AutoRadiomics: A Framework for Reproducible Radiomics Research

Replication study of PD-L1 status prediction in NSCLC using PET/CT radiomics

This study investigates the predictive capability of radiomics in determining programmed cell death ligand 1 (PD-L1) expression (>=1%) status in non-small cell lung cancer (NSCLC) patients using a newly collected [18F]FDG PET/CT dataset. We aimed to replicate and validate the radiomics-based machine learning (ML) model proposed by Zhao et al. [1] predicting PD-L1 status from PET/CT-imaging. An independent cohort of 254 NSCLC patients underwent [18F]FDG PET/CT imaging, with primary tumor segmentation conducted using lung tissue window (LTW) and more conservative soft tissue window (STW) methods. Radiomics models ("Rad-score" and "complex model") and a clinical-stage model from Zhao et al. were evaluated via 10-fold cross-validation and AUC analysis, alongside a benchmark-study comparing different ML-model pipelines. Clinicopathological data were collected from medical records. On our data, the Rad-score model yielded mean AUCs of 0.593 (STW) and 0.573 (LTW), below Zhao et al.'s 0.761. The complex model achieved mean AUCs of 0.505 (STW) and 0.519 (LTW), lower than Zhao et al.'s 0.769. The clinical model showed a mean AUC of 0.555, below Zhao et al.'s 0.64. All models performed significantly lower than Zhao et al.'s findings. Our benchmark study on four ML pipelines revealed consistently low performance across all configurations. Our study failed to replicate original findings, suggesting poor model performance and questioning predictive value of radiomics features in classifying PD-L1 expression from PET/CT imaging. These results highlight challenges in replicating radiomics-based ML models and stress the need for rigorous validation.

Repeatability of radiomic features in myocardial T1 and T2 mapping

PURPOSE

To investigate the test-retest repeatability of radiomic features in myocardial native T1 and T2 mapping.

METHODS

In this prospective study, 50 healthy volunteers (29 women and 21 men, mean age 39.4 ± 13.7 years) underwent two identical cardiac magnetic resonance imaging (MRI) examinations at 1.5 T. The protocol included native T1 and T2 mapping in both short-axis and long-axis orientation. For T1 mapping, we investigated standard (1.9 × 1.9 mm) and high (1.4 × 1.4 mm) spatial resolution. After manual segmentation of the left ventricular myocardium, 100 radiomic features from seven feature classes were extracted and analyzed. Test-retest repeatability of radiomic features was assessed using the intraclass correlation coefficient (ICC) and classified as poor (ICC < 0.50), moderate (0.50-0.75), good (0.75-0.90), and excellent (> 0.90).

RESULTS

For T1 maps acquired in short-axis orientation at standard resolution, repeatability was excellent for 6 features, good for 29 features, moderate for 19 features, and poor for 46 features. We identified 15 features from 6 classes which showed good to excellent reproducibility for T1 mapping in all resolutions and all orientations. For short-axis T2 maps, repeatability was excellent for 6 features, good for 25 features, moderate for 23 features, and poor for 46 features. 12 features from 5 classes were found to have good to excellent repeatability in T2 mapping independent of slice orientation.

CONCLUSION

We have identified a subset of features with good to excellent repeatability independent of slice orientation and spatial resolution. We recommend using these features for further radiomics research in myocardial T1 and T2 mapping.

KEY POINTS

Question The study addresses the need for reliable radiomic features for quantitative analysis of the myocardium to ensure diagnostic consistency in cardiac MRI. Findings We have identified a subset of radiomic features demonstrating good to excellent repeatability in native T1 and T2 mapping independent of slice orientation and resolution. Clinical relevanceRadiomic features have been proposed as diagnostic and prognostic biomarkers in various heart diseases. By identifying a subset of particularly reproducible radiomic features our study serves to inform the selection of radiomic features in future research and clinical applications.

Radiomics signature for automatic hydronephrosis detection in unenhanced Low-Dose CT

PURPOSE

To investigate the diagnostic performance of an automatic pipeline for detection of hydronephrosis on kidney's parenchyma on unenhanced low-dose CT of the abdomen.

METHODS

This retrospective study included 95 patients with confirmed unilateral hydronephrosis in an unenhanced low-dose CT of the abdomen. Data were split into training (n = 67) and test (n = 28) cohorts. Both kidneys for each case were included in further analyses, whereas the kidney without hydronephrosis was used as control. Using the training cohort, we developed a pipeline consisting of a deep-learning model for automatic segmentation (a Convolutional Neural Network based on nnU-Net architecture) of the kidney's parenchyma and a radiomics classifier to detect hydronephrosis. The models were assessed using standard classification metrics, such as area under the ROC curve (AUC), sensitivity and specificity, as well as semantic segmentation metrics, including Dice coefficient and Jaccard index.

RESULTS

Using manual segmentation of the kidney's parenchyma, hydronephrosis can be detected with an AUC of 0.84, a sensitivity of 75% and a specificity of 82%, a PPV of 81% and a NPV of 77%. Automatic kidney segmentation achieved a mean Dice score of 0.87 and 0.91 for the right and left kidney, respectively. Additionally, automatic segmentation achieved an AUC of 0.83, a sensitivity of 86%, specificity of 64%, PPV of 71%, and NPV of 82%.

CONCLUSION

Our proposed radiomics signature using automatic kidney's parenchyma segmentation allows for accurate hydronephrosis detection on unenhanced low-dose CT scans of the abdomen independently of widened renal pelvis. This method could be used in clinical routine to highlight hydronephrosis to radiologists as well as clinicians, especially in patients with concurrent parapelvic cysts and might reduce time and costs associated with diagnosing hydronephrosis.

A novel radiomics approach for predicting TACE outcomes in hepatocellular carcinoma patients using deep learning for multi-organ segmentation

Transarterial chemoembolization (TACE) represent the standard of therapy for non-operative hepatocellular carcinoma (HCC), while prediction of long term treatment outcomes is a complex and multifactorial task. In this study, we present a novel machine learning approach utilizing radiomics features from multiple organ volumes of interest (VOIs) to predict TACE outcomes for 252 HCC patients. Unlike conventional radiomics models requiring laborious manual segmentation limited to tumoral regions, our approach captures information comprehensively across various VOIs using a fully automated, pretrained deep learning model applied to pre-TACE CT images. Evaluation of radiomics random survival forest models against clinical ones using Cox proportional hazard demonstrated comparable performance in predicting overall survival. However, radiomics outperformed clinical models in predicting progression-free survival. Explainable analysis highlighted the significance of non-tumoral VOI features, with their cumulative importance superior to features from the largest liver tumor. The proposed approach overcomes the limitations of manual VOI segmentation, requires no radiologist input and highlight the clinical relevance of features beyond tumor regions. Our findings suggest the potential of this radiomics models in predicting TACE outcomes, with possible implications for other clinical scenarios.

Perfect Match: Radiomics and Artificial Intelligence in Cardiac Imaging

Cardiovascular diseases remain a significant health burden, with imaging modalities like echocardiography, cardiac computed tomography, and cardiac magnetic resonance imaging playing a crucial role in diagnosis and prognosis. However, the inherent heterogeneity of these diseases poses challenges, necessitating advanced analytical methods like radiomics and artificial intelligence. Radiomics extracts quantitative features from medical images, capturing intricate patterns and subtle variations that may elude visual inspection. Artificial intelligence techniques, including deep learning, can analyze these features to generate knowledge, define novel imaging biomarkers, and support diagnostic decision-making and outcome prediction. Radiomics and artificial intelligence thus hold promise for significantly enhancing diagnostic and prognostic capabilities in cardiac imaging, paving the way for more personalized and effective patient care. This review explores the synergies between radiomics and artificial intelligence in cardiac imaging, following the radiomics workflow and introducing concepts from both domains. Potential clinical applications, challenges, and limitations are discussed, along with solutions to overcome them.

Imaging-based risk stratification of patients with pulmonary embolism based on dual-energy CT-derived radiomics

BACKGROUND

Technological progress in the acquisition of medical images and the extraction of underlying quantitative imaging data has introduced exciting prospects for the diagnostic assessment of a wide range of conditions. This study aims to investigate the diagnostic utility of a machine learning classifier based on dual-energy computed tomography (DECT) radiomics for classifying pulmonary embolism (PE) severity and assessing the risk for early death.

METHODS

Patients who underwent CT pulmonary angiogram (CTPA) between January 2015 and March 2022 were considered for inclusion in this study. Based on DECT imaging, 107 radiomic features were extracted for each patient using standardized image processing. After dividing the dataset into training and test sets, stepwise feature reduction based on reproducibility, variable importance and correlation analyses were performed to select the most relevant features; these were used to train and validate the gradient-boosted tree models.

RESULTS

The trained machine learning classifier achieved a classification accuracy of .90 for identifying high-risk PE patients with an area under the receiver operating characteristic curve of .59. This CT-based radiomics signature showed good diagnostic accuracy for risk stratification in individuals presenting with central PE, particularly within higher risk groups.

CONCLUSION

Models utilizing DECT-derived radiomics features can accurately stratify patients with pulmonary embolism into established clinical risk scores. This approach holds the potential to enhance patient management and optimize patient flow by assisting in the clinical decision-making process. It also offers the advantage of saving time and resources by leveraging existing imaging to eliminate the necessity for manual clinical scoring.

Post-mortem CT radiomics for the prediction of time since death

OBJECTIVES

Post-mortem interval (PMI) estimation has long been relying on sequential post-mortem changes on the body as a function of extrinsic, intrinsic, and environmental factors. Such factors are difficult to account for in complicated death scenes; thus, PMI estimation can be compromised. Herein, we aimed to evaluate the use of post-mortem CT (PMCT) radiomics for the differentiation between early and late PMI.

METHODS

Consecutive whole-body PMCT examinations performed between 2016 and 2021 were retrospectively included (n = 120), excluding corpses without an accurately reported PMI (n = 23). Radiomics data were extracted from liver and pancreas tissue and randomly split into training and validation sets (70:30%). Following data preprocessing, significant features were selected (Boruta selection) and three XGBoost classifiers were built (liver, pancreas, combined) to differentiate between early (< 12 h) and late (> 12 h) PMI. Classifier performance was assessed with receiver operating characteristics (ROC) curves and areas under the curves (AUC), which were compared by bootstrapping.

RESULTS

A total of 97 PMCTs were included, representing individuals (23 females and 74 males) with a mean age of 47.1 ± 23.38 years. The combined model achieved the highest AUC reaching 75% (95%CI 58.4-91.6%) (p = 0.03 compared to liver and p = 0.18 compared to pancreas). The liver-based and pancreas-based XGBoost models achieved AUCs of 53.6% (95%CI 34.8-72.3%) and 64.3% (95%CI 46.7-81.9%) respectively (p > 0.05 for the comparison between liver- and pancreas-based models).

CONCLUSION

The use of radiomics analysis on PMCT examinations differentiated early from late PMI, unveiling a novel image-based method with important repercussions in forensic casework.

CLINICAL RELEVANCE STATEMENT

This paper introduces the employment of radiomics in forensic diagnosis by presenting an effective automated alternative method of estimating post-mortem interval from targeted tissues, thus paving the way for improvement in speed and quality of forensic investigations.

KEY POINTS

• A combined liver-pancreas radiomics model differentiated early from late post-mortem intervals (using a 12-h threshold) with an area under the curve of 75% (95%CI 58.4-91.6%). • XGBoost models based on liver-only or pancreas-only radiomics demonstrated inferior performance to the combined model in predicting the post-mortem interval.

Fusion-based tensor radiomics using reproducible features: Application to survival prediction in head and neck cancer

BACKGROUND

Numerous features are commonly generated in radiomics applications as applied to medical imaging, and identification of robust radiomics features (RFs) can be an important step to derivation of reliable, reproducible solutions. In this work, we utilize a tensor radiomics (TR) framework, where numerous fusions are explored, to generate different flavours of RFs, and we aimed to identify RFs that are robust to fusion techniques in head and neck cancer. Overall, we aimed to predict progression-free survival (PFS) using Hybrid Machine Learning Systems (HMLS) and reproducible RFs.

METHODS

The study was performed on 408 patients with head and neck cancer from The Cancer Imaging Archive. After image preprocessing, 15 fusion techniques were employed to combine Positron Emission Tomography (PET) and Computed Tomography (CT) images. Subsequently, 215 RFs were extracted through a standardized radiomics software, with 17 'flavours' generated using PET-only, CT-only, and 15 fused PET&CT images. The variability of RFs across flavours was studied using the Intraclass Correlation Coefficient (ICC). Furthermore, the features were categorized into seven reliability groups, 106 reproducible RFs with ICC>0.75 were selected, highly correlated flavours were removed, Principal Component Analysis was used to convert 17 flavours to 1 attribute, the polynomial function was utilized to increase RFs, and Analysis of variance (ANOVA) was used to select the relevant attributes. Finally, 3 classifiers including Random Forest (RFC), Logistic regression (LR), and Multi-layer perceptron were applied to the preselected relevant attributes to predict binary PFS. In 5-fold cross-validation, 80% of 4 divisions were utilized to train the model, and the remaining 20% was utilized to evaluate the model. Further, the remaining fold was used for external nested testing.

RESULTS

Reliability analysis indicated that most morphological features belong to the high-reliability category. By contrast, local intensity and statistical features extracted from images belong to the low-reliability category. In the tensor framework, the highest 5-fold cross-validation accuracy of 76.7%±3.3% with an external nested testing of 70.6%±6.7% resulted from the reproducible TR+polynomial function+ANOVA+LR algorithm while the accuracy of 70.0%±4.2% with the external nested testing of 67.7%±4.9% was achieved through the PCA fusion+RFC (non-tensor paradigm).

CONCLUSIONS

This study demonstrated that using reproducible RFs as utilized within a tensor fusion radiomics framework, linked with ANOVA and LR, added value to prediction of progression-free survival outcome in head and neck cancer patients.

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

OBJECTIVES

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.

METHODS

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).

RESULTS

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).

CONCLUSION

Both traditional radiomics features and transfer-learning deep radiomics features provide relevant and complementary information for non-invasive N-staging in lung cancer.

QuantImage v2: a comprehensive and integrated physician-centered cloud platform for radiomics and machine learning research

BACKGROUND

Radiomics, the field of image-based computational medical biomarker research, has experienced rapid growth over the past decade due to its potential to revolutionize the development of personalized decision support models. However, despite its research momentum and important advances toward methodological standardization, the translation of radiomics prediction models into clinical practice only progresses slowly. The lack of physicians leading the development of radiomics models and insufficient integration of radiomics tools in the clinical workflow contributes to this slow uptake.

METHODS

We propose a physician-centered vision of radiomics research and derive minimal functional requirements for radiomics research software to support this vision. Free-to-access radiomics tools and frameworks were reviewed to identify best practices and reveal the shortcomings of existing software solutions to optimally support physician-driven radiomics research in a clinical environment.

RESULTS

Support for user-friendly development and evaluation of radiomics prediction models via machine learning was found to be missing in most tools. QuantImage v2 (QI2) was designed and implemented to address these shortcomings. QI2 relies on well-established existing tools and open-source libraries to realize and concretely demonstrate the potential of a one-stop tool for physician-driven radiomics research. It provides web-based access to cohort management, feature extraction, and visualization and supports "no-code" development and evaluation of machine learning models against patient-specific outcome data.

CONCLUSIONS

QI2 fills a gap in the radiomics software landscape by enabling "no-code" radiomics research, including model validation, in a clinical environment. Further information about QI2, a public instance of the system, and its source code is available at https://medgift.github.io/quantimage-v2-info/ . Key points As domain experts, physicians play a key role in the development of radiomics models. Existing software solutions do not support physician-driven research optimally. QuantImage v2 implements a physician-centered vision for radiomics research. QuantImage v2 is a web-based, "no-code" radiomics research platform.

Radiomics for the Prediction of Overall Survival in Patients with Bladder Cancer Prior to Radical Cystectomy

(1) Background: To evaluate radiomics features as well as a combined model with clinical parameters for predicting overall survival in patients with bladder cancer (BCa). (2) Methods: This retrospective study included 301 BCa patients who received radical cystectomy (RC) and pelvic lymphadenectomy. Radiomics features were extracted from the regions of the primary tumor and pelvic lymph nodes as well as the peritumoral regions in preoperative CT scans. Cross-validation was performed in the training cohort, and a Cox regression model with an elastic net penalty was trained using radiomics features and clinical parameters. The models were evaluated with the time-dependent area under the ROC curve (AUC), Brier score and calibration curves. (3) Results: The median follow-up time was 56 months (95% CI: 48−74 months). In the follow-up period from 1 to 7 years after RC, radiomics models achieved comparable predictive performance to validated clinical parameters with an integrated AUC of 0.771 (95% CI: 0.657−0.869) compared to an integrated AUC of 0.761 (95% CI: 0.617−0.874) for the prediction of overall survival (p = 0.98). A combined clinical and radiomics model stratified patients into high-risk and low-risk groups with significantly different overall survival (p < 0.001). (4) Conclusions: Radiomics features based on preoperative CT scans have prognostic value in predicting overall survival before RC. Therefore, radiomics may guide early clinical decision-making.