Deep Radiomics Research Articles

Multicenter evaluation of CT deep radiomics model in predicting Leibovich score risk groups for non-metastatic clear cell renal cell carcinoma

BackgroundNon-metastatic clear cell renal cell carcinoma (nccRCC) poses a significant risk of postoperative recurrence and metastasis, underscoring the importance of accurate preoperative risk assessment. While the Leibovich score is effective, it relies on postoperative histopathological data. This study aims to evaluate the efficacy of CT radiomics and deep learning models in predicting Leibovich score risk groups in nccRCC, and to explore the interrelationship between CT and pathological features. Patients and MethodsThis research analyzed 600 nccRCC patients from four datasets, dividing them into low (Leibovich scores of 0–2) and intermediate to high risk (Leibovichscores exceeding 3) groups. Radiological models were developed from CT subjective features, and radiomics and deep learning models were constructed from CT images. Additionally, a deep radiomics model and a fusion model integrating radiological, radiomics, and deep learning features were introduced. Model performance was assessed by AUC values, while survival differences across predicted groups were analyzed using survival curves and the DeLong test. Moreover, the research investigated the connection between CT and pathological features derived from whole-slide pathological images in the second validation dataset. ResultsWithin the training dataset, four radiological, three radiomics, and thirteen deep learning features were selected to develop models predicting nccRCC Leibovich score risk groups. The deep radiomics model demonstrated superior predictive accuracy, evidenced by AUC values of 0.881, 0.829, and 0.819 in external validation datasets. Notably, significant differences in overall survival were observed among patients classified by this model (DeLong test p < 0.05 across all datasets). Furthermore, a correlation and complementarity were observed between deep radiomics features and pathological deep learning features. ConclusionsThe CT deep radiomics model precisely predicts nccRCC Leibovich score risk groups preoperatively and highlights the synergistic effect between CT and pathological data.

The Role of Radiomics in the Prediction of Clinically Significant Prostate Cancer in the PI-RADS v2 and v2.1 Era: A Systematic Review.

Early detection of clinically significant prostate cancer (csPCa) has substantially improved with the latest PI-RADS versions. However, there is still an overdiagnosis of indolent lesions (iPCa), and radiomics has emerged as a potential solution. The aim of this systematic review is to evaluate the role of handcrafted and deep radiomics in differentiating lesions with csPCa from those with iPCa and benign lesions on prostate MRI assessed with PI-RADS v2 and/or 2.1. The literature search was conducted in PubMed, Cochrane, and Web of Science databases to select relevant studies. Quality assessment was carried out with Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2), Radiomic Quality Score (RQS), and Checklist for Artificial Intelligence in Medical Imaging (CLAIM) tools. A total of 14 studies were deemed as relevant from 411 publications. The results highlighted a good performance of handcrafted and deep radiomics methods for csPCa detection, but without significant differences compared to radiologists (PI-RADS) in the few studies in which it was assessed. Moreover, heterogeneity and restrictions were found in the studies and quality analysis, which might induce bias. Future studies should tackle these problems to encourage clinical applicability. Prospective studies and comparison with radiologists (PI-RADS) are needed to better understand its potential.

Radiomics of pituitary adenoma using computer vision: a review.

Pituitary adenomas (PA) represent the most common type of sellar neoplasm. Extracting relevant information from radiological images is essential for decision support in addressing various objectives related to PA. Given the critical need for an accurate assessment of the natural progression of PA, computer vision (CV) and artificial intelligence (AI) play a pivotal role in automatically extracting features from radiological images. The field of "Radiomics" involves the extraction of high-dimensional features, often referred to as "Radiomic features," from digital radiological images. This survey offers an analysis of the current state of research in PA radiomics. Our work comprises a systematic review of 34 publications focused on PA radiomics and other automated information mining pertaining to PA through the analysis of radiological data using computer vision methods. We begin with a theoretical exploration essential for understanding the theoretical background of radionmics, encompassing traditional approaches from computer vision and machine learning, as well as the latest methodologies in deep radiomics utilizing deep learning (DL). Thirty-four research works under examination are comprehensively compared and evaluated. The overall results achieved in the analyzed papers are high, e.g., the best accuracy is up to 96% and the best achieved AUC is up to 0.99, which establishes optimism for the successful use of radiomic features. Methods based on deep learning seem to be the most promising for the future. In relation to this perspective DL methods, several challenges are remarkable: It is important to create high-quality and sufficiently extensive datasets necessary for training deep neural networks. Interpretability of deep radiomics is also a big open challenge. It is necessary to develop and verify methods that will explain to us how deep radiomic features reflect various physics-explainable aspects.

Integration of pre-treatment computational radiomics, deep radiomics, and transcriptomics enhances soft-tissue sarcoma patient prognosis

Our objective was to capture subgroups of soft-tissue sarcoma (STS) using handcraft and deep radiomics approaches to understand their relationship with histopathology, gene-expression profiles, and metastatic relapse-free survival (MFS). We included all consecutive adults with newly diagnosed locally advanced STS (N = 225, 120 men, median age: 62 years) managed at our sarcoma reference center between 2008 and 2020, with contrast-enhanced baseline MRI. After MRI postprocessing, segmentation, and reproducibility assessment, 175 handcrafted radiomics features (h-RFs) were calculated. Convolutional autoencoder neural network (CAE) and half-supervised CAE (HSCAE) were trained in repeated cross-validation on representative contrast-enhanced slices to extract 1024 deep radiomics features (d-RFs). Gene-expression levels were calculated following RNA sequencing (RNAseq) of 110 untreated samples from the same cohort. Unsupervised classifications based on h-RFs, CAE, HSCAE, and RNAseq were built. The h-RFs, CAE, and HSCAE grouping were not associated with the transcriptomics groups but with prognostic radiological features known to correlate with lower survivals and higher grade and SARCULATOR groups (a validated prognostic clinical-histological nomogram). HSCAE and h-RF groups were also associated with MFS in multivariable Cox regressions. Combining HSCAE and transcriptomics groups significantly improved the prognostic performances compared to each group alone, according to the concordance index. The combined radiomic-transcriptomic group with worse MFS was characterized by the up-regulation of 707 genes and 292 genesets related to inflammation, hypoxia, apoptosis, and cell differentiation. Overall, subgroups of STS identified on pre-treatment MRI using handcrafted and deep radiomics were associated with meaningful clinical, histological, and radiological characteristics, and could strengthen the prognostic value of transcriptomics signatures.

Combining machine and deep transfer learning for mediastinal lymph node evaluation in patients with lung cancer

BackgroundThe prognosis and survival of patients with lung cancer are likely to deteriorate with metastasis. Using deep-learning in the detection of lymph node metastasis can facilitate the noninvasive calculation of the likelihood of such metastasis, thereby providing clinicians with crucial information to enhance diagnostic precision and ultimately improve patient survival and prognosis MethodsIn total, 623 eligible patients were recruited from two medical institutions. Seven deep learning models, namely Alex, GoogLeNet, Resnet18, Resnet101, Vgg16, Vgg19, and MobileNetv3 (small), were utilized to extract deep image histological features. The dimensionality of the extracted features was then reduced using the Spearman correlation coefficient (r ≥ 0.9) and Least Absolute Shrinkage and Selection Operator. Eleven machine learning methods, namely Support Vector Machine, K-nearest neighbor, Random Forest, Extra Trees, XGBoost, LightGBM, Naive Bayes, AdaBoost, Gradient Boosting Decision Tree, Linear Regression, and Multilayer Perceptron, were employed to construct classification prediction models for the filtered final features. The diagnostic performances of the models were assessed using various metrics, including accuracy, area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and negative predictive value. Calibration and decision-curve analyses were also performed. ResultsThe present study demonstrated that using deep radiomic features extracted from Vgg16, in conjunction with a prediction model constructed via a linear regression algorithm, effectively distinguished the status of mediastinal lymph nodes in patients with lung cancer. The performance of the model was evaluated based on various metrics, including accuracy, area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and negative predictive value, which yielded values of 0.808, 0.834, 0.851, 0.745, 0.829, and 0.776, respectively. The validation set of the model was assessed using clinical decision curves, calibration curves, and confusion matrices, which collectively demonstrated the model's stability and accuracy ConclusionIn this study, information on the deep radiomics of Vgg16 was obtained from computed tomography images, and the linear regression method was able to accurately diagnose mediastinal lymph node metastases in patients with lung cancer.

ViTR-SP: A CT-based vision transformer model for prediction of pneumonitis in patients with non-small cell lung cancer who received thoracic radiotherapy and immunotherapy.

e20034 Background: The combination of thoracic radiotherapy (TRT) and immunotherapy accompanied by the accumulation of pulmonary toxicity. This study aimed to construct a deep radiomics model based on sequential computed tomography (CT) images for radioimmune-associated pneumonitis prediction in non-small cell lung cancer (NSCLC) patients treated with TRT and immune checkpoint inhibitors (ICIs). Methods: Three models were constructed based on a vision transformer (ViT) network using sequential chest CT images, tumor-mask and lung-mask as inputs, respectively. NSCLC patients treated with TRT and ICIs at Shandong Cancer Hospital and Qilu Hospital were included for model training and external validation, respectively. The image-level model was trained based on ViT to identify predictive images, and the image-level probability distribution were performed. Then, a ViT-Recursive Neural Network (ViT-RNN) model framework was constructed at the patient-level to integrate the image-level information to derive the final prediction, which was evaluated in an independent validation cohort. Results: A total of 230 NSCLC patients from Shandong Cancer Hospital were randomly assigned in a 7:3 ratio for model training and validation, respectively. Another 42 patients from Qilu Hospital were used as an independent external validation cohort. The ViT-RNN model for symptomatic pneumonitis (ViTR-SP) based on sequential CT images achieved better area under curve (AUC), sensitivity, and specificity of 0.725, 0.750, and 0.886 than lung-mask based model (0.575, 0.625, and 0.705) and tumor-mask based model (0.658, 0.667, and 0.773) in the internal validation set. In the external validation set, the proposed model achieved the AUC, sensitivity, and specificity of 0.654, 0.818, 0.727, while the lung-mask and tumor-mask based models had the AUC, sensitivity, and specificity of 0.540, 0.545, 0.677, and 0.630, 0.545, 0.742, respectively. The performance of the model was improved after integrating clinical features, with internal and external validation AUC of 0.782 and 0.672, respectively. Conclusions: In conclusion, the clinical combined ViTR-SP neural network risk prediction model based on sequential chest CT images can be used for predicting the risk of developing radioimmune-associated pneumonitis in NSCLC patients.

Structural- and DTI- MRI enable automated prediction of IDH Mutation Status in CNS WHO Grade 2–4 glioma patients: a deep Radiomics Approach

BackgroundThe role of isocitrate dehydrogenase (IDH) mutation status for glioma stratification and prognosis is established. While structural magnetic resonance image (MRI) is a promising biomarker, it may not be sufficient for non-invasive characterisation of IDH mutation status. We investigated the diagnostic value of combined diffusion tensor imaging (DTI) and structural MRI enhanced by a deep radiomics approach based on convolutional neural networks (CNNs) and support vector machine (SVM), to determine the IDH mutation status in Central Nervous System World Health Organization (CNS WHO) grade 2–4 gliomas.MethodsThis retrospective study analyzed the DTI-derived fractional anisotropy (FA) and mean diffusivity (MD) images and structural images including fluid attenuated inversion recovery (FLAIR), non-enhanced T1-, and T2-weighted images of 206 treatment-naïve gliomas, including 146 IDH mutant and 60 IDH-wildtype ones. The lesions were manually segmented by experienced neuroradiologists and the masks were applied to the FA and MD maps. Deep radiomics features were extracted from each subject by applying a pre-trained CNN and statistical description. An SVM classifier was applied to predict IDH status using imaging features in combination with demographic data.ResultsWe comparatively assessed the CNN-SVM classifier performance in predicting IDH mutation status using standalone and combined structural and DTI-based imaging features. Combined imaging features surpassed stand-alone modalities for the prediction of IDH mutation status [area under the curve (AUC) = 0.846; sensitivity = 0.925; and specificity = 0.567]. Importantly, optimal model performance was noted following the addition of demographic data (patients’ age) to structural and DTI imaging features [area under the curve (AUC) = 0.847; sensitivity = 0.911; and specificity = 0.617].ConclusionsImaging features derived from DTI-based FA and MD maps combined with structural MRI, have superior diagnostic value to that provided by standalone structural or DTI sequences. In combination with demographic information, this CNN-SVM model offers a further enhanced non-invasive prediction of IDH mutation status in gliomas.

Hybrid deep multi-task learning radiomics approach for predicting EGFR mutation status of non-small cell lung cancer in CT images

Objective. Epidermal growth factor receptor (EGFR) mutation genotyping plays a pivotal role in targeted therapy for non-small cell lung cancer (NSCLC). We aimed to develop a computed tomography (CT) image-based hybrid deep radiomics model to predict EGFR mutation status in NSCLC and investigate the correlations between deep image and quantitative radiomics features. Approach. First, we retrospectively enrolled 818 patients from our centre and 131 patients from The Cancer Imaging Archive database to establish a training cohort (N = 654), an independent internal validation cohort (N = 164) and an external validation cohort (N = 131). Second, to predict EGFR mutation status, we developed three CT image-based models, namely, a multi-task deep neural network (DNN), a radiomics model and a feature fusion model. Third, we proposed a hybrid loss function to train the DNN model. Finally, to evaluate the model performance, we computed the areas under the receiver operating characteristic curves (AUCs) and decision curve analysis curves of the models. Main results. For the two validation cohorts, the feature fusion model achieved AUC values of 0.86 ± 0.03 and 0.80 ± 0.05, which were significantly higher than those of the single-task DNN and radiomics models (all P < 0.05). There was no significant difference between the feature fusion and the multi-task DNN models (P > 0.8). The binary prediction scores showed excellent prognostic value in predicting disease-free survival (P = 0.02) and overall survival (P < 0.005) for validation cohort 2. Significance. The results demonstrate that (1) the feature fusion and multi-task DNN models achieve significantly higher performance than that of the conventional radiomics and single-task DNN models, (2) the feature fusion model can decode the imaging phenotypes representing NSCLC heterogeneity related to both EGFR mutation and patient NSCLC prognosis, and (3) high correlations exist between some deep image and radiomics features.

Deep radiomics-based fusion model for prediction of bevacizumab treatment response and outcome in patients with colorectal cancer liver metastases: a multicentre cohort study

Accurate tumour response prediction to targeted therapy allows for personalised conversion therapy for patients with unresectable colorectal cancer liver metastases (CRLM). In this study, we aimed to develop and validate a multi-modal deep learning model to predict the efficacy of bevacizumab in patients with initially unresectable CRLM using baseline PET/CT, clinical data, and colonoscopy biopsy specimens. In this multicentre cohort study, we retrospectively collected data of 307 patients with CRLM from the BECOME study (NCT01972490) (Zhongshan Hospital of Fudan University, Shanghai) and two independent Chinese cohorts (internal validation cohort from January 1, 2018 to December 31, 2018at Zhongshan Hospital of Fudan University; external validation cohort from January 1, 2020 to December 31, 2020at Zhongshan Hospital-Xiamen, Shanghai, and the First Hospital of Wenzhou Medical University, Wenzhou). The main inclusion criteria were that patients with CRLM had pre-treatment PET/CT images as well as colonoscopy specimens. After extracting PET/CT features with deep neural networks (DNN) and selecting related clinical factors using LASSO analysis, a random forest classifier was built as the Deep Radiomics Bevacizumab efficacy predicting model (DERBY). Furthermore, by combining histopathological biomarkers into DERBY, we established DERBY+. The performance of model was evaluated using area under the curve (AUC), sensitivity, specificity, positive predictive value, and negative predictive value. DERBY achieved promising performance in predicting bevacizumab sensitivity with an AUC of 0.77 and 95% confidence interval (CI) [0.67-0.87]. After combining histopathological features, we developed DERBY+, which had more robust accuracy for predicting tumour response in external validation cohort (AUC 0.83 and 95% CI [0.75-0.92], sensitivity 80.4%, specificity 76.8%). DERBY+ also had prognostic value: the responders had longer progression-free survival (median progression-free survival: 9.6 vs 6.3 months, p=0.002) and overall survival (median overall survival: 27.6 vs 18.5 months, p=0.010) than non-responders. This multi-modal deep radiomics model, using PET/CT, clinical data and histopathological data, was able to identify patients with bevacizumab-sensitive CRLM, providing a favourable approach for precise patient treatment. To further validate and explore the clinical impact of this work, future prospective studies with larger patient cohorts are warranted. The National Natural Science Foundation of China; Fujian Provincial Health Commission Project; Xiamen Science and Technology Agency Program; Clinical Research Plan of SHDC; Shanghai Science and Technology Committee Project; Clinical Research Plan of SHDC; Zhejiang Provincial Natural Science Foundation of China; and National Science Foundation of Xiamen.

Develop a Deep Radiomics Model for Predicting the Response to Neoadjuvant Chemoradiotherapy (nCRT) in Patients with Locally Advanced Esophageal Cancer Using Three-Stage Longitudinal CT Images

To develop a deep radiomics model for predicting the response to neoadjuvant chemoradiotherapy (nCRT) of patients with locally advanced esophageal cancer using three stage longitudinal CT images. In this study, 189 patients were used to train our model, 80 patients were used to test the performance of the trained model. All patients enrolled in this study underwent nCRT followed by esophagectomy, and all patients underwent three stage longitudinal CT scans (before nCRT, after nCRT and before esophagectomy). The number of radiomics features for each CT images was 2153, the number of longitudinal radiomic features for each patient was 6459. Then, we used the least absolute shrinkage and selection operator regression for feature importance analysis. After this, the selected features of each stage were feed to the disentangled representation network to explore the relationship between the dynamic changes of tumors before and after nCRT and the pathological complete response (pCR). In addition, in order to further evaluate the effect of the time interval between nCRT and esophagectomy on the response to nCRT, we conducted subgroup analysis on different time frames. The performance of our model was evaluated by area under curve (AUC), accuracy, sensitivity and specificity. Compared with only using single-stage CT images (the AUC of only using before nCRT CT images, after nCRT CT images, before esophagectomy CT images were 73.27%, 74.21%, and 74.95%, respectively), effectively exploring the dynamic changes of the tumor can achieve better performance in predicting the response to nCRT in the testing cohort (the AUC was 84.29%, 95% CI, 81.14%-87.44%). In addition, the performance of our proposed method outperforms any combinations of two stage CT images (the AUC using the CT images of before nCRT and after nCRT was 77.92%, the AUC using the CT images of before nCRT and before esophagectomy was 79.31%, the AUC using the CT images of after nCRT and before esophagectomy was 80.01%). Finally, the results showed that exploring the dynamic changes of the tumor using the three-stage CT images outperformed using single-stage CT images and any combinations of two-stage CT images in predicting the response to nCRT. The study also found that the time interval between nCRT and esophagectomy had some influence on the accuracy of pCR prediction, with the prediction accuracy tending to increase from 1 to 6 weeks and stabilizing after 6 weeks. By exploring the dynamic changes of tumors, the designed disentangled representation network can effectively predict the response to nCRT of patients with esophageal cancer. In addition, the time interval between nCRT and esophagectomy also has a certain impact on the response to nCRT.

Transfer-Learning Deep Radiomics and Hand-Crafted Radiomics for Classifying Lymph Nodes from Contrast-Enhanced Computed Tomography in Lung Cancer.

Positron emission tomography (PET) is currently considered the non-invasive reference standard for lymph node (N-)staging in lung cancer. However, not all patients can undergo this diagnostic procedure due to high costs, limited availability, and additional radiation exposure. The purpose of this study was to predict the PET result from traditional contrast-enhanced computed tomography (CT) and to test different feature extraction strategies. In this study, 100 lung cancer patients underwent a contrast-enhanced 18F-fluorodeoxyglucose (FDG) PET/CT scan between August 2012 and December 2019. We trained machine learning models to predict FDG uptake in the subsequent PET scan. Model inputs were composed of (i) traditional "hand-crafted" radiomics features from the segmented lymph nodes, (ii) deep features derived from a pretrained EfficientNet-CNN, and (iii) a hybrid approach combining (i) and (ii). In total, 2734 lymph nodes [555 (20.3%) PET-positive] from 100 patients [49% female; mean age 65, SD: 14] with lung cancer (60% adenocarcinoma, 21% plate epithelial carcinoma, 8% small-cell lung cancer) were included in this study. The area under the receiver operating characteristic curve (AUC) ranged from 0.79 to 0.87, and the scaled Brier score (SBS) ranged from 16 to 36%. The random forest model (iii) yielded the best results [AUC 0.871 (0.865-0.878), SBS 35.8 (34.2-37.2)] and had significantly higher model performance than both approaches alone (AUC: p < 0.001, z = 8.8 and z = 22.4; SBS: p < 0.001, z = 11.4 and z = 26.6, against (i) and (ii), respectively). Both traditional radiomics features and transfer-learning deep radiomics features provide relevant and complementary information for non-invasive N-staging in lung cancer.

Abstract 5435: Correlating and combining computational radiomics, deep radiomics and transcriptomics data in soft-tissue sarcomas (STS) patients highlight complementary prognostic information

Abstract Introduction: STS are mostly prognosticated through nomograms relying on age, size, histotype and grade. Radiomics approaches, complemented with deep-learning, deep-radiomics and gene-expression profiling, could help understanding the bridge between STS radiophenotypes and molecular features and provide more efficient prognostic tools. Our goals were to investigate correlations between imaging and transcriptomics patterns, and to develop supervised prognostic models for STS patients. Methods: We included all consecutive adult patients with newly-diagnosed locally-advanced STS managed at our sarcoma reference center between 2008 and 2020, with contrast-enhanced baseline MRI. After MRI dataset homogenization, we reduced the dimensions of the MRI data space by extracting 138 radiomics features (RFs) and 1024 deep-RFs with computational approach and autoencoder neural networks. Patient RNA was extracted from untreated samples. Following transcriptomic sequence analysis, gene expression levels for each patient were calculated. Complexity Index in Sarcoma (CINSARC) signature was extracted. Unsupervised classifications of patients based on radiomics, deep-radiomics and transcriptomics datasets were built using consensus hierarchical clustering. Differential Gene Expression and oncogenetic pathways analyses were performed. Associations between the 3 classifications, CINSARC, grade, histotypes and SARCULATOR were explored, as well as their prognostic value. The main outcome was the metastatic-relapse free survival (MFS). The SARCULATOR nomogram and prognostic semantic-radiological features were extracted for benchmarking and understanding models outputs. Results: 220 patients were included (111 men, median age: 62 years); 60 patients developed metastases after completing curative treatments (data are being updated with 2 additional follow-up years). Transcriptomic analysis was achieved in 54 patients and is being updated with 56 additional samples. So far, no significant associations were found between the radiomics-based classifications and the transcriptomics-based (including CINSARC). Nevertheless, the computational radiomics, deep-radiomics, and transcriptomics classifications were associated with MFS, though transcriptomic significance was dampened by the small sample size (P=0.008 [N=220], 0.006 [N=220] and 0.070 [N=54], respectively), suggesting complementary prognostic information. Supervised models using data-splitting, cross-validated algorithms training, and various combinations of input data are being elaborated to improve the MFS prediction. Conclusion: Integrating complementary multiomics datasets with computational and deep radiomics should pave the way for better performing and personalized prognostications in STS patients. Citation Format: Amandine Crombe, Carlo Lucchesi, Frédéric Bertolo, Michèle Kind, Raul Perret, Francois Le Loarer, Mariella Spalato-Ceruso, Maud Toulmonde, Audrey Laroche, Vanessa Chaire, Aurelien Bourdon, Antoine Italiano. Correlating and combining computational radiomics, deep radiomics and transcriptomics data in soft-tissue sarcomas (STS) patients highlight complementary prognostic information. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5435.

Fused deep learning paradigm for the prediction of o6-methylguanine-DNA methyltransferase genotype in glioblastoma patients: A neuro-oncological investigation

BackgroundThe O6-methylguanine-DNA methyltransferase (MGMT) is a deoxyribonucleic acid (DNA) repairing enzyme that has been established as an essential clinical brain tumor biomarker for Glioblastoma Multiforme (GBM). Knowing the status of MGMT methylation biomarkers using multi-parametric MRI (mp-MRI) helps neuro-oncologists to analyze GBM and its treatment plan. MethodThe hand-crafted radiomics feature extraction of GBM's subregions, such as edema(ED), tumor core (TC), and enhancing tumor (ET) in the machine learning (ML) framework, was investigated using support vector machine(SVM), K-Nearest Neighbours (KNN), random forest (RF), LightGBM, and extreme gradient boosting (XGB). For tissue-level analysis of the promotor genes in GBM, we used the deep residual neural network (ResNet-18) with 3D architecture, followed by EfficientNet-based investigation for variants as B0 and B1. Lastly, we analyzed the fused deep learning (FDL) framework that combines ML and DL frameworks. ResultStructural mp-MRI consisting of T1, T2, FLAIR, and T1GD having a size of 400 and 185 patients, respectively, for discovery and replication cohorts. Using the CV protocol in the ResNet-3D framework, MGMT methylation status prediction in mp-MRI gave the AUC of 0.753 (p < 0.0001) and 0.72 (p < 0.0001) for the discovery and replication cohort, respectively. We presented that the FDL is ∼7% superior to solo DL and ∼15% to solo ML. ConclusionThe proposed study aims to provide solutions for building an efficient predictive model of MGMT for GBM patients using deep radiomics features obtained from mp-MRI with the end-to-end ResNet-18 3D and FDL imaging signatures.

Graphic Intelligent Diagnosis of Hypoxic-Ischemic Encephalopathy Using MRI-Based Deep Learning Model

Introduction: Heterogeneous MRI manifestations restrict the efficiency and consistency of neuroradiologists in diagnosing hypoxic-ischemic encephalopathy (HIE) due to complex injury patterns. This study aimed to develop and validate an intelligent HIE identification model (termed as DLCRN, deep learning clinical-radiomics nomogram) based on conventional structural MRI and clinical characteristics. Methods: In this retrospective case-control study, full-term neonates with HIE and healthy controls were collected in two different medical centers from January 2015 to December 2020. Multivariable logistic regression analysis was implemented to establish the DLCRN model based on conventional MRI sequences and clinical characteristics. Discrimination, calibration, and clinical applicability were used to evaluate the model in the training and validation cohorts. Grad-class activation map algorithm was implemented to visualize the DLCRN. Results: 186 HIE patients and 219 healthy controls were assigned to the training, internal validation, and independent validation cohorts. Birthweight was incorporated with deep radiomics signatures to create the final DLCRN model. The DLCRN model achieved better discriminatory power than simple radiomics models, with an area under the curve (AUC) of 0.868, 0.813, and 0.798 in the training, internal validation, and independent validation cohorts, respectively. The DLCRN model was well calibrated and has clinical potential. Visualization of the DLCRN highlighted the lesion areas that conformed to radiological identification. Conclusion: Visualized DLCRN may be a useful tool in the objective and quantitative identification of HIE. Scientific application of the optimized DLCRN model may save time for screening early mild HIE, improve the consistency of HIE diagnosis, and guide timely clinical management.

CT-based deep learning radiomics signature for the preoperative prediction of the muscle-invasive status of bladder cancer.

Although the preoperative assessment of whether a bladder cancer (BCa) indicates muscular invasion is crucial for adequate treatment, there currently exist some challenges involved in preoperative diagnosis of BCa with muscular invasion. The aim of this study was to construct deep learning radiomic signature (DLRS) for preoperative predicting the muscle invasion status of BCa. A retrospective review covering 173 patients revealed 43 with pathologically proven muscle-invasive bladder cancer (MIBC) and 130 with non-muscle-invasive bladder cancer (non- MIBC). A total of 129 patients were randomly assigned to the training cohort and 44 to the test cohort. The Pearson correlation coefficient combined with the least absolute shrinkage and selection operator (LASSO) was utilized to reduce radiomic redundancy. To decrease the dimension of deep learning features, Principal Component Analysis (PCA) was adopted. Six machine learning classifiers were finally constructed based on deep learning radiomics features, which were adopted to predict the muscle invasion status of bladder cancer. The area under the curve (AUC), accuracy, sensitivity and specificity were used to evaluate the performance of the model. According to the comparison, DLRS-based models performed the best in predicting muscle violation status, with MLP (Train AUC: 0.973260 (95% CI 0.9488-0.9978) and Test AUC: 0.884298 (95% CI 0.7831-0.9855)) outperforming the other models. In the test cohort, the sensitivity, specificity and accuracy of the MLP model were 0.91 (95% CI 0.551-0.873), 0.78 (95% CI 0.594-0.863) and 0.58 (95% CI 0.729-0.827), respectively. DCA indicated that the MLP model showed better clinical utility than Radiomics-only model, which was demonstrated by the decision curve analysis. A deep radiomics model constructed with CT images can accurately predict the muscle invasion status of bladder cancer.

The deep radiomic analytics pipeline.

Radiomics refers to the process of extracting useful imaging features from radiological data. Conventional radiomics like standard uptake value, intensity histograms, or phase images involve hand-crafted (manual) or automated regions of interest (computer generated), however, artificial intelligence (AI) segmentation (AI-augmented radiomics) has recently emerged. Radiomic feature extraction extends image insights beyond simply data quantitation and provides additional insights to aid semantic reporting. Deeper layers of a convolutional neural network produce more abstract radiomic features that are referred to as deep radiomics. The application of radiomics in veterinary radiology is already firmly entrenched using hand-crafted and automated computer-generated radiomic features in X-ray, nuclear medicine, CT, ultrasound, and MRI. There is an opportunity for veterinary radiology to capitalize on advances in AI, machine learning, and deep learning to enrich imaging interpretation using deep radiomic feature extraction. This manuscript aims to provide a general understanding of radiomics and deep radiomics, and to arm readers with the vernacular to progress discussion and development of deep radiomics in veterinary imaging.

NIMG-83. MULTIMODAL PET/MRI RADIOMICS AND CLINICAL PARAMETERS FOR OVERALL SURVIVAL PREDICTION IN PATIENTS WITH IDH WILDTYPE GLIOBLASTOMA

Abstract BACKGROUND Currently, most radiomics studies on survival prediction in brain tumor patients are based on MRI only. The goal of our study was to evaluate multimodal radiomics derived from amino acid PET/MRI and clinical parameters for survival prediction in patients with newly diagnosed IDH wildtype glioblastoma. METHODS Sixty-three patients with newly diagnosed IDH wildtype glioblastoma were evaluated retrospectively. At initial diagnosis, all patients underwent structural MRI and O-(2-[18F]fluoroethyl)-L-tyrosine (FET) PET. Tumor volumes were automatically segmented using a deep learning-based tool followed by visual inspection. Predefined and deep radiomics features were extracted from both imaging modalities. Feature repeatability analyses and feature selection were performed to avoid overfitting. Cox regression models for overall survival were built from clinical parameters such as age or the extent of resection, radiomics features, and combinations thereof, and finally validated using 5-fold cross-validation. Further evaluation of the model in an external test dataset is ongoing. RESULTS The median overall survival was 12 months (range, 0-64 months). Higher age and larger FET PET tumor volumes were significantly correlated with shorter overall survival (age, r=-0.39, p&amp;lt; 0.001; volume, r=-0.31, p&amp;lt; 0.05). Models solely based on predefined FET PET or MRI radiomics features showed a similar mean concordance index (C-index) as the model based on clinical parameters (C-indices, 0.68±0.04; 0.64±0.03; and 0.69±0.08, respectively). Multimodal radiomics based on predefined and deep features yielded improved C-indices of 0.75±0.06 and 0.72±0.09, respectively. A model based on multimodal radiomics and clinical parameters achieved the best prognostic performance (C-index, 0.80±0.04). CONCLUSION Our results suggest an added clinical value of multimodal FET PET/MRI radiomics with clinical parameters for the non-invasive survival prediction in patients with IDH wildtype glioblastoma.

53P Baseline PET/CT deep radiomics signature apply for identifying bevacizumab sensitivity of RAS-mutant colorectal cancer liver metastases patients

Bevacizumab has significantly improved the resectability, response rate and survival of patients with RAS-mutant colorectal cancer liver metastases (CRLM). However, more than half of these patients were insensitive to bevacizumab therapy. Identification of patients who are sensitive to bevacizumab therapy may improve the response rate and reduce adverse events. In this study, we aimed to construct and validate a PET/CT deep radiomics signature to predict bevacizumab efficacy in initially unresectable RAS-mutant CRLM patients.

The transformational potential of molecular radiomics.

Conventional radiomics in nuclear medicine involve hand-crafted and computer-assisted regions of interest. Recent developments in artificial intelligence (AI) have seen the emergence of AI-augmented segmentation and extraction of lower order traditional radiomic features. Deep learning (DL) affords the opportunity to extract abstract radiomic features directly from input tensors (images) without the need for segmentation. These fourth-order, high dimensional radiomics produce deep radiomics and are well suited to the data density associated with the molecular nature of hybrid imaging. Molecular radiomics and deep molecular radiomics provide insights beyond images and quantitation typical of semantic reporting. While the application of molecular radiomics using hand-crafted and computer-generated features is integrated into decision-making in nuclear medicine, the acceptance of deep molecular radiomics is less universal. This manuscript aims to provide an understanding of the language and principles associated with radiomics and deep radiomics in nuclear medicine.

Radiomics Analysis of Lymph Nodes with Esophageal Squamous Cell Carcinoma Based on Deep Learning.

Purpose To assess the role of multiple radiomic features of lymph nodes in the preoperative prediction of lymph node metastasis (LNM) in patients with esophageal squamous cell carcinoma (ESCC). Methods Three hundred eight patients with pathologically confirmed ESCC were retrospectively enrolled (training cohort, n = 216; test cohort, n = 92). We extracted 207 handcrafted radiomic features and 1000 deep radiomic features of lymph nodes from their computed tomography (CT) images. The t-test and least absolute shrinkage and selection operator (LASSO) were used to reduce the dimensions and select key features. Handcrafted radiomics, deep radiomics, and clinical features were combined to construct models. Models I (handcrafted radiomic features), II (Model I plus deep radiomic features), and III (Model II plus clinical features) were built using three machine learning methods: support vector machine (SVM), adaptive boosting (AdaBoost), and random forest (RF). The best model was compared with the results of two radiologists, and its performance was evaluated in terms of sensitivity, specificity, accuracy, area under the curve (AUC), and receiver operating characteristic (ROC) curve analysis. Results No significant differences were observed between cohorts. Ten handcrafted and 12 deep radiomic features were selected from the extracted features (p < 0.05). Model III could discriminate between patients with and without LNM better than the diagnostic results of the two radiologists. Conclusion The combination of handcrafted radiomic features, deep radiomic features, and clinical features could be used clinically to assess lymph node status in patients with ESCC.