To systematically identify radiomics features on CT enterography (CTE) scans which can accurately diagnose active Crohn's disease across multiple sources of variation. Retrospective study of CTE scans curated between 2013 and 2015, comprising 164 subjects (65 male, 99 female; all patients were over the age of 18) with endoscopic confirmation for the presence or absence of active Crohn's disease. All patients had three distinct sets of scans available (full and reduced dose, where the latter had been reconstructed via two different methods), acquired on a single scanner at a single institution. Radiomics descriptors from annotated terminal ileum regions were individually and systematically evaluated for resilience to different imaging variations (changes in dose/reconstruction, batch effects, and simulated annotation differences) via multiple reproducibility measures. Multiple radiomics models (by accounting for each source of variation) were evaluated in terms of classifier area under the ROC curve (AUC) for identifying patients with active Crohn's disease, across separate discovery and hold-out validation cohorts. Radiomics descriptors selected based on resiliency to multiple sources of imaging variation yielded the highest overall classification performance in the discovery cohort (AUC = 0.79 ± 0.04) which also best generalized in hold-out validation (AUC = 0.81). Performance was maintained across multiple doses and reconstructions while also being significantly better (p < 0.001) than non-resilient descriptors or descriptors only resilient to a single source of variation. Radiomics features can accurately diagnose active Crohn's disease on CTE scans across multiple sources of imaging variation via systematic analysis of reproducibility measures. Clinical utility and translatability of radiomics features for diagnosis and characterization of Crohn's disease on CTE scans will be contingent on their reproducibility across multiple types and sources of imaging variation.

Radiomics allows for the quantification of medical images and facilitates precision medicine. Many radiomic features derived from computed tomography (CT) are sensitive to variations across scanners, reconstruction settings, and acquisition protocols. In this phantom study, eight different CT reconstruction parameters were varied to explore image- and feature-level harmonization approaches to improve tissue classification.

Varying reconstructions of an anthropomorphic radiopaque phantom containing three lesion categories (metastasis, hemangioma, and benign cyst) and normal liver tissue were used for evaluating two harmonization methods and their combination: (i) generative adversarial networks (GANs) at the image level; (ii) ComBat at the feature level, and (iii) a combination of (i) and (ii). A total of 93 texture and intensity features were extracted from each tissue class before and after image-level harmonization and were also harmonized at the feature level. Reproducibility and stability were assessed via the Concordance Correlation Coefficient (CCC) and pairwise comparisons using paired stability tests. The ability of features to discriminate between tissue classes was assessed by measuring the area under the receiver operating characteristic curve. The global reproducibility and discriminative power were assessed by averaging over the entire dataset and across all tissue types.

ComBat improved reproducibility by 31.58% and stability by 5.24%, while GAN increased reproducibility by 8% it reduced stability by 4.33%. Classification analysis revealed that ComBat increased average AUC by 15.19%, whereas GAN decreased AUC by 2.56%.

While GAN qualitatively enhances image harmonization, ComBat provides superior statistical improvements in feature stability and classification performance, highlighting the importance of robust feature-level harmonization in radiomics.

High-grade patterns, visceral pleural invasion, lymphovascular invasion, spread through air spaces, and lymph node metastasis are high-risk factors and associated with poor prognosis in lung adenocarcinomas (LUADs). This study aimed to construct and validate a radiomic model and a radiographic model derived from low-dose CT (LDCT) for predicting high-risk LUADs in solid and part-solid nodules.

This study retrospectively enrolled 658 pathologically confirmed LUADs from July 2018 to December 2022 from four centers, which were divided into training set (n=411), internal validation set (n=139), and external validation set (n=108). Radiomic features and radiographic features including maximal diameter, consolidation/tumor ratio (CTR), and semantic features, were obtained to construct a radiomic model and a radiographic model through multivariable logistic regression. Area under receiver operating characteristic curve (AUC) was utilized to assess the diagnostic performance of the models.

Three radiomic features (GLCM_Correlation, GLSZM_SmallAreaEmphasis, and GLDM_LargeDependenceHighGrayLevelEmphasis) and four radiographic features (maximal diameter, CTR, spiculation, and pleural indentation) were selected to build models. The radiomic model yielded AUCs of 0.916 in the internal validation set and 0.938 in the external validation set, which were significantly higher than the AUCs of the radiographic model (0.916 vs. 0.868, P=0.014 and 0.938 vs. 0.880, P=0.002).

Our LDCT-based radiomic model enabled non-invasive identification of high-risk LUADs in solid and part-solid nodules with good diagnostic performance and might assist in case-specific decision-making in lung cancer screening.

In cardiac MRI, the field of view (FOV) is adapted to the individual patient's size, influencing spatial resolution and myocardial radiomics. This study aimed to investigate the effects of parametric feature maps on radiomics derived from cine images acquired with different FOV sizes on individuals without myocardial pathologies. In the clinical setting, cardiac MRI scans from clinical care were screened retrospectively for patients without pathological findings, neither in the MRI nor the medical history or follow-up, resulting in 61 included patients. In the experimental setting, 12 healthy volunteers were prospectively examined on a 1.5 Tesla MRI scanner with cine images acquired with three different FOVs (256 × 329 mm, 279 × 359 mm, 302 × 390 mm). One midventricular end-diastolic short-axis slice of the non-enhanced cine images was extracted for healthy volunteers and patients. The left ventricular myocardium was encompassed with regions of interest (ROIs). Ninety-three features were extracted using PyRadiomics. Images were converted to parametric radiomic feature maps using pretested software. ROIs were copied to the maps to retrieve the feature quantity. The variability of features across the different FOVs from the original images and feature maps was assessed with coefficients of variation (COVs) and rated stable at up to 10%. When derived from the original images, out of the 93 extracted features, only 24 (patients) and 29 (volunteers) revealed COVs < 10%. When extracted from the parametric maps, the number of stable features increased by 63% and 66%, with 39 (patients) and 48 (volunteers) features showing COVs < 10%, respectively. Software-computed parametric feature maps improve the reproducibility of radiomics across different FOVs in cardiac cine images of individuals without myocardial pathologies. Prospective investigations with different FOVs of a patient collective with myocardial pathologies could enhance the generalizability of the findings.

Stereotactic radiosurgery (SRS) is widely used for managing brain metastases (BMs), but an adverse effect, radionecrosis, complicates post-SRS management. Differentiating radionecrosis from tumor recurrence non-invasively remains a major clinical challenge, as conventional imaging techniques often necessitate surgical biopsy for accurate diagnosis. Machine learning and deep learning models have shown potential in distinguishing radionecrosis from tumor recurrence. However, their clinical adoption is hindered by a lack of explainability, limiting understanding and trust in their diagnostic decisions.

To utilize a novel neural ordinary differential equation (NODE) model for discerning BM post-SRS radionecrosis from recurrence. This approach integrates image-deep features, genomic biomarkers, and non-image clinical parameters within a synthesized latent feature space. The trajectory of each data sample towards the diagnosis decision can be visualized within this feature space, offering a new angle on radiogenomic data analysis foundational for AI explainability.

By hypothesizing that deep feature extraction can be modeled as a spatiotemporally continuous process, we designed a novel model based on heavy ball NODE (HBNODE) in which deep feature extraction was governed by a second-order ODE. This approach enabled tracking of deep neural network (DNN) behavior by solving the HBNODE and observing the stepwise derivative evolution. Consequently, the trajectory of each sample within the Image-Genomic-Clinical (I-G-C) space became traceable. A decision-making field (F) was reconstructed within the feature space, with its gradient vectors directing the data samples' trajectories and intensities showing the potential. The evolution of F reflected the cumulative feature contributions at intermediate states to the final diagnosis, enabling quantitative and dynamic comparisons of the relative contribution of each feature category over time. A velocity curve was designed to determine key intermediate states (locoregional ∇F = 0) that are most predictive. Subsequently, a non-parametric model aggregated the optimal solutions from these key states to predict outcomes. Our dataset included 90 BMs from 62 NSCLC patients, and 3-month post-SRS T1+c MR image features, seven NSCLC genomic features, and seven clinical features were analyzed. An 8:2 train/test assignment was employed, and five independent models were trained to ensure robustness. Performance was benchmarked in sensitivity, specificity, accuracy, and ROCAUC, and results were compared against (1) a DNN using only image-based features, and (2) a combined "I+G+C" features without the HBNODE model.

The temporal evolution of gradient vectors and potential fields in F suggested that clinical features contribute the most during the initial stages of the HBNODE implementation, followed by imagery features taking dominance in the latter ones, while genomic features contribute the least throughout the process. The HBNODE model successfully identified and assembled key intermediate states, exhibiting competitive performance with an ROCAUC of 0.88 ± 0.04, sensitivity of 0.79 ± 0.02, specificity of 0.86 ± 0.01, and accuracy of 0.84 ± 0.01, where the uncertainties represent standard deviations. For comparison, the image-only DNN model achieved an ROCAUC of 0.71 ± 0.05 and sensitivity of 0.66 ± 0.32 (p = 0.086), while the "I+G+C" model without HBNODE reported an ROCAUC of 0.81 ± 0.02 and sensitivity of 0.58 ± 0.11 (p = 0.091).

The HBNODE model effectively identifies BM radionecrosis from recurrence, enhancing explainability within XAI frameworks. Its performance encourages further exploration in clinical settings and suggests potential applicability across various XAI domains.

Intratumoral tertiary lymphoid structures (iTLS) in hepatocellular carcinoma (HCC) are associated with improved survival and may influence treatment decisions. However, their non-invasive detection remains challenging in HCC. We aim to develop a non-invasive model using baseline contrast-enhanced MRI to predict the iTLS status.

A total of 660 patients with HCC who underwent surgery were retrospectively recruited from four centers between October 2015 and January 2023 and divided into training, internal test, and external validation sets. After features dimensionality and selection, corresponding features were used to construct transfer learning radiomic (TLR) models for diagnosing iTLS, and model interpretability was explored with pathway analysis in The Cancer Genome Atlas-Liver HCC. The performances of models were assessed using the area under the receiver operating characteristic curve (AUC). The log-rank test was used to evaluate the prognostic value of the TLR model. The combination therapy set of 101 patients with advanced HCC treated with first-line anti-programmed death 1 or ligand 1 plus antiangiogenic treatment between January 2021 and January 2024 was used to investigate the value of the TLR model for evaluating the treatment response.

The presence of iTLS was identified in 46.0% (n=308) patients. The TLR model demonstrated excellent performance in predicting the presence of iTLS in training (AUC=0.91, 95% CI: 0.87, 0.94), internal test (AUC=0.85, 95% CI: 0.77, 0.93) and external validation set (AUC=0.85, 95% CI: 0.81, 0.90). The TLR model-predicted iTLS group has favorable overall survival (HR=0.66; 95% CI: 0.48, 0.90; p=0.007) and relapse-free survival (HR=0.64; 95% CI: 0.48, 0.85; p=0.001) in the external validation set. The model-predicted iTLS status was associated with inflammatory response and specific tumor-associated signaling activation (all p<0.001). The proportion of treatment responders was significantly higher in the model-predicted group with iTLS than in the group without iTLS (36% vs 13.73%, p=0.009).

The TLR model has indicated accurate prediction of iTLS status, which may assist in the risk stratification for patients with HCC in clinical practice.

The reproducibility of radiomic features (RFs) is essential in lung nodule diagnosis. This study aimed to prospectively investigate the effects of computed tomography (CT) scanning parameters on the reproducibility of RFs in pulmonary nodules.

Patients with pulmonary nodules who underwent chest CT scans at the Cancer Hospital of the Chinese Academy of Medical Sciences between July 2018 and March 2019 were prospectively included in the study. Six sequences with three pairs of different scanning parameters, including the reconstruction algorithm [filtered back projection (FBP) vs. 50% adaptive statistical iterative reconstruction-V (ASiR-V)], radiation dose (low dose vs. standard dose), and contrast agent [contrast-enhanced (CE) CT vs. non-contrast enhanced (NE) CT], were used for each patient. When one of the scanning parameters was changed, the other two remained fixed. The nodules were classified into pure ground-glass nodules (pGGNs), part-solid nodules (PSNs), and solid nodules (SNs) according to the nodule consistency. RFs with an intraclass correlation coefficient (ICC) >0.75 were considered to have good retest reliability. All the RF values of the different scanning parameters and nodule consistency were investigated and compared.

A total of 150 pulmonary nodules, including 50 pGGNs, 50 PSNs, and 50 SNs, in 96 patients (mean age: 52±10 years; 62 females) were included in the study. In total, 320 RFs with an ICC >0.75 were evaluated. The proportion of RFs showed significant difference between FBP and 50% ASiR-V, low dose and standard dose, and CE and NE CT scans was 38.4% (123/320), 63.1% (202/320) and 54.1% (173/320), respectively. The radiation dose and contrast agent affected more RFs than the reconstruction algorithm (both P<0.001). In the subgroup analysis of nodule consistency, regardless of changes in the reconstruction algorithms, radiation doses, or contrast agents, the RFs showed significant difference among the pGGNs, PSNs, and SNs (all P<0.001).

The scanning parameters affected the reproducibility of the RFs, and nodules of different consistency were affected differently. The effects of these parameters should be fully considered in radiomic analysis.

Radiomics models trained on data from one center typically show a decline of performance when applied to data from external centers, hindering their introduction into large-scale clinical practice. Current expert recommendations suggest to use only reproducible radiomics features isolated by multiscanner test-retest experiments, which might help to overcome the problem of limited generalizability to external data.

To evaluate the influence of using only a subset of robust radiomics features, defined in a prior in vivo multi-MRI-scanner test-retest-study, on the performance and generalizability of radiomics models.

Retrospective.

Patients with monoclonal plasma cell disorders. Training set (117 MRIs from center 1); internal test set (42 MRIs from center 1); external test set (143 MRIs from center 2-8).

1.5T and 3.0T; T1-weighted turbo spin echo.

The task for the radiomics models was to predict plasma cell infiltration, determined by bone marrow biopsy, noninvasively from MRI. Radiomics machine learning models, including linear regressor, support vector regressor (SVR), and random forest regressor (RFR), were trained on data from center 1, using either all radiomics features, or using only reproducible radiomics features. Models were tested on an internal (center 1) and a multicentric external data set (center 2-8).

Pearson correlation coefficient r and mean absolute error (MAE) between predicted and actual plasma cell infiltration. Fisher's z-transformation, Wilcoxon signed-rank test, Wilcoxon rank-sum test; significance level P < 0.05.

When using only reproducible features compared with all features, the performance of the SVR on the external test set significantly improved (r = 0.43 vs. r = 0.18 and MAE = 22.6 vs. MAE = 28.2). For the RFR, the performance on the external test set deteriorated when using only reproducible instead of all radiomics features (r = 0.33 vs. r = 0.44, P = 0.29 and MAE = 21.9 vs. MAE = 20.5, P = 0.10).

Using only reproducible radiomics features improves the external performance of some, but not all machine learning models, and did not automatically lead to an improvement of the external performance of the overall best radiomics model.

Stage 2.

Conventional methods for detecting lung cancer early are often qualitative and subject to interpretation. Radiomics provides quantitative characteristics of pulmonary nodules (PNs) in medical images, but variability in medical image acquisition is an obstacle to consistent clinical application of these quantitative features. Correcting radiomic features' dependency on acquisition parameters is problematic when combining data from benign and malignant PNs, as is necessary when the goal is to diagnose lung cancer, because acquisition effects may differ between them due to their biological differences.

We evaluated whether we must account for biological differences between benign and malignant PNs when correcting the dependency of radiomic features on acquisition parameters, and we compared methods of doing this using ComBat harmonization.

This study used a dataset of 567 clinical chest CT scans containing both malignant and benign PNs. Scans were grouped as benign, malignant, or lung cancer screening (mixed benign and malignant). Preprocessing and feature extraction from ROIs were performed using PyRadiomics. Optimized Permutation Nested ComBat harmonization was performed on extracted features to account for variability in four imaging protocols: contrast enhancement, scanner manufacturer, acquisition voltage, focal spot size. Three methods were compared: harmonizing all data collectively in the standard manner, harmonizing all data with a covariate to preserve distinctions between subgroups, and harmonizing subgroups separately. A significant (p ≤ 0.05) Kruskal-Wallis test determined whether harmonization removed a feature's dependency on an acquisition parameter. A LASSO-SVM pipeline was trained using acquisition-independent radiomic features to predict whether PNs were malignant or benign. To evaluate the predictive information made available by each harmonization method, the trained harmonization estimators and predictive model were applied to a corresponding unseen test set. Harmonization and predictive performance metrics were assessed over 10 trials of 5-fold cross validation.

Kruskal-Wallis defined an average 2.1% of features (95% CI: 1.9-2.4%) as acquisition-independent when data were harmonized collectively, 27.3% of features (95% CI: 25.7-28.9%) as acquisition-independent when harmonized with a covariate, and 90.9% of features (95% CI: 90.4-91.5%) as acquisition-independent when harmonized separately. LASSO-SVM models trained on data harmonized separately or with a covariate had higher ROC-AUC for lung cancer screening scans than models trained on data harmonized without distinction between benign and malignant tissues (Delong test, Holm-Bonferroni adjusted p ≤ 0.05). There was not a conclusive difference in ROC-AUC between models trained on data harmonized separately and models trained on data harmonized with a covariate.

Radiomic features of benign and malignant PNs require different corrective transformations to recover acquisition-independent distributions. This can be done using separate harmonization or harmonization with a covariate. Separate harmonization enabled the greatest number of predictive features to be used in a machine learning model to retrospectively detect lung cancer. Features harmonized separately and features harmonized with a covariate enabled predictive models to achieve similar performance on lung cancer screening scans.

Objective. Sarcoidosis is a granulomatous disease affecting the lungs in over 90% of patients. Qualitative assessment of chest CT by radiologists is standard clinical practice and reliable quantification of disease from CT would support ongoing efforts to identify sarcoidosis phenotypes. Standard imaging feature engineering techniques such as radiomics suffer from extreme sensitivity to image acquisition and processing, potentially impeding generalizability of research to clinical populations. In this work, we instead investigate approaches to engineering variogram-based features with the intent to identify a robust, generalizable pipeline for image quantification in the study of sarcoidosis.Approach. For a cohort of more than 300 individuals with sarcoidosis, we investigated 24 feature engineering pipelines differing by decisions for image registration to a template lung, empirical and model variogram estimation methods, and feature harmonization for CT scanner model, and subsequently 48 sets of phenotypes produced through unsupervised clustering. We then assessed sensitivity of engineered features, phenotypes produced through unsupervised clustering, and sarcoidosis disease signal strength to pipeline.Main results. We found that variogram features had low to mild association with scanner model and associations were reduced by image registration. For each feature type, features were also typically robust to all pipeline decisions except image registration. Strength of disease signal as measured by association with pulmonary function testing and some radiologist visual assessments was strong (optimistic AUC ≈ 0.9,p≪0.0001in models for architectural distortion, conglomerate mass, fibrotic abnormality, and traction bronchiectasis) and fairly consistent across engineering approaches regardless of registration and harmonization for CT scanner.Significance. Variogram-based features appear to be a suitable approach to image quantification in support of generalizable research in pulmonary sarcoidosis.

Tumor fibrosis plays an important role in chemotherapy resistance in pancreatic ductal adenocarcinoma (PDAC); however, there remains a contradiction in the prognostic value of fibrosis. The authors aimed to investigate the relationship between tumor fibrosis and survival in patients with PDAC, classify patients into high- and low-fibrosis groups, and develop and validate a CT-based radiomics model to non-invasively predict fibrosis before treatment.

This retrospective, bicentric study included 295 patients with PDAC without any treatments before surgery. Tumor fibrosis was assessed using the collagen fraction (CF). Cox regression analysis was used to evaluate the associations of CF with overall survival (OS) and disease-free survival (DFS). Receiver operating characteristic (ROC) analyses were used to determine the rounded threshold of CF. An integrated model (IM) was developed by incorporating selected radiomic features and clinical-radiological characteristics. The predictive performance was validated in the test cohort (Center 2).

The CFs were 38.22±6.89% and 38.44±8.66% in center 1 (131 patients, 83 males) and center 2 (164 patients, 100 males), respectively ( P =0.814). Multivariable Cox regression revealed that CF was an independent risk factor in the OS and DFS analyses at both centers. ROCs revealed that 40% was the rounded cut-off value of CF. IM predicted CF with areas under the curves (AUCs) of 0.829 (95% CI: 0.753-0.889) and 0.751 (95% CI: 0.677-0.815) in the training and test cohorts, respectively. Decision curve analyses revealed that IM outperformed radiomics model and clinical-radiological model for CF prediction in both cohorts.

Tumor fibrosis was an independent risk factor for survival of patients with PDAC, and a rounded cut-off value of 40% provided a good differentiation of patient prognosis. The model combining CT-based radiomics and clinical-radiological features can satisfactorily predict survival-grade fibrosis in patients with PDAC.

This study developed and validated a novel deep learning radiomic biomarker to estimate response to immune checkpoint inhibitor (ICI) therapy in advanced non-small cell lung cancer (NSCLC) using real-world data (RWD) and clinical trial data.

Retrospective RWD of 1,829 patients with advanced NSCLC treated with PD-(L)1 ICIs were collected from 10 academic and community institutions in the United States and Europe. The RWD included data sets for discovery (Data Set A-Discovery, n = 1,173) and independent test (Data Set B, n = 458). A radiomic pipeline, containing a deep learning feature extractor and a survival model, generated the computed tomography (CT) response score (CTRS) applied to the pretreatment routine CT/positron emission tomography (PET)-CT scan. An enhanced CTRS (eCTRS) also incorporated age, sex, treatment line, and lesion annotations. Performance was evaluated against progression-free survival (PFS) and overall survival (OS). Biomarker generalizability was further evaluated using a secondary analysis of a prospective clinical trial (ClinicalTrials.gov identifier: NCT02573259) evaluating the PD-1 inhibitor sasanlimab in second or later line of treatment (Data Set C, n = 54).

In RWD Test Data Set B, the CTRS identified patients with a high probability of response to ICI with a PFS hazard ratio (HR) of 0.46 (95% CI, 0.26 to 0.82) and an OS HR of 0.50 (95% CI, 0.28 to 0.92) in the first-line ICI monotherapy cohort, after adjustment for baseline covariates including the PD-L1 tumor proportion score. In Clinical Trial Data Set C, the CTRS demonstrated an adjusted PFS HR of 1.03 (95% CI, 0.43 to 2.47) and an OS HR of 0.33 (95% CI, 0.14 to 0.91). The CTRS and eCTRS outperformed traditional imaging biomarkers of lesion size in PFS and OS for RWD Test Data Set B and in OS for the Clinical Trial Data Set.

The study developed and validated a deep learning radiomic biomarker using pretreatment routine CT/PET-CT scans to identify ICI benefit in advanced NSCLC.

We assessed the feasibility of using deep learning-based image harmonization to improve the reproducibility of radiomics features in abdominal CT scans. In CT imaging, harmonization adjusts images from different institutions to ensure consistency despite variations in scanners and acquisition protocols. This process is essential because such differences can lead to variability in radiomics features, affecting reproducibility and accuracy. Harmonizing images minimizes these inconsistencies, supporting more reliable and clinically applicable results across diverse settings. A pre-trained harmonization algorithm was applied to 63 dual-energy abdominal CT images, which were reconstructed into four different types, and 10 regions of interest (ROIs) were analyzed. From the original 455 radiomics features per ROI, 387 were used after excluding redundant features. Reproducibility was measured using the intraclass correlation coefficient (ICC), with a threshold of ICC ≥ 0.85 indicating acceptable reproducibility. The region-based analysis revealed significant improvements in reproducibility post-harmonization, especially in vessel features, which increased from 14% to 69%. Other regions, including the spleen, kidney, muscle, and liver parenchyma, also saw notable improvements, although air reproducibility slightly decreased from 95% to 94%, impacting only a few features. In patient-based analysis, reproducible features increased from 18% to 65%, with an average of 179 additional reproducible features per patient after harmonization. These results demonstrate that deep learning-based harmonization can significantly enhance the reproducibility of radiomics features in abdominal CT, offering promising potential for advancing radiomics development and its clinical applications.

High-resolution CT (HRCT) plays an important role in diagnosing and monitoring interstitial lung diseases (ILDs). Despite advances, predicting disease progression and treatment response remains challenging. HRCT enables noninvasive visualization and classification of patterns of lung injury and assessment of disease extent. Visual estimation of CT extent of fibrotic lung disease is an independent predictor of mortality and progression, but is subjective, with only modest interobserver agreement for radiologic interpretation of ILD. Machine learning-based textural analysis of fibrosis extent on baseline and serial HRCT scans shows robust correlations with physiologic measures and strong association with risk of disease progression or mortality across various fibrosing ILDs. In idiopathic pulmonary fibrosis, quantitative CT (QCT) assessment is associated with physiologic impairment and risk of progression and death, and increasing severity of fibrosis on longitudinal evaluation is associated with increased risk of progression and death. Similar results have been noted for fibrotic hypersensitivity pneumonitis and connective tissue disease. This review focuses on QCT techniques for ILDs. We describe the clinical need for quantification of lung disease and illustrate the role of conventional visual evaluation and of QCT approaches in defining disease severity, prognosis, and longitudinal progression, both in established disease and in preclinical interstitial abnormality.

Quantitative imaging biomarkers (QIB) are increasingly used in clinical research to advance precision medicine approaches in oncology. Computed tomography (CT) is a modality of choice for cancer diagnosis, prognosis, and response assessment due to its reliability and global accessibility. Here, we contribute to the cancer imaging community through The Cancer Imaging Archive (TCIA) by providing investigator-initiated, same-day repeat CT scan images of 32 non-small cell lung cancer (NSCLC) patients, along with radiologist-annotated lesion contours as a reference standard. Each scan was reconstructed into 6 image settings using various combinations of three slice thicknesses (1.25 mm, 2.5 mm, 5 mm) and two reconstruction kernels (lung, standard; GE CT equipment), which spans a wide range of CT imaging reconstruction parameters commonly used in lung cancer clinical practice and clinical trials. This holds considerable value for advancing the development of robust Radiomics, Artificial Intelligence (AI) and machine learning (ML) methods.

Generative artificial intelligence (AI) has been applied to images for image quality enhancement, domain transfer, and augmentation of training data for AI modeling in various medical fields. Image-generative AI can produce large amounts of unannotated imaging data, which facilitates multiple downstream deep-learning tasks. However, their evaluation methods and clinical utility have not been thoroughly reviewed. This article summarizes commonly used generative adversarial networks and diffusion models. In addition, it summarizes their utility in clinical tasks in the field of radiology, such as direct image utilization, lesion detection, segmentation, and diagnosis. This article aims to guide readers regarding radiology practice and research using image-generative AI by 1) reviewing basic theories of image-generative AI, 2) discussing the methods used to evaluate the generated images, 3) outlining the clinical and research utility of generated images, and 4) discussing the issue of hallucinations.

This narrative review explores texture analysis as a valuable technique in dentomaxillofacial diagnosis, providing an advanced method for quantification and characterization of different image modalities. The traditional imaging techniques rely primarily on visual assessment, which may overlook subtle variations in tissue structure. In contrast, texture analysis uses sophisticated algorithms to extract quantitative information from imaging data, thus offering deeper insights into the spatial distribution and relationships of pixel intensities. This process identifies unique "texture signatures", serving as markers for accurately characterizing tissue changes or pathological processes. The synergy between texture analysis and radiomics allows radiologists to transcend traditional size-based or semantic descriptors, offering a comprehensive understanding of imaging data. This method enhances diagnostic accuracy, particularly for the assessment of oral and maxillofacial pathologies. The integration of texture analysis with radiomics expands the potential for precise tissue characterization by moving beyond the limitations of human eye evaluations. This article reviews the current trends and methodologies in texture analysis within the field of dentomaxillofacial imaging, highlights its practical applications, and discusses future directions for research and dental clinical practice.

To systematically review the literature on radiomics for predicting intracranial aneurysm rupture and conduct a meta-analysis to obtain evidence confirming the value of radiomics in this prediction.

A systematic literature search was conducted in PubMed, Web of Science, Embase, and The Cochrane Library databases up to March 2024. The QUADAS-2 tool was used to assess study quality. Stata 15.0 and Review Manager 5.4.1 were used for statistical analysis. Outcomes included combined sensitivity (Sen), specificity (Spe), positive likelihood ratio (+LR), negative likelihood ratio (-LR), diagnostic odds ratio (DOR), and their 95% confidence intervals (CI), as well as pre-test and post-test probabilities. The SROC curve was plotted, and the area under the curve (AUC) was calculated. Publication bias and small-study effects were assessed using the Deeks' funnel plot.

The 9 included studies reported 4,284 patients, with 1,411 patients with intracranial aneurysm rupture (prevalence 32.9%). The overall performance of radiomics for predicting intracranial aneurysm rupture showed a combined Sen of 0.78 (95% CI: 0.74-0.82), Spe of 0.74 (95% CI: 0.70-0.78), +LR of 3.0 (95% CI: 2.7-3.4), -LR of 0.29 (95% CI: 0.25-0.35), DOR of 10 (95% CI: 9-12), and AUC of 0.83 (95% CI: 0.79-0.86). Significant heterogeneity was observed in both Sen (I2 = 90.93, 95% CI: 89.00-92.87%) and Spe (I2 = 94.28, 95% CI: 93.21-95.34%).

Radiomics can improve the diagnostic efficacy of intracranial aneurysm rupture. More large-sample, prospective, multicenter clinical studies are needed to further evaluate its predictive value.

https://www.crd.york.ac.uk/PROSPERO/.

This study aims to externally validate a predictive model for distant metastasis (DM) with computed tomography (CT)-based radiomics features in prospectively enrolled non-small-cell lung cancer patients undergoing dynamic tumor-tracking stereotactic body radiation therapy (DTT-SBRT).

The study collected retrospective data from 567 patients across 11 institutions as the training dataset and prospectively enrolled 42 patients from four institutions as the external test dataset. Four clinical features were collected, and 944 CT-based radiomic features were extracted from gross tumor volumes. After standardization and feature selection, DM predictive models were developed using fine and gray regression (FG) and random survival forest (RSF), incorporating clinical and radiomic features, and their combinations within the training dataset. Then, the model was applied to the test dataset, dividing patients into high- and low-risk groups based on medians of risk scores. Model performance was assessed using the concordance index (C-index), and the statistical significance between groups was evaluated using Gray's test.

In the training dataset, 122 of 567 patients (21.5%) developed DM, compared to 9 of 42 patients (21.4%) in the test dataset. In the test dataset, the C-indices of the clinical, radiomics, and hybrid models with FG were 0.559, 0.544, and 0.560, respectively, whereas those with RSF were 0.576, 0.604, and 0.627, respectively. The hybrid model with RSF, which exhibited the best predictive performance of all models, identified 7 of 23 patients (30.4%) as high risk and 2 of 19 patients (10.5%) as low risk for DM incidence in the test dataset (p = 0.116).

Although predictive models for DM lack significance when applied to prospectively enrolled cases undergoing DTT-lung SBRT, the model with RSF exhibits a consistent capacity to effectively classify patients at a high risk of developing DM.

To develop a novel deep ensemble learning model for accurate prediction of brain metastasis (BM) local control outcomes after stereotactic radiosurgery (SRS).

A total of 114 brain metastases (BMs) from 82 patients were evaluated, including 26 BMs that developed biopsy-confirmed local failure post-SRS. The SRS spatial dose distribution (Dmap) of each BM was registered to the planning contrast-enhanced T1 (T1-CE) magnetic resonance imaging (MRI). Axial slices of the Dmap, T1-CE, and planning target volume (PTV) segmentation (PTVseg) intersecting the BM center were extracted within a fixed field of view determined by the 60% isodose volume in Dmap. A spherical projection was implemented to transform planar image content onto a spherical surface using multiple projection centers, and the resultant T1-CE/Dmap/PTVseg projections were stacked as a 3-channel variable. Four Visual Geometry Group (VGG-19) deep encoders were used in an ensemble design, with each submodel using a different spherical projection formula as input for BM outcome prediction. In each submodel, clinical features after positional encoding were fused with VGG-19 deep features to generate logit results. The ensemble's outcome was synthesized from the 4 submodel results via logistic regression. In total, 10 model versions with random validation sample assignments were trained to study model robustness. Performance was compared with (1) a single VGG-19 encoder, (2) an ensemble with a T1-CE MRI as the sole image input after projections, and (3) an ensemble with the same image input design without clinical feature inclusion.

The ensemble model achieved an excellent area under the receiver operating characteristic curve (AUCROC: 0.89 ± 0.02) with high sensitivity (0.82 ± 0.05), specificity (0.84 ± 0.11), and accuracy (0.84 ± 0.08) results. This outperformed the MRI-only VGG-19 encoder (sensitivity: 0.35 ± 0.01, AUCROC: 0.64 ± 0.08), the MRI-only deep ensemble (sensitivity: 0.60 ± 0.09, AUCROC: 0.68 ± 0.06), and the 3-channel ensemble without clinical feature fusion (sensitivity: 0.78 ± 0.08, AUCROC: 0.84 ± 0.03).

Facilitated by the spherical image projection method, a deep ensemble model incorporating Dmap and clinical variables demonstrated excellent performance in predicting BM post-SRS local failure. Our novel approach could improve other radiation therapy outcome models and warrants further evaluation.

To further optimize radiotherapy, a more personalized treatment towards individual patient's risk profiles, dissecting both patient-specific tumor and normal tissue response to multimodality treatments is needed. Novel developments in radiobiology, using in vitro patient-specific complex tissue resembling 3D models and multiomics approaches at a spatial single-cell level, may provide unprecedented insight into the radiation responses of tumors and normal tissue. Here, we describe the necessary team effort, including all disciplines in radiation oncology, to integrate such data into clinical prediction models and link the relatively "big data" from the clinical practice, allowing accurate patient stratification for personalized treatment approaches.

To evaluate radiomics in soft tissue sarcomas (STSs) for diagnostic accuracy, grading, and treatment response assessment, with a focus on clinical relevance.

In this diagnostic accuracy study, radiomics was applied using multiple MRI sequences and AI classifiers, with histopathological diagnosis as the reference standard. Statistical analysis involved meta-analysis, random-effects model, and Deeks' funnel plot asymmetry test.

Among 579 unique titles and abstracts, 24 articles were included in the systematic review, with 21 used for meta-analysis. Radiomics demonstrated a pooled sensitivity of 84% (95% CI: 80-87) and specificity of 63% (95% CI: 56-70), AUC of 0.93 for diagnosis, sensitivity of 84% (95% CI: 82-87) and specificity of 73% (95% CI: 68-77), AUC of 0.91 for grading, and sensitivity of 83% (95% CI: 67-94) and specificity of 67% (95% CI: 59-74), AUC of 0.87 for treatment response assessment.

Radiomics exhibits potential for accurate diagnosis, grading, and treatment response assessment in STSs, emphasizing the need for standardization and prospective trials.

Radiomics offers precise tools for STS diagnosis, grading, and treatment response assessment, with implications for optimizing patient care and treatment strategies in this complex malignancy.

The purpose of this study was to determine and compare the performance of pre-treatment clinical risk score (CRS), radiomics models based on computed (CT), and their combination for predicting time to recurrence (TTR) and disease-specific survival (DSS) in patients with colorectal cancer liver metastases.

We retrospectively analyzed a prospectively maintained registry of 241 patients treated with systemic chemotherapy and surgery for colorectal cancer liver metastases. Radiomics features were extracted from baseline, pre-treatment, contrast-enhanced CT images. Multiple aggregation strategies were investigated for cases with multiple metastases. Radiomics signatures were derived using feature selection methods. Random survival forests (RSF) and neural network survival models (DeepSurv) based on radiomics features, alone or combined with CRS, were developed to predict TTR and DSS. Leveraging survival models predictions, classification models were trained to predict TTR within 18 months and DSS within 3 years. Classification performance was assessed with area under the receiver operating characteristic curve (AUC) on the test set.

For TTR prediction, the concordance index (95% confidence interval) was 0.57 (0.57-0.57) for CRS, 0.61 (0.60-0.61) for RSF in combination with CRS, and 0.70 (0.68-0.73) for DeepSurv in combination with CRS. For DSS prediction, the concordance index was 0.59 (0.59-0.59) for CRS, 0.57 (0.56-0.57) for RSF in combination with CRS, and 0.60 (0.58-0.61) for DeepSurv in combination with CRS. For TTR classification, the AUC was 0.33 (0.33-0.33) for CRS, 0.77 (0.75-0.78) for radiomics signature alone, and 0.58 (0.57-0.59) for DeepSurv score alone. For DSS classification, the AUC was 0.61 (0.61-0.61) for CRS, 0.57 (0.56-0.57) for radiomics signature, and 0.75 (0.74-0.76) for DeepSurv score alone.

Radiomics-based survival models outperformed CRS for TTR prediction. More accurate, noninvasive, and early prediction of patient outcome may help reduce exposure to ineffective yet toxic chemotherapy or high-risk major hepatectomies.

Cardiovascular disease affects the carotid arteries, coronary arteries, aorta and the peripheral arteries. Radiomics involves the extraction of quantitative data from imaging features that are imperceptible to the eye. Radiomics analysis in cardiovascular disease has largely focused on CT and MRI modalities. This scoping review aims to summarise the existing literature on radiomic analysis techniques in cardiovascular disease.

MEDLINE and Embase databases were searched for eligible studies evaluating radiomic techniques in living human subjects derived from CT, MRI or PET imaging investigating atherosclerotic disease. Data on study population, imaging characteristics and radiomics methodology were extracted.

Twenty-nine studies consisting of 5753 patients (3752 males) were identified, and 78.7% of patients were from coronary artery studies. Twenty-seven studies employed CT imaging (19 CT carotid angiography and 6 CT coronary angiography (CTCA)), and two studies studied PET/CT. Manual segmentation was most frequently undertaken. Processing techniques included voxel discretisation, voxel resampling and filtration. Various shape, first-order, second-order and higher-order radiomic features were extracted. Logistic regression was most commonly used for machine learning.

Most published evidence was feasibility/proof of concept work. There was significant heterogeneity in image acquisition, segmentation techniques, processing and analysis between studies. There is a need for the implementation of standardised imaging acquisition protocols, adherence to published reporting guidelines and economic evaluation.

Texture analysis derived from computed tomography (CT) may provide clinically relevant imaging biomarkers associated with tumor histopathology. Perihilar cholangiocarcinoma is a malignant disease with an overall poor prognosis.

The present study sought to elucidate possible associations between texture features derived from CT images with grading, tumor markers, and survival in extrahepatic, perihilar cholangiocarcinomas tumors.

This retrospective study included 22 patients (10 females, 45%) with a mean age of 71.8 ± 8.7 years. Texture analysis was performed using the free available Mazda software. All tumors were histopathologically confirmed. Survival and clinical parameters were used as primary study outcomes.

In discrimination analysis, "S(1,1)SumVarnc" was statistically significantly different between patients with long-term survival and nonlong-term survival (mean 275.8 ± 32.6 vs. 239.7 ± 26.0, p = 0.01). The first-order parameter "skewness" was associated with the tumor marker "carcinoembryonic antigen" (CEA) (r = -0.7, p = 0.01). A statistically significant correlation of the texture parameter "S(5,0)SumVarnc" with tumor grading was identified (r = -0.6, p < 0.01). Several other texture features correlated with tumor markers CA-19-9 and AFP, as well as with T and N stage of tumors.

Several texture features derived from CT images were associated with tumor characteristics and survival in patients with perihilar cholangiocarcinomas. CT texture features could be used as valuable novel imaging markers in clinical routine.

Definitive chemoradiotherapy (dCRT) is the cornerstone for locally advanced non-small cell lung cancer (LA-NSCLC). The study aimed to construct a multi-omics model integrating baseline clinical data, computed tomography (CT) images and genetic information to predict the prognosis of dCRT in LA-NSCLC patients.

The study retrospectively enrolled 105 stage III LA-NSCLC patients who had undergone dCRT. The pre-treatment CT images were collected, and the primary tumor was delineated as a region of interest (ROI) on the image using 3D-Slicer, and the radiomics features were extracted. The least absolute shrinkage and selection operator (LASSO) was employed for dimensionality reduction and selection of features. Genomic information was obtained from the baseline tumor tissue samples. We then constructed a multi-omics model by combining baseline clinical data, radiomics and genomics features. The predictive performance of the model was evaluated by the area under the curve (AUC) of the receiver operating characteristic (ROC) and the concordance index (C-index).

The median follow-up time was 30.1 months, and the median progression-free survival (PFS) was 10.60 months. Four features were applied to construct the radiomics model. Multivariable analysis demonstrated the Rad-score, KEAP1 and MET mutations were independent prognostic factors for PFS. The C-index of radiomics model, genomics model and radiogenomics model all performed well in the training group (0.590 vs. 0.606 vs. 0.663) and the validation group (0.599 vs. 0.594 vs. 0.650).

The radiomics model, genomics model and radiogenomics model can all predict the prognosis of dCRT for LA-NSCLC, and the radiogenomics model is superior to the single type model.

Radiomics is not yet used in clinical practice due to concerns regarding its susceptibility to technical factors. We aimed to assess the stability and interscan and interreader reproducibility of myocardial radiomic features between energy-integrating detector computed tomography (EID-CT) and photon-counting detector CT (PCD-CT) in patients undergoing coronary CT angiography (CCTA) on both systems.

Consecutive patients undergoing clinically indicated CCTA on an EID-CT were prospectively enrolled for a PCD-CT CCTA within 30 days. Virtual monoenergetic images (VMI) at various keV levels and polychromatic images (T3D) were generated for PCD-CT, with image reconstruction parameters standardized between scans. Two readers performed myocardial segmentation and 110 radiomic features were compared intraindividually between EID-CT and PDC-CT series. The agreement of parameters was assessed using the intraclass correlation coefficient and paired t-test for the stability of the parameters.

Eighteen patients (15 males) aged 67.6 ± 9.7 years (mean ± standard deviation) were included. Besides polychromatic PCD-CT reconstructions, 60- and 70-keV VMIs showed the highest feature stability compared to EID-CT (96%, 90%, and 92%, respectively). The interscan reproducibility of features was moderate even in the most favorable comparisons (median ICC 0.50 [interquartile range 0.20-0.60] for T3D; 0.56 [0.33-0.74] for 60 keV; 0.50 [0.36-0.62] for 70 keV). Interreader reproducibility was excellent for the PCD-CT series and good for EID-CT segmentations.

Most myocardial radiomic features remain stable between EID-CT and PCD-CT. While features demonstrated moderate reproducibility between scanners, technological advances associated with PCD-CT may lead to greater reproducibility, potentially expediting future standardization efforts.

While the use of PCD-CT may facilitate reduced interreader variability in radiomics analysis, the observed interscanner variations in comparison to EID-CT should be taken into account in future research, with efforts being made to minimize their impact in future radiomics studies.

Most myocardial radiomic features resulted in being stable between EID-CT and PCD-CT on certain VMIs. The reproducibility of parameters between detector technologies was limited. PCD-CT improved interreader reproducibility of myocardial radiomic features.

Radiomics features (RFs) serve as quantitative metrics to characterize shape, density/intensity, and texture patterns in radiological images. Despite their promise, RFs exhibit reproducibility challenges across acquisition settings, thus limiting implementation into clinical practice. In this investigation, we evaluate the effects of different CT scanners and CT acquisition protocols (KV, mA, field-of-view, and reconstruction kernel settings) on RFs extracted from lumbar vertebrae of a cadaveric trunk. Employing univariate and multivariate Generalized Linear Models (GLM), we evaluated the impact of each acquisition parameter on RFs. Our findings indicate that variations in mA had negligible effects on RFs, while alterations in kV resulted in exponential changes in several RFs, notably First Order (94.4%), GLCM (87.5%), and NGTDM (100%). Moreover, we demonstrated that a tailored GLM model was superior to the ComBat algorithm in harmonizing CT images. GLM achieved R2 > 0.90 in 21 RFs (19.6%), contrasting ComBat's mean R2 above 0.90 in only 1 RF (0.9%). This pioneering study unveils the effects of CT acquisition parameters on bone RFs in cadaveric specimens, highlighting significant variations across parameters and scanner datasets. The proposed GLM model presents a robust solution for mitigating these differences, potentially advancing harmonization efforts in Radiomics-based studies across diverse CT protocols and vendors.

Neoadjuvant chemoradiotherapy is the standard treatment for patients with locally advanced rectal cancers owing to its ability to downstage primary tumours. Some patients can achieve pathological complete response after neoadjuvant therapy, and can adopt a "watch and wait" treatment strategy to avoid overtreatment. Therefore, it is essential to develop strategies for predicting responses to neoadjuvant therapy. Radiomics has shown great potential in extracting tumour features from high-throughput medical images for the construction of mathematics models for predicting the effects of anticancerous therapies. Herein, we explored MRI-based radiomics and found that it can predict responses of locally advanced rectal cancers to chemoradiation. Efficient radiomics model allow early-stage prediction of the effect of neoadjuvant chemoradiotherapy on locally advanced rectal cancers. It helps clinicians to make informed therapeutic decisions. In this review, we discuss the workflow of radiomics, and summarize the clinical application of MRI-based radiomics in predicting pathological complete response to neoadjuvant chemoradiotherapy of locally advanced rectal cancer.

This single-centre retrospective study aims to determine the incidence of therapy-induced surgical benefit in patients with non-metastatic gastrointestinal stromal tumour (GIST) treated with neoadjuvant tyrosine kinase inhibitors (TKI) and evaluate whether this can be predicted by radiological response criteria.

Thirty-nine non-metastatic GIST patients were treated with neoadjuvant TKI treatment, followed by curative-intended surgery, and monitored using contrast-enhanced computed tomography (CE-CT). Surgical benefit was independently assessed by two surgical oncologists and was defined by de-escalation of surgical strategy or reduced surgical complexity. Radiological response between baseline and the last preoperative scan was determined through RECIST 1.1, Choi and volumetric criteria.

In this patient cohort, median neoadjuvant treatment interval was 8.3 (IQR, 3.9-10.6) months. Surgical benefit was gained in 22/39 patients. When comparing radiological criteria to findings on surgical benefit, accuracy, sensitivity, and specificity for RECIST 1.1 (90 %, 100.0 % and 82 %), Choi (64 %, 24 %, and 96 %) and volumetry (95 %, 100.0 %, and 91 %) were calculated. In 30/39 patients, temporal changes in tumour size over the course of treatment was assessed. Tumour volume reduced significantly in the surgical-benefit group compared to the non-benefit group (72 % vs. 25 %, p < 0.01) within three months. 14/19 surgical-benefit patients had an initial volume reduction above 66 %, after which volume reduced slightly with a median 3.1 % (IQR, 2.1-7.8 %) reduction.

Surgical benefit after neoadjuvant treatment was achieved in 56 % of patients and was most accurately reflected by size-based response criteria. In patients with therapy-induced surgical benefit, nearly all treatment-induced volume reductions were achieved within three months.

Artificial intelligence (AI) stands at the threshold of revolutionizing clinical oncology, with considerable potential to improve early cancer detection and risk assessment, and to enable more accurate personalized treatment recommendations. However, a notable imbalance exists in the distribution of the benefits of AI, which disproportionately favour those living in specific geographical locations and in specific populations. In this Perspective, we discuss the need to foster the development of equitable AI tools that are both accurate in and accessible to a diverse range of patient populations, including those in low-income to middle-income countries. We also discuss some of the challenges and potential solutions in attaining equitable AI, including addressing the historically limited representation of diverse populations in existing clinical datasets and the use of inadequate clinical validation methods. Additionally, we focus on extant sources of inequity including the type of model approach (such as deep learning, and feature engineering-based methods), the implications of dataset curation strategies, the need for rigorous validation across a variety of populations and settings, and the risk of introducing contextual bias that comes with developing tools predominantly in high-income countries.

Extracorporeal shock wave lithotripsy (ESWL) is widely considered the primary approach for managing urinary tract stones. This study aimed to assess the predictive factors associated with non-contrast computed tomography (NCCT)-based parameters of upper urinary stones in relation to the outcomes of ESWL.

A systematic search was conducted in PubMed, ScienceDirect, Web of Science, and Cochrane Library to identify all relevant studies published up to June 3, 2023. Several NCCT-based parameters to predict ESWL outcomes, comprised of mean stone density (MSD), skin-to-stone distance (SSD), and stone size, were extracted and analyzed using Review Manager software.

Out of 979 publications screened, a total of 39 publications, involving 7869 patients, were enrolled in the analysis. The pooled estimate demonstrated significant differences between MSD, and stone size between successful and failure of stone fragmentation groups, in which lower values of these parameters are associated with successful ESWL outcomes.

The results from the current study suggested that lower NCCT parameters, notably MSD, SSD, and stone size, are significantly associated with successful ESWL outcome. However, additional large-scale prospective studies are required to utilize these parameters effectively, and the optimal cutoff value should be determined.

Artificial intelligence (AI)-based auto-segmentation models hold promise for enhanced efficiency and consistency in organ contouring for adaptive radiation therapy and radiation therapy planning. However, their performance on pediatric computed tomography (CT) data and cross-scanner compatibility remain unclear. This study aimed to evaluate the performance of AI-based auto-segmentation models trained on adult CT data when applied to pediatric data sets and explore the improvement in performance gained by including pediatric training data. It also examined their ability to accurately segment CT data acquired from different scanners.

Using the nnU-Net framework, segmentation models were trained on data sets of adult, pediatric, and combined CT scans for 7 pelvic/thoracic organs. Each model was trained on 290 to 300 cases per category and organ. Training data sets included a combination of clinical data and several open repositories. The study incorporated a database of 459 pediatric (0-16 years) CT scans and 950 adults (>18 years), ensuring all scans had human expert ground-truth contours of the selected organs. Performance was evaluated based on Dice similarity coefficients (DSC) of the model-generated contours.

AI models trained exclusively on adult data underperformed on pediatric data, especially for the 0 to 2 age group: mean DSC was below 0.5 for the bladder and spleen. The addition of pediatric training data demonstrated significant improvement for all age groups, achieving a mean DSC of above 0.85 for all organs in every age group. Larger organs like the liver and kidneys maintained consistent performance for all models across age groups. No significant difference emerged in the cross-scanner performance evaluation, suggesting robust cross-scanner generalization.

For optimal segmentation across age groups, it is important to include pediatric data in the training of segmentation models. The successful cross-scanner generalization also supports the real-world clinical applicability of these AI models. This study emphasizes the significance of data set diversity in training robust AI systems for medical image interpretation tasks.

Radiomics analysis extracts quantitative data (features) from medical images. These features could potentially reflect biological characteristics and act as imaging biomarkers within precision medicine. However, there is a lack of cross-comparison and validation of radiomics outputs which is paramount for clinical implementation. In this study, we compared radiomics outputs across two computed tomography (CT)-based preclinical scanners.

Cone beam CT (CBCT) and µCT scans were acquired using different preclinical CT imaging platforms. The reproducibility of radiomics features on each scanner was assessed using a phantom across imaging energies (40 & 60 kVp) and segmentation volumes (44-238 mm3). Retrospective mouse scans were used to compare feature reliability across varying tissue densities (lung, heart, bone), scanners and after voxel size harmonisation. Reliable features had an intraclass correlation coefficient (ICC) > 0.8.

First order and GLCM features were the most reliable on both scanners across different volumes. There was an inverse relationship between tissue density and feature reliability, with the highest number of features in lung (CBCT=580, µCT=734) and lowest in bone (CBCT=110, µCT=560). Comparable features for lung and heart tissues increased when voxel sizes were harmonised. We have identified tissue-specific preclinical radiomics signatures in mice for the lung (133), heart (35), and bone (15).

Preclinical CBCT and µCT scans can be used for radiomics analysis to support the development of meaningful radiomics signatures. This study demonstrates the importance of standardisation and emphasises the need for multi-centre studies.

Synchronous liver metastasis (SLM) is a significant contributor to morbidity in colorectal cancer (CRC). There are no effective predictive device integration algorithms to predict adverse SLM events during the diagnosis of CRC.

To explore the risk factors for SLM in CRC and construct a visual prediction model based on gray-level co-occurrence matrix (GLCM) features collected from magnetic resonance imaging (MRI).

Our study retrospectively enrolled 392 patients with CRC from Yichang Central People's Hospital from January 2015 to May 2023. Patients were randomly divided into a training and validation group (3:7). The clinical parameters and GLCM features extracted from MRI were included as candidate variables. The prediction model was constructed using a generalized linear regression model, random forest model (RFM), and artificial neural network model. Receiver operating characteristic curves and decision curves were used to evaluate the prediction model.

Among the 392 patients, 48 had SLM (12.24%). We obtained fourteen GLCM imaging data for variable screening of SLM prediction models. Inverse difference, mean sum, sum entropy, sum variance, sum of squares, energy, and difference variance were listed as candidate variables, and the prediction efficiency (area under the curve) of the subsequent RFM in the training set and internal validation set was 0.917 [95% confidence interval (95%CI): 0.866-0.968] and 0.09 (95%CI: 0.858-0.960), respectively.

A predictive model combining GLCM image features with machine learning can predict SLM in CRC. This model can assist clinicians in making timely and personalized clinical decisions.