed. The differences in features among different groups were analyzed by the two-way Mann-Whitney test, and receiver operating characteristic (ROC) analysis was used to estimate the diagnostic efficacy.

Results:

Inertia_GLCM was significantly lower in gastric cancer than that in gastric lymphoma (4975.61 vs. 11,425.30, z = −3.238, P = 0.001), and it was found to be the most discriminating texture feature in differentiating gastric lymphoma and gastric cancer. The area under the curve (AUC) of inertia_GLCM was higher than the AUCs of SUVmax and SUVmean (0.714 vs. 0.649 and 0.666, respectively). SUVmax and SUVmean were significantly lower in low-grade gastric lymphoma than those in high grade gastric lymphoma (3.30 vs. 11.80, 2.40 vs. 7.50, z = −2.792 and −3.007, P = 0.005 and 0.003, respectively). SUVs and first-order grey-level intensity features were not significantly different between low-grade gastric lymphoma and gastric cancer. Entropy_GLCM12 was significantly lower in low-grade gastric lymphoma than that in gastric cancer (6.95 vs. 9.14, z = −2.542, P = 0.011) and had an AUC of 0.770 in the ROC analysis of differentiating low-grade gastric lymphoma and gastric cancer.

Conclusions:

Inertia_GLCM and entropy_GLCM were the most discriminating features in differentiating gastric lymphoma from gastric cancer and low-grade gastric lymphoma from gastric cancer, respectively. PET TA can improve the differential diagnosis of gastric neoplasms, especially in tumors with similar degrees of fluorodeoxyglucose uptake.

Addition of HER2 and CD44 to 18F-FDG PET-based clinico-radiomic models enhances prediction of neoadjuvant chemoradiotherapy response in esophageal cancer

OBJECTIVES

To assess the complementary value of human epidermal growth factor receptor 2 (HER2)-related biological tumor markers to clinico-radiomic models in predicting complete response to neoadjuvant chemoradiotherapy (NCRT) in esophageal cancer patients.

METHODS

Expression of HER2 was assessed by immunohistochemistry in pre-treatment tumor biopsies of 96 patients with locally advanced esophageal cancer. Five other potentially active HER2-related biological tumor markers in esophageal cancer were examined in a sub-analysis on 43 patients. Patients received at least four of the five cycles of chemotherapy and full radiotherapy regimen followed by esophagectomy. Three reference clinico-radiomic models based on ¹⁸F-FDG PET were constructed to predict pathologic response, which was categorized into complete versus incomplete (Mandard tumor regression grade 1 vs. 2-5). The complementary value of the biological tumor markers was evaluated by internal validation through bootstrapping.

RESULTS

Pathologic examination revealed 21 (22%) complete and 75 (78%) incomplete responders. HER2 and cluster of differentiation 44 (CD44), analyzed in the sub-analysis, were univariably associated with pathologic response. Incorporation of HER2 and CD44 into the reference models improved the overall performance (R²s of 0.221, 0.270, and 0.225) and discrimination AUCs of 0.759, 0.857, and 0.816. All models exhibited moderate to good calibration. The remaining studied biological tumor markers did not yield model improvement.

CONCLUSIONS

Incorporation of HER2 and CD44 into clinico-radiomic prediction models improved NCRT response prediction in esophageal cancer. These biological tumor markers are promising in initial response evaluation.

KEY POINTS

• A multimodality approach, integrating independent genomic and radiomic information, is promising to improve prediction of γpCR in patients with esophageal cancer. • HER2 and CD44 are potential biological tumor markers in the initial work-up of patients with esophageal cancer. • Prediction models combining ¹⁸F-FDG PET radiomic features with HER2 and CD44 may be useful in the decision to omit surgery after neoadjuvant chemoradiotherapy in patients with esophageal cancer.

Assessment of Intratumoral and Peritumoral Computed Tomography Radiomics for Predicting Pathological Complete Response to Neoadjuvant Chemoradiation in Patients With Esophageal Squamous Cell Carcinoma

IMPORTANCE

For patients with locally advanced esophageal squamous cell carcinoma, neoadjuvant chemoradiation has been shown to improve long-term outcomes, but the treatment response varies among patients. Accurate pretreatment prediction of response remains an urgent need.

OBJECTIVE

To determine whether peritumoral radiomics features derived from baseline computed tomography images could provide valuable information about neoadjuvant chemoradiation response and enhance the ability of intratumoral radiomics to estimate pathological complete response.

DESIGN, SETTING, AND PARTICIPANTS

A total of 231 patients with esophageal squamous cell carcinoma, who underwent baseline contrast-enhanced computed tomography and received neoadjuvant chemoradiation followed by surgery at 2 institutions in China, were consecutively included. This diagnostic study used single-institution data between April 2007 and December 2018 to extract radiomics features from intratumoral and peritumoral regions and established intratumoral, peritumoral, and combined radiomics models using different classifiers. External validation was conducted using independent data collected from another hospital during the same period. Radiogenomics analysis using gene expression profile was done in a subgroup of the training set for pathophysiological explanation. Data were analyzed from June to December 2019.

EXPOSURES

Computed tomography-based radiomics.

MAIN OUTCOMES AND MEASURES

The discriminative performances of radiomics models were measured by area under the receiver operating characteristic curve.

RESULTS

Among the 231 patients included (192 men [83.1%]; mean [SD] age, 59.8 [8.7] years), the optimal intratumoral and peritumoral radiomics models yielded similar areas under the receiver operating characteristic curve of 0.730 (95% CI, 0.609-0.850) and 0.734 (0.613-0.854), respectively. The combined model was composed of 7 intratumoral and 6 peritumoral features and achieved better discriminative performance, with an area under the receiver operating characteristic curve of 0.852 (95% CI, 0.753-0.951), accuracy of 84.3%, sensitivity of 90.3%, and specificity of 79.5% in the test set. Gene sets associated with the combined model mainly involved lymphocyte-mediated immunity. The association of peritumoral area with response identification might be partially attributed to type I interferon-related biological process.

CONCLUSIONS AND RELEVANCE

A combination of peritumoral radiomics features appears to improve the predictive performance of intratumoral radiomics to estimate pathological complete response after neoadjuvant chemoradiation in patients with esophageal squamous cell carcinoma. This study underlines the significant application of peritumoral radiomics to assess treatment response in clinical practice.

Characterization of FDG PET Images Using Texture Analysis in Tumors of the Gastro-Intestinal Tract: A Review

Radiomics or textural feature extraction obtained from positron emission tomography (PET) images through complex mathematical models of the spatial relationship between multiple image voxels is currently emerging as a new tool for assessing intra-tumoral heterogeneity in medical imaging. In this paper, available literature on texture analysis using FDG PET imaging in patients suffering from tumors of the gastro-intestinal tract is reviewed. While texture analysis of FDG PET images appears clinically promising, due to the lack of technical specifications, a large variability in the implemented methodology used for texture analysis and lack of statistical robustness, at present, no firm conclusions can be drawn regarding the predictive or prognostic value of FDG PET texture analysis derived indices in patients suffering from gastro-enterologic tumors. In order to move forward in this field, a harmonized image acquisition and processing protocol as well as a harmonized protocol for texture analysis of tumor volumes, allowing multi-center studies excluding statistical biases should be considered. Furthermore, the complementary and additional value of CT-imaging, as part of the PET/CT imaging technique, warrants exploration.

Residual disease after neoadjuvant chemoradiotherapy for oesophageal cancer: locations undetected by endoscopic biopsies in the preSANO trial

Background

Active surveillance has been proposed for patients with oesophageal cancer in whom there is a complete clinical response after neoadjuvant chemoradiotherapy (nCRT). However, endoscopic biopsies have limited negative predictive value in detecting residual disease. This study determined the location of residual tumour following surgery to improve surveillance and endoscopic strategies.

Methods

The present study was based on patients who participated in the prospective preSANO trial with adenocarcinoma or squamous cell carcinoma of the oesophagus or oesophagogastric junction treated in four Dutch hospitals between 2013 and 2016. Resection specimens and endoscopic biopsies taken during clinical response evaluations after nCRT were reviewed by two expert gastrointestinal pathologists. The exact location of residual disease in the oesophageal wall was determined in resection specimens. Endoscopic biopsies were assessed for the presence of structures representing the submucosal layer of the oesophageal wall.

Results

In total, 119 eligible patients underwent clinical response evaluations after nCRT followed by standard surgery. Residual tumour was present in endoscopic biopsies from 70 patients, confirmed on histological analysis of the resected organ. Residual tumour was present in the resection specimen from 27 of the other 49 patients, despite endoscopic biopsies being negative. Of these 27 patients, residual tumour was located in the mucosa in 18, and in the submucosa beneath tumour‐free mucosa in eight. One patient had tumour in muscle beneath tumour‐free mucosa and submucosa.

Conclusion

Most residual disease after nCRT missed by endoscopic biopsies was located in the mucosa. Active surveillance could be improved by more sampling and considering submucosal biopsies.

Breast lesion characterization using Quantitative Ultrasound (QUS) and derivative texture methods

Purpose

Accurate and timely diagnosis of breast cancer is extremely important because of its high incidence and high morbidity. Early diagnosis of breast cancer through screening can improve overall prognosis. Currently, biopsy remains as the gold standard for tumor pathological confirmation. Development of diagnostic imaging techniques for rapid and accurate characterization of breast lesions is required. We aim to evaluate the usefulness of texture-derivate features of QUS spectral parametric images for non-invasive characterization of breast lesions.

Methods

QUS Spectroscopy was used to determine parametric images of mid-band fit (MBF), spectral slope (SS), spectral intercept (SI), average scatterer diameter (ASD), and average acoustic concentration (AAC) in 204 patients with suspicious breast lesions. Subsequently, texture analysis techniques were used to generate texture maps from parametric images to quantify heterogeneities of QUS parametric images. Further, a second-pass texture analysis was applied to obtain texture-derivate features. QUS parameters, texture-parameters and texture-derivate parameters were determined from both tumor core and a 5-mm tumor margin and were used in comparison to histopathological analysis in order to develop a diagnostic model for classifying breast lesions as either benign or malignant. Both leave-one-out and hold-out cross-validations were used to evaluate the performance of the diagnostic model. Three standard classification algorithms including a linear discriminant analysis (LDA), k-nearest neighbors (KNN), and support vector machines-radial basis function (SVM-RBF) were evaluated.

Results

Core and margin information using the SVM-RBF attained the best classification performance of 90% sensitivity, 92% specificity, 91% accuracy, and 0.93 AUC utilizing QUS parameters and their texture derivatives, evaluated using leave-one-out cross-validation. Implementation of hold-out cross-validation using combination of both core and margin information and SVM-RBF achieved average accuracy and AUC of 88% and 0.92, respectively.

Conclusions

QUS-based framework and derivative texture methods enable accurate classification of breast lesions. Evaluation of the proposed technique on a large cohort using hold-out cross-validation demonstrates its robustness and its generalization.

Predicting response to radiotherapy in tumors with PET/CT: when and how?

Radiotherapy is one of the main methods for tumor treatment, with the improved radiotherapy delivery technique to combat cancer, there is a growing interest for finding effective and feasible ways to predict tumor radiosensitivity. Based on a series of changes in metabolism, microvessel density, hypoxic microenvironment, and cytokines of tumors after radiotherapy, a variety of radiosensitivity detection methods have been studied. Among the detection methods, positron emission tomography-computed tomography (PET/CT) is a feasible tool for response evaluation following definitive radiotherapy for cancers with a high negative predictive value. The prognostic or predictive value of PET/CT is currently being studied widely. However, there are many unresolved issues, such as the optimal probe of PET/CT for radiosensitivity prediction, the selection of the most useful PET/CT parameters and their optimal cut-offs such as total lesion glycolysis (TLG), metabolic tumor volume (MTV) and standardized uptake value (SUV), and the optimal timing of PET/CT pre-treatment, during or following RT. Different radiosensitivity of tumors, modes of radiotherapy action and fraction scheduling may complicate the appropriate choice. In this study, we will discuss the diverse methods for evaluating radiosensitivity, and will also focus on the selection of the optimal probe, timing, cut-offs and parameters of PET/CT for evaluating the radiotherapy response.

Development and validation of a radiomics signature on differentially expressed features of 18F-FDG PET to predict treatment response of concurrent chemoradiotherapy in thoracic esophagus squamous cell carcinoma

BACKGROUND AND PURPOSE

To investigate potential image markers for early prediction of treatment response on thoracic esophagus squamous cell carcinoma (ESCC) treated with concurrent chemoradiotherapy (CCRT).

MATERIALS AND METHODS

159 thoracic ESCC patients enrolled from two institutions were divided into training and validation sets. A total of 944 radiomics features were extracted from pretreatment ¹⁸F-FDG PET images. We first performed the inter-observer reproducibility test in 10 pairs of patients (responders vs. nonresponders), and the limma package was used to identify differentially expressed features (DEFs). Then the least absolute shrinkage and selection operator (LASSO) logistic regression model with 10-fold cross-validation was used to construct a treatment response related radiomics signature. Finally, the performance was assessed in both sets with receiver operating characteristic (ROC) curves and Kaplan-Meier analysis.

RESULTS

After the inter-observer test, 691 features were considered reproducible and been retained (ICC > 0.9). 61 DEFs were selected from limma and entered into the LASSO logistic regression model. The radiomics signature was significantly associated with treatment response in the training (p < 0.001) and validation set (p = 0.026), which achieved area under curve (AUC) values of 0.844 and 0.835, respectively. Delong test results of two ROCs showed no significant difference (p = 0.918). The cut-off value of the radiomics signature could successfully divide patients into high-risk and low-risk groups in both sets.

CONCLUSION

This study indicated that the proposed radiomics signature could be a useful image marker to predict the therapeutic response of thoracic ESCC patients treated with CCRT.

Preoperative Prediction of Pathologic Response to Neoadjuvant Chemoradiotherapy in Patients With Esophageal Cancer Using 18F-FDG PET/CT and DW-MRI: A Prospective Multicenter Study

PURPOSE

Accurate preoperative prediction of pathologic response to neoadjuvant chemoradiotherapy (nCRT) in patients with esophageal cancer could enable omission of esophagectomy in patients with a pathologic complete response (pCR). This study aimed to evaluate the individual and combined value of ¹⁸F-fluorodeoxyglucose positron emission tomography with integrated computed tomography (¹⁸F-FDG PET/CT) and diffusion-weighted magnetic resonance imaging (DW-MRI) during and after nCRT to predict pathologic response in patients with esophageal cancer.

METHODS AND MATERIALS

In this multicenter prospective study, patients scheduled to receive nCRT followed by esophagectomy for esophageal cancer underwent ¹⁸F-FDG PET/CT and DW-MRI scanning before the start of nCRT, during nCRT, and before esophagectomy. Response to nCRT was based on histopathologic evaluation of the resection specimen. Relative changes in ¹⁸F-FDG PET/CT and DW-MRI parameters were compared between patients with pCR and non-pCR groups. Multivariable ridge regression analyses with bootstrapped c-indices were performed to evaluate the individual and combined value of ¹⁸F-FDG PET/CT and DW-MRI.

RESULTS

pCR was found in 26.1% of 69 patients. Relative changes in ¹⁸F-FDG PET/CT parameters after nCRT (Δ standardized uptake value [SUV]_mean,postP = .016, and Δ total lesion glycolysis _postP = .024), as well as changes in DW-MRI parameters during nCRT (Δ apparent diffusion coefficient [ADC]_duringP = .008) were significantly different between pCR and non-pCR. A c-statistic of 0.84 was obtained for a model with ΔADC_during, ΔSUV_mean,post, and histology in classifying patients as pCR (versus 0.82 for ΔADC_during and 0.79 for ΔSUV_mean,post alone).

CONCLUSIONS

Changes on ¹⁸F-FDG PET/CT after nCRT and early changes on DW-MRI during nCRT can help identify pCR to nCRT in esophageal cancer. Moreover, ¹⁸F-FDG PET/CT and DW-MRI might be of complementary value in the assessment of pCR.

Prediction of Anal Cancer Recurrence After Chemoradiotherapy Using Quantitative Image Features Extracted From Serial 18F-FDG PET/CT

We extracted image features from serial ¹⁸F-labeled fluorodeoxyglucose (FDG) positron emission tomography (PET) / computed tomography (CT) scans of anal cancer patients for the prediction of tumor recurrence after chemoradiation therapy (CRT). Seventeen patients (4 recurrent and 13 non-recurrent) underwent three PET/CT scans at baseline (Pre-CRT), in the middle of the treatment (Mid-CRT) and post-treatment (Post-CRT) were included. For each patient, Mid-CRT and Post-CRT scans were aligned to Pre-CRT scan. Comprehensive image features were extracted from CT and PET (SUV) images within manually delineated gross tumor volume, including geometry features, intensity features and texture features. The difference of feature values between two time points were also computed and analyzed. We employed univariate logistic regression model, multivariate model, and naïve Bayesian classifier to analyze the image features and identify useful tumor recurrent predictors. The area under the receiver operating characteristic (ROC) curve (AUC) was used to evaluate the accuracy of the prediction. In univariate analysis, six geometry, three intensity, and six texture features were identified as significant predictors of tumor recurrence. A geometry feature of Roundness between Post-CRT and Pre-CRT CTs was identified as the most important predictor with an AUC value of 1.00 by multivariate logistic regression model. The difference of Number of Pixels on Border (geometry feature) between Post-CRT and Pre-CRT SUVs and Elongation (geometry feature) of Post-CRT CT were identified as the most useful feature set (AUC = 1.00) by naïve Bayesian classifier. To investigate the early prediction ability, we used features only from Pre-CRT and Mid-CRT scans. Orientation (geometry feature) of Pre-CRT SUV, Mean (intensity feature) of Pre-CRT CT, and Mean of Long Run High Gray Level Emphasis (LRHGLE) (texture feature) of Pre-CRT CT were identified as the most important feature set (AUC = 1.00) by multivariate logistic regression model. Standard deviation (intensity feature) of Mid-CRT SUV and difference of Mean of LRHGLE (texture feature) between Mid-CRT and Pre-CRT SUVs were identified as the most important feature set (AUC = 0.86) by naïve Bayesian classifier. The experimental results demonstrated the potential of serial PET/CT scans in early prediction of anal tumor recurrence.

Prediction of Immunohistochemistry of Suspected Thyroid Nodules by Use of Machine Learning-Based Radiomics

OBJECTIVE. The purpose of this study was to develop and validate a radiomics model for evaluating immunohistochemical characteristics in patients with suspected thyroid nodules. MATERIALS AND METHODS. A total of 103 patients (training cohort-to-validation cohort ratio, ≈ 3:1) with suspected thyroid nodules who had undergone thyroidectomy and immunohistochemical analysis were enrolled. The immunohistochemical markers were cytokeratin 19, galectin 3, thyroperoxidase, and high-molecular-weight cytokeratin. All patients underwent CT before surgery, and a 3D slicer was used to analyze images of the surgical specimen. Test-retest and Spearman correlation coefficient (ρ) were used to select reproducible and nonredundant features. The Kruskal-Wallis test (p < 0.05) was used for feature selection, and a feature-based model was built by support vector machine methods. The performance of the radiomic models was assessed with respect to accuracy, sensitivity, specificity, corresponding AUC, and independent validation. RESULTS. Eighty-six reproducible and nonredundant features selected from the 828 features were used to build the model. The best performance of the cytokeratin 19 model yielded accuracy of 84.4% in the training cohort and 80.0% in the validation cohort. The thyroperoxidase and galectin 3 predictive models yielded accuracies of 81.4% and 82.5% in the training cohort and 84.2% and 85.0% in the validation cohort. The performance of the high-molecular-weight cytokeratin predictive model was not good (accuracy, 65.7%) and could not be validated. CONCLUSION. A radiomics model with excellent performance was developed for individualized noninvasive prediction of the presence of cytokeratin 19, galectin 3, and thyroperoxidase based on CT images. This model may be used to identify benign and malignant thyroid nodules.

Does Measurement of First-Order and Heterogeneity Parameters Improve Response Assessment of Bone Metastases in Breast Cancer Compared to SUVmax in [18F]fluoride and [18F]FDG PET?

PURPOSE

To establish whether first-order statistical features from [18F]fluoride and 2-deoxy-2-[18F] fluoro-D-glucose ([18F]FDG) positron emission tomography/x-ray computed tomography (PET/CT) demonstrate incremental value in skeletal metastasis response assessment compared with maximum standardised uptake value (SUVmax).

PROCEDURES

Sixteen patients starting endocrine treatment for de novo or progressive breast cancer bone metastases were prospectively recruited to undergo [18F]fluoride and [18F]FDG PET/CT scans before and 8 weeks after treatment. Percentage changes in SUV parameters, metabolic tumour volume (MTV), total lesion metabolism (TLM), standard deviation (SD), entropy, uniformity and absolute changes in kurtosis and skewness, from the same ≤ 5 index lesions, were measured. Clinical response to 24 weeks, assessed by two experienced oncologists blinded to PET/CT imaging findings, was used as a reference standard and associations were made between parameters and progression free and overall survival.

RESULTS

[18F]fluoride PET/CT: In four patients (20 lesions) with progressive disease (PD), TLM and kurtosis predicted PD better than SUVmax on a patient basis (4, 4 and 3 out of 4, respectively) and TLM, entropy, uniformity and skewness on a lesion basis (18, 16, 16, 18 and 15 out of 20, respectively). Kurtosis was independently associated with PFS (p = 0.033) and OS (p = 0.008) on Kaplan-Meier analysis. [18F]FDG PET: No parameter provided incremental value over SUVmax in predicting PD or non-PD. TLM was significantly associated with OS (p = 0.041) and skewness with PFS (p = 0.005). Interlesional heterogeneity of response was seen in 11/16 and 8/16 patients on [18F]fluoride and [18F]FDG PET/CT, respectively.

CONCLUSION

With [18F]fluoride PET/CT, some first-order features, including those that take into account lesion volume but also some heterogeneity parameters, provide incremental value over SUVmax in predicting clinical response and survival in breast cancer patients with bone metastases treated with endocrine therapy. With [18F]FDG PET/CT, no first-order parameters were more accurate than SUVmax although TLM and skewness were associated with OS and PFS, respectively. Intra-patient heterogeneity of response occurs commonly between metastases with both tracers and most parameters.

Prognostic Value of Functional Parameters of 18F-FDG-PET Images in Patients with Primary Renal/Adrenal Lymphoma

Objectives

The aim of this study is to explore the textural features that may identify the morphological changes in the lymphoma region and predict the prognosis of patients with primary renal lymphoma (PRL) and primary adrenal lymphoma (PAL).

Methods

This retrospective study comprised nineteen non-Hodgkin's lymphoma (NHL) patients undergoing ¹⁸F-FDG-PET/CT at West China Hospital from December 2013 to May 2017. ¹⁸F-FDG-PET images were reviewed independently by two board certificated radiologists of nuclear medicine, and the texture features were extracted from LifeX packages. The prognostic value of PET FDG-uptake parameters, patients' baseline characteristics, and textural parameters were analyzed using Kaplan–Meier analysis. Cox regression analysis was used to identify the independent prognostic factors among the imaging and clinical features.

Results

The overall survival of included patients was 18.84 ± 13.40 (mean ± SD) months. Univariate Cox analyses found that the tumor stage, GLCM (gray-level co-occurrence matrix) entropy, GLZLM_GLNU (gray-level nonuniformity), and GLZLM_ZLNU (zone length nonuniformity), values were significant predictors for OS. Among them, GLRLM_RLNU ≥216.6 demonstrated association with worse OS at multivariate analysis (HR 9.016, 95% CI 1.041–78.112, p=0.046).

Conclusions

The texture analysis of ¹⁸F-FDG-PET images could potentially serve as a noninvasive strategy to predict the overall survival of patients with PRL and PAL.

Radiomics in esophageal and gastric cancer

Esophageal, esophago-gastric, and gastric cancers are major causes of cancer morbidity and cancer death. For patients with potentially resectable disease, multi-modality treatment is recommended as it provides the best chance of survival. However, quality of life may be adversely affected by therapy, and with a wide variation in outcome despite multi-modality therapy, there is a clear need to improve patient stratification. Radiomic approaches provide an opportunity to improve tumor phenotyping. In this review we assess the evidence to date and discuss how these approaches could improve outcome in esophageal, esophago-gastric, and gastric cancer.

A machine-learning-based prediction model of fistula formation after interstitial brachytherapy for locally advanced gynecological malignancies

PURPOSE

External beam radiotherapy combined with interstitial brachytherapy is commonly used to treat patients with bulky, advanced gynecologic cancer. However, the high radiation dose needed to control the tumor may result in fistula development. There is a clinical need to identify patients at high risk for fistula formation such that treatment may be managed to prevent this toxic side effect. This work aims to develop a fistula prediction model framework using machine learning based on patient, tumor, and treatment features.

METHODS AND MATERIALS

This retrospective study included 35 patients treated at our institution using interstitial brachytherapy for various gynecological malignancies. Five patients developed rectovaginal fistula and two developed both rectovaginal and vesicovaginal fistula. For each patient, 31 clinical features of multiple data types were collected to develop a fistula prediction framework. A nonlinear support vector machine was used to build the prediction model. Sequential backward feature selection and sequential floating backward feature selection methods were used to determine optimal feature sets. To overcome data imbalance issues, the synthetic minority oversampling technique was used to generate synthetic fistula cases for model training.

RESULTS

Seven mixed data features were selected by both sequential backward selection and sequential floating backward selection methods. Our prediction model using these features achieved a high prediction accuracy, that is, 0.904 area under the curve, 97.1% sensitivity, and 88.5% specificity.

CONCLUSIONS

A machine-learning-based prediction model of fistula formation has been developed for patients with advanced gynecological malignancies treated using interstitial brachytherapy. This model may be clinically impactful pending refinement and validation in a larger series.

Deep Learning for Variational Multimodality Tumor Segmentation in PET/CT

Positron emission tomography/computed tomography (PET/CT) imaging can simultaneously acquire functional metabolic information and anatomical information of the human body. How to rationally fuse the complementary information in PET/CT for accurate tumor segmentation is challenging. In this study, a novel deep learning based variational method was proposed to automatically fuse multimodality information for tumor segmentation in PET/CT. A 3D fully convolutional network (FCN) was first designed and trained to produce a probability map from the CT image. The learnt probability map describes the probability of each CT voxel belonging to the tumor or the background, and roughly distinguishes the tumor from its surrounding soft tissues. A fuzzy variational model was then proposed to incorporate the probability map and the PET intensity image for an accurate multimodality tumor segmentation, where the probability map acted as a membership degree prior. A split Bregman algorithm was used to minimize the variational model. The proposed method was validated on a non-small cell lung cancer dataset with 84 PET/CT images. Experimental results demonstrated that: 1). Only a few training samples were needed for training the designed network to produce the probability map; 2). The proposed method can be applied to small datasets, normally seen in clinic research; 3). The proposed method successfully fused the complementary information in PET/CT, and outperformed two existing deep learning-based multimodality segmentation methods and other multimodality segmentation methods using traditional fusion strategies (without deep learning); 4). The proposed method had a good performance for tumor segmentation, even for those with Fluorodeoxyglucose (FDG) uptake inhomogeneity and blurred tumor edges (two major challenges in PET single modality segmentation) and complex surrounding soft tissues (one major challenge in CT single modality segmentation), and achieved an average dice similarity indexes (DSI) of 0.86 ± 0.05, sensitivity (SE) of 0.86 ± 0.07, positive predictive value (PPV) of 0.87 ± 0.10, volume error (VE) of 0.16 ± 0.12, and classification error (CE) of 0.30 ± 0.12.

Tumor co-segmentation in PET/CT using multi-modality fully convolutional neural network

Automatic tumor segmentation from medical images is an important step for computer-aided cancer diagnosis and treatment. Recently, deep learning has been successfully applied to this task, leading to state-of-the-art performance. However, most of existing deep learning segmentation methods only work for a single imaging modality. PET/CT scanner is nowadays widely used in the clinic, and is able to provide both metabolic information and anatomical information through integrating PET and CT into the same utility. In this study, we proposed a novel multi-modality segmentation method based on a 3D fully convolutional neural network (FCN), which is capable of taking account of both PET and CT information simultaneously for tumor segmentation. The network started with a multi-task training module, in which two parallel sub-segmentation architectures constructed using deep convolutional neural networks (CNNs) were designed to automatically extract feature maps from PET and CT respectively. A feature fusion module was subsequently designed based on cascaded convolutional blocks, which re-extracted features from PET/CT feature maps using a weighted cross entropy minimization strategy. The tumor mask was obtained as the output at the end of the network using a softmax function. The effectiveness of the proposed method was validated on a clinic PET/CT dataset of 84 patients with lung cancer. The results demonstrated that the proposed network was effective, fast and robust and achieved significantly performance gain over CNN-based methods and traditional methods using PET or CT only, two V-net based co-segmentation methods, two variational co-segmentation methods based on fuzzy set theory and a deep learning co-segmentation method using W-net.

A pilot study using kernelled support tensor machine for distant failure prediction in lung SBRT

We developed a kernelled support tensor machine (KSTM)-based model with tumor tensors derived from pre-treatment PET and CT imaging as input to predict distant failure in early stage non-small cell lung cancer (NSCLC) treated with stereotactic body radiation therapy (SBRT). The patient cohort included 110 early stage NSCLC patients treated with SBRT, 25 of whom experienced failure at distant sites. Three-dimensional tumor tensors were constructed and used as input for the KSTM-based classifier. A KSTM iterative algorithm with a convergent proof was developed to train the weight vectors for every mode of the tensor for the classifier. In contrast to conventional radiomics approaches that rely on handcrafted imaging features, the KSTM-based classifier uses 3D imaging as input, taking full advantage of the imaging information. The KSTM-based classifier preserves the intrinsic 3D geometry structure of the medical images and the correlation in the original images and trains the classification hyper-plane in an adaptive feature tensor space. The KSTM-based predictive algorithm was compared with three conventional machine learning models and three radiomics approaches. For PET and CT, the KSTM-based predictive method achieved the highest prediction results among the seven methods investigated in this study based on 10-fold cross validation and independent testing.