Whole-orbit radiomics: machine learning-based multi- and fused- region radiomics signatures for intravenous glucocorticoid response prediction in thyroid eye disease.

BACKGROUND: Radiomics analysis of orbital magnetic resonance imaging (MRI) shows preliminary potential for intravenous glucocorticoid (IVGC) response prediction of thyroid eye disease (TED). The current region of interest segmentation contains only a single organ as extraocular muscles (EOMs). It would be of great value to consider all orbital soft tissues and construct a better prediction model. METHODS: In this retrospective study, we enrolled 127 patients with TED that received 4·5 g IVGC therapy and had complete follow-up examinations. Pre-treatment orbital T2-weighted imaging (T2WI) was acquired for all subjects. Using multi-organ segmentation (MOS) strategy, we contoured the EOMs, lacrimal gland (LG), orbital fat (OF), and optic nerve (ON), respectively. By fused-organ segmentation (FOS), we contoured the aforementioned structures as a cohesive unit. Whole-orbit radiomics (WOR) models consisting of a multi-regional radiomics (MRR) model and a fused-regional radiomics (FRR) model were further constructed using six machine learning (ML) algorithms. RESULTS: The support vector machine (SVM) classifier had the best performance on the MRR model (AUC = 0·961). The MRR model outperformed the single-regional radiomics (SRR) models (highest AUC = 0·766, XGBoost on EOMs, or LR on OF) and conventional semiquantitative imaging model (highest AUC = 0·760, NaiveBayes). The application of different ML algorithms for the comparison between the MRR model and the FRR model (highest AUC = 0·916, LR) led to different conclusions. CONCLUSIONS: The WOR models achieved a satisfactory result in IVGC response prediction of TED. It would be beneficial to include more orbital structures and implement ML algorithms while constructing radiomics models. The selection of separate or overall segmentation of orbital soft tissues has not yet attained its final optimal result.

White Matter Alterations of Visual Pathway in Thyroid Eye Disease: A Fixel-Based Analysis.

BACKGROUND: Thyroid eye disease (TED), particularly its sight-threatening complication, dysthyroid optic neuropathy (DON), profoundly impacts patients' visual health. The pathological changes in the white matter (WM) fibers within the intracranial visual pathway in TED have been infrequently studied. Understanding these changes holds crucial importance for exploring the pathogenesis and prognosis of TED. PURPOSE: To utilize fixel-based analysis (FBA) to clarify the type of microstructural damage occurring in the visual pathway in TED. STUDY TYPE: Prospective. SUBJECTS: 28 TED with DON patients (11 males and 17 females), 28 TED without DON (non-DON) patients (12 males and 16 females), and 28 healthy controls (HCs) (12 males and 16 females). FIELD STRENGTH/SEQUENCE: 3 T; multishell diffusion MRI using echo planar imaging. ASSESSMENT: Fiber density (FD) and fiber-bundle cross-section (FC) were calculated to characterize WM microstructural alteration in TED visual pathway. The correlations between FBA metrics and visual field index and mean deviation were examined. STATISTICAL TESTS: One-way analysis of variance, Kruskal-Wallis, t-tests, Mann-Whitney U, Chi-square, and Pearson correlation, were conducted with false discovery rate and family wise error corrections. Significance was set at P < 0.05. RESULTS: Both DON and non-DON groups showed significant FD loss in the right optic tract compared with HCs, with DON patients experiencing more severe FD loss. Only DON patients had FD loss in the right optic radiation (OR) compared with the non-DON patients and HCs, with no FC difference across groups. FD in DON patients' ORs significantly correlated with visual field index (r = 0.857) and mean deviation (r = 0.751). DATA CONCLUSION: Both DON and non-DON affect the WM microstructure of the visual pathway to varying extents. Visual field metrics can reflect the severity of FD damage to the OR in the visual pathway of DON patients. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: Stage 3.

MRI-based radiomics nomogram for distinguishing solitary fibrous tumor from schwannoma in the orbit: a two-center study.

OBJECTIVES: To investigate the value of magnetic resonance imaging (MRI) radiomics for distinguishing solitary fibrous tumor (SFT) from schwannoma in the orbit. MATERIALS AND METHODS: A total of 140 patients from two institutions were retrospectively included. All patients from institution 1 were randomized into a training cohort (n = 69) and a validation cohort (n = 35), and patients from institution 2 were used as an external testing cohort (n = 36). One hundred and six features were extracted from T2-weighted imaging (T2WI) and contrast-enhanced T1-weighted imaging (CET1WI). A radiomics model was built for each sequence using least absolute shrinkage and selection operator logistic regression, and radiomics scores were calculated. A combined model was constructed and displayed as a radiomics nomogram. Two radiologists jointly assessed tumor category based on MRI findings. The performances of the radiomics models and visual assessment were compared via area under the curve (AUC). RESULTS: The performances of the radiomics nomogram combining T2WI and CET1WI radiomics scores were superior to those of the pooled readers in the training (AUC 0.986 vs. 0.807, p < 0.001), validation (AUC 0.989 vs. 0.788, p = 0.009), and the testing (AUC 0.903 vs. 0.792, p = 0.093), although significant difference was not found in the testing cohort. Decision curve analysis demonstrated that the radiomics nomogram had better clinical utility than visual assessment. CONCLUSION: MRI radiomics nomogram can be used for distinguishing between orbital SFT and schwannoma, which may help tumor management by clinicians. CLINICAL RELEVANCE STATEMENT: It is of great importance and challenging for distinguishing solitary fibrous tumor from schwannoma in the orbit. In the present study, an MRI-based radiomics nomogram were developed and independently validated, which could help the discrimination of the two entities. KEY POINTS: • It is challenging to differentiate solitary fibrous tumor from schwannoma in the orbit due to similar clinical and image features. • A radiomics nomogram based on T2-weighted imaging and contrast-enhanced T1-weighted imaging has advantages over radiologists. • Radiomics can provide a non-invasive diagnostic tool for differentiating between the two entities.

A c-MET-Targeted Topical Fluorescent Probe cMBP-ICG Improves Oral Squamous Cell Carcinoma Detection in Humans.

INTRODUCTION: The postoperative survival of oral squamous cell carcinoma (SCC) relies on precise detection and complete resection of original tumors. The mucosal extension of the tumor is evaluated visually during surgery, but small and flat foci are difficult to detect. Real-time fluorescence imaging may improve detection of tumor margins. MATERIALS AND METHODS: In the current study, a peptide-based near-infrared (NIR) fluorescence dye, c-MET-binding peptide-indocyanine green (cMBP-ICG), which specifically targets tumor via c-MET binding, was synthetized. A prospective pilot clinical trial then was conducted with oral SCC patients and intraoperatively to assess the feasibility of cMBP-ICG used to detect tumors margins. Fluorescence was histologically correlated to determine sensitivity and specificity. RESULTS: The immunohistochemistry (IHC) results demonstrated increased c-Met expression in oral SCC compared with normal mucosa. Tumor-to-background ratios ranged from 2.71 ± 0.7 to 3.11 ± 1.2 in different concentration groups. From 10 patients with oral SCC, 60 specimens were collected from tumor margins. The sensitivity and specificity of discriminative value derived from cMBP-ICG application in humans were respectively 100% and 75%. CONCLUSIONS: Topical application of cMBP-ICG is feasible and safe for optimizing intraoperative visualization and tumor margin detection in oral SCC patients, which could clinically increase the probability of complete resections and improve oncologic outcomes.

Differentiation of geniculate ganglion venous malformation from schwannoma: dynamic T1-weighted imaging provides unique diagnostic value.

OBJECTIVE: To distinguish geniculate ganglion venous malformation (GGVM) from schwannoma (GGS) by using high-resolution CT (HRCT), routine MRI, and dynamic T1-weighted imaging (T1WI) characteristics. METHODS: Surgically confirmed GGVMs and GGSs between 2016 and 2021 were retrospectively included. Preoperative HRCT, routine MR, and dynamic T1WI were performed on all patients. Clinical data, imaging characteristics including lesion size, involvement of facial nerve (FN), signal intensity, enhancement pattern on dynamic T1WI, and bone destruction on HRCT were evaluated. Logistic regression model was developed to identify independent factors for GGVMs, and the diagnostic performance was accessed by receiving operative curve (ROC) analysis. Histological characteristics were explored for both GGVMs and GGSs. RESULTS: Twenty GGVMs and 23 GGSs with mean age of 31 were included. On dynamic T1WI, 18 GGVMs (18/20) showed "pattern A" enhancement (a progressive filling enhancement), while all 23 GGSs showed "pattern B" enhancement (a gradual whole-lesion enhancement) (p < 0.001). Thirteen GGVMs (13/20) showed the "honeycomb" sign whereas all GGS (23/23) showed extensive bone changes on HRCT (p < 0.001). Lesion size, involvement of FN segment, signal intensity on non-contrast T1WI and T2-weighted imaging (T2WI), and homogeneity on enhanced T1WI were obviously differed between two lesions (p < 0.001, p = 0.002, p < 0.001, p = 0.01, p = 0.02, respectively). Regression model showed the "honeycomb" sign and "pattern A" enhancement were independent risk factors. Histologically, GGVM was characterized by interwoven dilated and tortuous veins, while GGS was characterized by abundant spindle cells with dense arterioles or capillaries. CONCLUSIONS: The "honeycomb" sign on HRCT and "pattern A" enhancement on dynamic T1WI are the most promising imaging characteristics for differentiating GGVM from GGS. CLINICAL RELEVANCE STATEMENT: The characteristic sign and enhancement pattern on HRCT and dynamic T1-weighted imaging allow preoperative differentiation of geniculate ganglion venous malformation and schwannoma feasible, which will improve clinical management and benefit patient prognosis. KEY POINTS: • The "honeycomb" sign on HRCT is a reliable finding to differentiate GGVM from GGS. • GGVM typically shows "pattern A" enhancement (focal enhancement of the tumor on early dynamic T1WI, followed by progressive contrast filling of the tumor in the delayed phase), while "pattern B" enhancement (gradual heterogeneous or homogeneous enhancement of the whole lesion) is observed in GGS on dynamic T1WI.

T Cell Subsets are Associated with Clinical Activity and Efficacy of 4.5g Intravenous Glucocorticoid for Moderate-To-Severe Thyroid-Associated Ophthalmopathy.

BACKGROUND: Intravenous glucocorticoid (IVGC) remains the main treatment for moderate-to-severe and active thyroid-associated ophthalmopathy (TAO). However, a substantial number (20-30%) of active moderate-to-severe TAO patients may not respond to IVGC. Some patients may have disease progression despite IVGC treatment or relapse after steroid withdrawal. OBJECTIVES: To analyze risk factors for clinical activity and predictive factors for clinical outcomes of 4.5 g IVGC therapy in patients with moderate-to-severe TAO. DESIGN AND METHODS: Our study was performed in two steps: step 1 involved 110 moderate-to-severe TAO patients and analyzed risk factors for TAO activity; step 2 involved 53 active moderate-to-severe TAO patients from step 1 who were treated with 4.5 g IVGC therapy and analyzed predictive factors for clinical outcomes of IVGC therapy. Multivariate logistic regression analysis was used to identify the independent predictors and establish the predictive model. RESULTS: Abnormal TRAb (OR = 4.717; P = 0.019) and the percentage of CD3+CD4+ T cell (OR = 1.092; P = 0.028) were independently associated with the activity of moderate-to-severe TAO patients. The pretreatment CAS-max in both eyes (OR = 7.221; P = 0.013) and the percentage of pretreatment CD3+T cell (OR = 0.718; P = 0.037) were independently associated with therapeutic efficacy. The pretreatment CAS-max in both eyes (OR = 156.53; P = 0.028) and the percentage of post-treatment CD3+T cell (OR = 0.554; P = 0.043) were independently associated with therapeutic efficacy. Besides, multivariable prediction models were established, which were better in the forecasting aspect than single-variable prediction models. CONCLUSIONS: Based on the findings of this study, we should monitor the peripheral blood T cell subsets for TAO, which could be helpful to timely judge the condition of clinical manifestation and effect of treatment for TAO patients. Multivariable prediction models have been established, which have great significance for clinical work.

Multimodal ultrasound for preoperative evaluation of dermatofibrosarcoma protuberans: a series of 40 cases.

BACKGROUND: Dermatofibrosarcoma protuberans (DFSP) is a rare, low to intermediate-grade sarcoma, which needs imaging examination. Small series of ultrasound findings in DFSP have been published; however, the usefulness of elastography and contrast-enhanced ultrasound (CEUS) in DFSP has not been studied. We aim to study multimodal ultrasound findings and report the correlation between imagings and tiny extension in DFSP for preoperative evaluation. METHODS: Two-D ultrasound, 3-D color ultrasound, elastography, and CEUS findings were retrospectively evaluated. Forty histopathologically confirmed DFSPs were studied. RESULTS: On 2-D ultrasound, 26(65%) appeared as mostly hypoechoic lesions with occasional hyperechoic dots within the tumor matrix and lobulated lateral borders. Eight (20%) lesions were multilayered. Ninety-five percent of lesions showed increased vascularity. On 3-D ultrasound, DFSPs showed branch-shaped, striped, and wrapped color patterns. Power Doppler showed mainly artery of a moderate arterial peak systolic blood flow and low resistance index. DFSP is hard on elastography. On CEUS, DFSPs showed a long peak time, low peak and a small amount of perfusion around the tumor, 73.7% (14/19) of lesions showed a heterogeneous contrast enhancement and 89.5% (17/19) of lesions showed hyper-enhancement. CEUS showed better concordance than US with histology on the maximum diameter and depth (P < 0.05). CONCLUSIONS: Multimodal ultrasound showed significant characteristics in DFSP, which would improve diagnostic accuracy. CEUS could be an effective tool to determine tiny tumor extension.

Histogram analysis of diffusion-weighted imaging and dynamic contrast-enhanced MRI for predicting occult lymph node metastasis in early-stage oral tongue squamous cell carcinoma.

OBJECTIVES: To investigate the feasibility of whole-tumor histogram analysis of diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) MRI for predicting occult lymph node metastasis (LNM) in early-stage oral tongue squamous cell cancer (OTSCC). MATERIALS AND METHODS: This retrospective study included 55 early-stage OTSCC (cT1-2N0M0) patients; 34 with pathological LNM and 21 without. Eight whole-tumor histogram features were extracted from quantitative apparent diffusion coefficient (ADC) maps and two semi-quantitative DCE parametric maps (wash-in and wash-out). The clinicopathological factors and histogram features were compared between the two groups. Stepwise logistic regression was used to identify independent predictors. Receiver operating characteristic curves were generated to assess the performances of significant variables and a combined model for predicting occult LNM. RESULTS: MRI-determined depth of invasion and ADCentropy was significantly higher in the LNM group, with respective areas under the curve (AUCs) of 0.67 and 0.69, and accuracies of 0.73 and 0.73. ADC10th. ADCuniformity and wash-inskewness were significantly lower in the LNM group, with respective AUCs of 0.68, 0.71, and 0.69, and accuracies of 0.65, 0.71, and 0.64. Histogram features from wash-out maps were not significantly associated with cervical node status. In the logistic regression analysis, ADC10th, ADCuniformity, and wash-inskewness were independent predictors. The combined model yielded the best predictive performance, with an AUC of 0.87 and an accuracy of 0.82. CONCLUSIONS: Whole-tumor histogram analysis of ADC and wash-in maps is a feasible tool for preoperative evaluation of cervical node status in early-stage OTSCC. KEY POINTS: • Histogram analysis of parametric maps from DWI and DCE-MRI may assist the prediction of occult LNM in early-stage OTSCC. • ADC10th, ADCuniformity, and wash-inskewness were independent predictors. • The combined model exhibited good predictive performance, with an accuracy of 0.82.

Assessment of tumor depth in oral tongue squamous cell carcinoma with multiparametric MRI: correlation with pathology.

OBJECTIVES: To compare the correlation of depth of invasion (DOI) measured on multiple magnetic resonance imaging (MRI) sequences and pathological DOI, in order to determine the optimal MRI sequence for measurement. METHODS: A total of 122 oral tongue squamous cell carcinoma (OTSCC) patients were retrospectively analyzed, who had received preoperative MRI and surgical resection. DOIs measured on fat-suppressed T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), dynamic enhanced-T1 high-resolution insotropic volume examination (e-THRIVE), and contrast-enhanced fat-suppressed T1WI (CE-T1WI) were respectively compared to those measured in pathologic specimens. The cutoff value of the best correlated MRI sequence was determined, and the T staging accuracy of MRI-derived DOI was evaluated. RESULTS: DOI derived from e-THRIVE showed the best correlation (r = 0.936, p < 0.001) with pathological DOI. The area under the curve values of MRI-derived DOI distinguishing T1 stage from T2 stage and distinguishing T2 stage from T3 stage were 0.969 and 0.974, respectively. The T staging criteria of MRI-derived DOI were 6.2 mm and 11.4 mm, with a staging accuracy of 86.9% compared to pathological DOI criteria of 5 mm and 10 mm. CONCLUSION: E-THRIVE was the optimal MR sequence to measure the MR-derived DOI, and DOI derived from e-THRIVE could serve as a potential cut-off value as a clinical T staging indicator of OTSCC. KEY POINTS: • Multiparametric MRI helps radiologists to assess the neoplasm invasion in patients with oral tongue squamous cell carcinoma. • Retrospective study indicated that measurement was most accurate on enhanced-T1 high-resolution insotropic volume examination dynamic contrast enhancement images. • T staging of oral tongue squamous cell carcinoma was accurate according to the dynamic contrast enhancement MRI-derived depth of invasion.

Clinical effectiveness of contrast medium injection protocols for 80-kV coronary and craniocervical CT angiography-a prospective multicenter observational study.

BACKGROUND AND OBJECTIVE: Decreasing X-ray tube voltage is an effective way to reduce radiation and contrast dose, especially in non-obese patients. The current study focuses on CTA in non-obese patients to evaluate image quality and feasibility of 80-kV acquisition protocols with varying iodine delivery rates (IDR) and contrast concentrations in routine clinical practice. METHODS: A prospective observational study in patients ≥ 18 years and ≤ 90 kg referred for coronary or craniocervical CTA at 10 centers in China (ClinicalTrials.gov: NCT02840903). Patients were divided into four groups: a standard 100-kV protocol (370 mgI/ml, IDR 1.48 gI/s), and three 80-kV protocols (370 mgI/ml, IDR 1.2 gI/s; 300 mgI/ml, IDR 1.2 gI/s; 300 mgI/ml, IDR 0.96gI/s). The primary outcome was contrast opacification of target vascular segments. Secondary outcomes were image quality (contrast-to-noise ratio, signal-to-noise ratio, visual image quality, and diagnostic confidence assessment), radiation, and iodine dose. RESULTS: From July 2016 to July 2017, 1213 patients were enrolled: 614 coronary and 599 craniocervical CTA. The mean contrast opacification was ≥ 300 HU for 80-kV 1.2 gI/s IDR scanned segments; IDR 0.96 gI/s led to lower opacification. Image quality and diagnostic confidence were fair to excellent (≥ 98% of images), despite lower contrast-to-noise ratios and signal-to-noise ratios in 80-kV images. Compared to the standard protocol, 80-kV protocols led to 44-52% radiation dose reductions (p < 0.001) and 19% iodine dose reductions (p < 0.001). CONCLUSION: Eighty-kilovolt 1.2 gI/s IDR protocols can be recommended for coronary and craniocervical CTA in non-obese patients, reducing radiation and iodine dose without compromising image quality. KEY POINTS: • Using low-voltage scanning CTA protocols, in which tube voltage and iodine delivery rate are reduced proportionally (voltage: 80 kV, IDR: 1.2 gI/s), reduces radiation and contrast dose without compromising image quality in routine clinical practice. • Reducing iodine delivery rate beyond direct proportionality to tube voltage is not beneficial.

Estrogen receptor variant ER-α36 facilitates estrogen signaling via EGFR in glioblastoma.

Glioblastoma (GBM) is a deadly and common primary brain tumor. Poor prognosis is linked to high proliferation and cell heterogeneity. Sex differences may play a role in patient outcome. Previous studies showed that ER-α36, a variant of the estrogen receptor (ER), mediated non-genomic estrogen signaling and is highly expressed in many ER-negative malignant tumors. ER-α36 also associates with epidermal growth factor receptor (EGFR). The primary purpose of this study is to investigate the cross talk between ER-α36 and EGFR in estrogen-mediated GBM cell proliferation. Here, we showed that ER-α36 was highly expressed and confirmed that ER-α36 co-labels with EGFR in human GBM samples using immunohistochemical techniques. We also investigated the mechanisms of estrogen-induced proliferation in ER-α-negative cell lines. We found that GBM cells showed varying responsive to mitogenic estrogen signaling which correlated with ER-α36 expression, and knockdown of ER-α36 diminished the response. Exposure to estrogen also caused upregulation of cyclin protein expression in vitro. We also found that low concentration of estrogen promoted SRC-Y-416 and inhibited SRC-Y-527 phosphorylation, corresponding with activated SRC signaling. Inhibiting SRC or EGFR abolished estrogen-induced mitogenic signaling, including cyclin expression and MAPK phosphorylation. Cumulatively, our results demonstrate that ER-α36 promotes non-genomic estrogen signaling via the EGFR/SRC/MAPK pathway in GBM. This may be important for the treatment of ER-α-negative GBMs that retain high level of ER-α36, since estrogen may be a viable therapeutic target for these patients.

Dual-molecular targeted NIR II probe with enhanced response for head and neck squamous cell carcinoma imaging.

In this research, a fluorescent probe of 7-(diethylamine) coumarin derivatives with multiple binding sites to detect biothiols in tumor cell with strong NIR II luminescencein vivowas synthesized. The biothiols include cysteine (Cys) and glutathione (GSH) in tumor cells, and the tumor-response luminescence was proved by the cell experiment. Importantly, the monolayer functional phospholipid (DSPE-PEG) coating and aggregation induced emission (AIE) dye of TPE modification made the probe have good stability and biocompatibility with little luminescence quenching in aqueous phase, which was proved byin vitroandin vivoexperiments. The final aqueous NIR II probe combined with bevacizumab (for VEGF recognition in the cancer cells) and Capmatinib (for Met protein recognition in the cancer cells) has stronger targeted imaging on head and neck squamous cell carcinoma (HNSCC) cancer with intravenous injection. This GSH/Cys detection in the tumor cell and strong dual-molecular NIR II bioimagingin vivomay provide new strategy to tumor detection.

Dual-energy CT-based radiomics nomogram in predicting histological differentiation of head and neck squamous carcinoma: a multicenter study.

PURPOSE: To develop and validate a dual-energy CT (DECT)-based radiomics nomogram from multicenter trials for predicting the histological differentiation of head and neck squamous cell carcinoma (HNSCC). METHODS: A total of 178 patients (112 in the training and 66 in the validation cohorts) from eight institutions with histologically proven HNSCCs were included in this retrospective study. Radiomics-signature models were constructed from features extracted from virtual monoenergetic images (VMI) and iodine-based material decomposition images (IMDI), reconstructed from venous-phase DECT images. Clinical factors were also assessed to build a clinical model. Multivariate logistic regression analysis was used to develop a nomogram combining the radiomics signature models and clinical model for predicting poorly differentiated HNSCC and moderately well-differentiated HNSCC. The predictive performance of the clinical model, radiomics signature models, and nomogram was compared. The calibration degree of the nomogram was also assessed. RESULTS: The tumor location, VMI-signature, and IMDI-signature were associated with the degree of HNSCC differentiation, and areas under the ROC curves (AUCs) were 0.729, 0.890, and 0.833 in the training cohort and 0.627, 0.859, and 0.843 in the validation cohort, respectively. The nomogram incorporating tumor location and two radiomics-signature models yielded the best performance in training (AUC = 0.987) and validation (AUC = 0.968) cohorts with a good calibration degree. CONCLUSION: The nomogram that integrated the DECT-based radiomics-signature models and tumor location showed good performance in predicting histological differentiation degree of HNSCC, providing a novel combination for predicting HNSCC differentiation.

Clinical evaluation of AI software for rib fracture detection and its impact on junior radiologist performance.

BACKGROUND: The detection of rib fractures (RFs) on computed tomography (CT) images is time-consuming and susceptible to missed diagnosis. An automated artificial intelligence (AI) detection system may be helpful to improve the diagnostic efficiency for junior radiologists. PURPOSE: To compare the diagnostic performance of junior radiologists with and without AI software for RF detection on chest CT images. MATERIALS AND METHODS: Six junior radiologists from three institutions interpreted 393 CT images of patients with acute chest trauma, with and without AI software. The CT images were randomly split into two sets at each institution, with each set assigned to a different radiologist First, the detection of all fractures (AFs), including displaced fractures (DFs), non-displaced fractures and buckle fractures, was analyzed. Next, the DFs were selected for analysis. The sensitivity and specificity of the radiologist-only and radiologist-AI groups at the patient level were set as primary endpoints, and secondary endpoints were at the rib and lesion level. RESULTS: Regarding AFs, the sensitivity difference between the radiologist-AI group and the radiologist-only group were significant at different levels (patient-level: 26.20%; rib-level: 22.18%; lesion-level: 23.74%; P < 0.001). Regarding DFs, the sensitivity difference was 16.67%, 14.19%, and 16.16% at the patient, rib, and lesion levels, respectively (P < 0.001). No significant difference was found in the specificity between the two groups for AFs and DFs at the patient and rib levels (P > 0.05). CONCLUSION: AI software improved the sensitivity of RF detection on CT images for junior radiologists and reduced the reading time by approximately 1 min per patient without decreasing the specificity.

Value of Dynamic Contrast-Enhanced (DCE) MRI in Predicting Response to Foam Sclerotherapy of Venous Malformations.

BACKGROUND: Preoperative imaging assessment of venous malformations (VMs) and prediction of foam sclerotherapy efficacy might be achievable by DCE-MRI but elaborate quantitive analysis was absent. PURPOSE: To evaluate the value of DCE-MRI in predicting the effectiveness of foam sclerotherapy in VMs. STUDY TYPE: Retrospective. POPULATION: Fifty-five patients (M:F = 17:38; mean age ± SD, 15.4 ± 13.0 years) with VMs. FIELD STRENGTH/SEQUENCE: Three Tesla MRI with 3D T1 -weighted volume interpolated body examination. ASSESSMENT: Patients who underwent pretreatment DCE-MRI were divided into "effective" and "ineffective" groups according to the response to foam sclerotherapy. Clinical characteristics and morphologic features were assessed. The semiquantitative parameters, such as maximum intensity time ratio (MITR), enhancement ratio (ER), and Slope, were obtained from ROI and volume of interest (VOI). The quartile and mean values of these parameters were acquired from VOI, while mean values denoted as Mean# were acquired from ROI. Establishment of two predictive models was based on ROI and VOI respectively. Model 1 was based on morphologic parameters and ROI semiquantitative parameters, while model 2 was based on morphologic parameters and VOI semiquantitative parameters. STATISTICAL ANALYSIS: Mann-Whitney U-test, Cohen's kappa, multivariate logistic regression analysis (backward stepwise), and ROC analyses. RESULTS: The lesion classification, presence of phlebolith, semiquantitative parameters of VOI (quartile and mean of MITR), and semiquantitative parameters of ROI (Slopemean# , MITRmean# ) were significantly different between two groups. Lesion classification (P = 0.002) and MITRmean# (P = 0.027) were independent predictors for poor efficacy in model 1 as determined by multivariate binary logistic regression analysis. For model 2, lesion classification (P = 0.006) and MITR25 (P = 0.001) were independent predictors. The predictive model based on VOI (AUC = 0.961) performed better than that based on ROI (AUC = 0.909) in predicting therapeutic response. DATA CONCLUSION: DCE-MRI is promising in predicting the response to foam sclerotherapy for VMs. The whole lesion VOI-based model showed better performance and could instruct surgical approach in the future. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 4.

Machine learning-based MRI texture analysis to predict occult lymph node metastasis in early-stage oral tongue squamous cell carcinoma.

OBJECTIVES: To develop and compare several machine learning models to predict occult cervical lymph node (LN) metastasis in early-stage oral tongue squamous cell cancer (OTSCC) from preoperative MRI texture features. MATERIALS AND METHODS: We retrospectively enrolled 116 patients with early-stage OTSCC (cT1-2N0) who had been surgically treated by tumor excision and elective neck dissection (END). For each patient, we extracted 86 texture features from T2-weighted imaging (T2WI) and contrast-enhanced T1-weighted imaging (ceT1WI), respectively. Dimension reduction was performed in three consecutive steps: reproducibility analysis, collinearity analysis, and information gain algorithm. Models were created using six machine learning methods, including logistic regression (LR), random forest (RF), naïve Bayes (NB), support vector machine (SVM), AdaBoost, and neural network (NN). Their performance was assessed using tenfold cross-validation. RESULTS: Occult LN metastasis was pathologically detected in 42.2% (49/116) of the patients. No significant association was identified between node status and patients' gender, age, or clinical T stage. Dimension reduction steps selected 6 texture features. The NB model gave the best overall performance, which correctly classified the nodal status in 74.1% (86/116) of the carcinomas, with an AUC of 0.802. CONCLUSION: Machine learning-based MRI texture analysis offers a feasible tool for preoperative prediction of occult cervical node metastasis in early-stage OTSCC. KEY POINTS: • A machine learning-based MRI texture analysis approach was adopted to predict occult cervical node metastasis in early-stage OTSCC with no evidence of node involvement on conventional images. • Six texture features from T2WI and ceT1WI of preoperative MRI were selected to construct the predictive model. • After comparing six machine learning methods, naïve Bayes (NB) achieved the best performance by correctly identifying the node status in 74.1% of the patients, using tenfold cross-validation.

Determination of Cervical Lymph Nodes Metastasis and Extra Nodal Extension Status by Quantitative Assessment of Border Irregularity and Apparent Diffusion Coefficient in Patients With Tongue Squamous Cell Carcinoma.

OBJECTIVE: The objective of this study was to determine the diagnostic value of quantitative border irregularity assessment and apparent diffusion coefficient (ADC) in patients with squamous cell carcinoma of the tongue (SCCT). METHODS: Cervical lymph nodes (n = 192) from 63 patients with SCCT were examined preoperatively by magnetic resonance imaging, including routine head and neck sequences, dynamic contrast-enhanced magnetic resonance imaging, diffusion-weighted imaging, ADC, surface regularity (SR), and visually assessed variables, and evaluated pathologically after surgery. RESULTS: Necrosis, lymphatic hilum, unclear margin, higher SR, long to short axis ratio, and ADC were associated with metastasis in cervical lymph nodes (M-cLNs) and extranodal extension (ENE), and thickened nodal rim with ENE alone. Apparent diffusion coefficient, SR, unclear margin, and visible necrosis were strongly associated with M-cLN, whereas SR, unclear margin, and visible necrosis were associated with ENE status on logistic regression analysis. CONCLUSIONS: Quantitative SR and ADC data greatly improved diagnosis of M-cLNs and ENE, relative to visible variables alone in patients with SCCT.

Histogram analysis of dynamic contrast-enhanced magnetic resonance imaging in the differential diagnosis of parotid tumors.

OBJECTIVE: Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) histograms were used to investigate whether their parameters can distinguish between benign and malignant parotid gland tumors and further differentiate tumor subgroups. MATERIALS AND METHODS: A total of 117 patients (32 malignant and 85 benign) who had undergone DCE-MRI for pretreatment evaluation were retrospectively included. Histogram parameters including mean, median, entropy, skewness, kurtosis and 10th, 90th percentiles were calculated from time to peak (TTP) (s), wash in rate (WIR) (l/s), wash out rate (WOR) (l/s), and maximum relative enhancement (MRE) (%) mono-exponential models. The Mann-Whitney U test was used to compare the differences between the benign and malignant groups. The diagnostic value of each significant parameter was determined on Receiver operating characteristic (ROC) analysis. Multivariate stepwise logistic regression analysis was used to identify the independent predictors of the different tumor groups. RESULTS: For both the benign and malignant groups and the comparisons among the subgroups, the parameters of TTP and MRE showed better performance among the various parameters. WOR can be used as an indicator to distinguish Warthin's tumors from other tumors. Warthin's tumors showed significantly lower values on 10th MRE and significantly higher values on skewness TTP and 10th WOR, and the combination of 10th MRE, skewness TTP and 10th WOR showed optimal diagnostic performance (AUC, 0.971) and provided 93.12% sensitivity and 96.70% specificity. After Warthin's tumors were removed from among the benign tumors, malignant parotid tumors showed significantly lower values on the 10th TTP (AUC, 0.847; sensitivity 90.62%; specificity 69.09%; P < 0.05) and higher values on skewness MRE (AUC, 0.777; sensitivity 71.87%; specificity 76.36%; P < 0.05). CONCLUSION: DCE-MRI histogram parameters, especially TTP and MRE parameters, show promise as effective indicators for identifying and classifying parotid tumors. Entropy TTP and kurtosis MRE were found to be independent differentiating variables for malignant parotid gland tumors. The 10th WOR can be used as an indicator to distinguish Warthin's tumors from other tumors.

Radiomics of apparent diffusion coefficient maps to predict histologic grade in squamous cell carcinoma of the oral tongue and floor of mouth: a preliminary study.

BACKGROUND: Histologic grade assessment plays an important part in the clinical decision making and prognostic evaluation of squamous cell carcinoma (SCC) of the oral tongue and floor of mouth (FOM). PURPOSE: To assess the value of apparent diffusion coefficient (ADC)-based radiomics in discriminating between low- and high-grade SCC of the oral tongue and FOM. MATERIAL AND METHODS: We included data from 88 patients (training cohort: n = 59; testing cohort: n = 29) who underwent diffusion-weighted imaging with a 3.0-T magnetic resonance imaging scanner before treatment. A total of 526 radiomics features were extracted from ADC maps to construct a radiomics signature with least absolute shrinkage and selection operator logistic regression. Receiver operating characteristic curves and areas under the curve (AUCs) were used to evaluate the performance of radiomic signature. RESULTS: Five features were selected to construct the radiomics signature for predicting histologic grade. The ADC-based radiomics signature performed well for discriminating between low- and high-grade tumors, with AUCs of 0.83 in both cohorts. Based on the cut-off value of the training cohort, the radiomics signature achieved accuracies of 0.78 and 0.79, sensitivities of 0.65 and 0.71, and specificities of 0.85 and 0.82 in the training and testing cohorts, respectively. CONCLUSION: ADC-based radiomics can be a useful and promising non-invasive method for predicting histologic grade of SCC of the oral tongue and FOM.

Multiparametric Magnetic Resonance Imaging Features That Predict Treatment Response to Endovascular Sclerotherapy in Craniofacial Venous Malformations.

PURPOSE: Craniofacial venous malformations (VMs) cause severe psychological and physiological burden to patients, and treatment is meaningful only when the benefits of treatment outweigh the risks. Therefore, it is very important to predict the treatment response before treatment. This study was performed to explore the value of multiparametric magnetic resonance imaging for predicting treatment response to endovascular sclerotherapy in VMs. MATERIALS AND METHODS: We designed and implemented a case-control study and enrolled a sample from patients with VM treated by endovascular sclerotherapy at our hospital from January 2014 to January 2018. The primary predictor variables were pretreatment volume (prevolume), lesion classification, phleboliths, initial slope of increase (ISI), gender, age, and sclerosants. The primary outcome variable was treatment response (positive response or negative response). Descriptive, univariate and multivariate binary logistic regressions, and Firth's penalized maximum likelihood estimate were computed to measure the association between predictor variables and treatment response. The level of statistical significance was set at a P value less than or equal to .05. RESULTS: The sample was composed of 42 patients with a median age of 17.50 years, and 33.3% were males. There were 27 and 15 patients in the positive and negative response groups, respectively. There were significant differences between the 2 groups for ISI (adjusted odds ratio [OR], 2.184; P = .0268; 95% confidence interval [95% CI], 1.094 to 4.360), lesion classification (adjusted OR, 9.072; P = .0226; 95% CI, 1.363 to 60.400), and prevolume (adjusted OR, 1.020; P = .0268; 95% CI, 1.002 to 1.038). The cutoff point for prevolume and ISI was 40.42 cm3 and 2.61. CONCLUSIONS: Multiparametric magnetic resonance imaging could provide an approach for predicting treatment response in craniofacial VMs. When the prevolume was greater than 40.42 cm3, ISI was greater than 2.61, and the classification was infiltrating type, the response to sclerotherapy was negative.