Prediction of the Ki-67 expression level in head and neck squamous cell carcinoma with machine learning-based multiparametric MRI radiomics: a multicenter study.

BACKGROUND: This study aimed to develop and validate a machine learning (ML)-based fusion model to preoperatively predict Ki-67 expression levels in patients with head and neck squamous cell carcinoma (HNSCC) using multiparametric magnetic resonance imaging (MRI). METHODS: A total of 351 patients with pathologically proven HNSCC from two medical centers were retrospectively enrolled in the study and divided into training (n = 196), internal validation (n = 84), and external validation (n = 71) cohorts. Radiomics features were extracted from T2-weighted images and contrast-enhanced T1-weighted images and screened. Seven ML classifiers, including k-nearest neighbors (KNN), support vector machine (SVM), logistic regression (LR), random forest (RF), linear discriminant analysis (LDA), naive Bayes (NB), and eXtreme Gradient Boosting (XGBoost) were trained. The best classifier was used to calculate radiomics (Rad)-scores and combine clinical factors to construct a fusion model. Performance was evaluated based on calibration, discrimination, reclassification, and clinical utility. RESULTS: Thirteen features combining multiparametric MRI were finally selected. The SVM classifier showed the best performance, with the highest average area under the curve (AUC) of 0.851 in the validation cohorts. The fusion model incorporating SVM-based Rad-scores with clinical T stage and MR-reported lymph node status achieved encouraging predictive performance in the training (AUC = 0.916), internal validation (AUC = 0.903), and external validation (AUC = 0.885) cohorts. Furthermore, the fusion model showed better clinical benefit and higher classification accuracy than the clinical model. CONCLUSIONS: The ML-based fusion model based on multiparametric MRI exhibited promise for predicting Ki-67 expression levels in HNSCC patients, which might be helpful for prognosis evaluation and clinical decision-making.

The role of metabolic reprogramming of oxygen-induced macrophages in the dynamic changes of atherosclerotic plaques.

Atherosclerosis (As) is a chronic vascular inflammatory disease. Macrophages are the most important immune cells in atherosclerotic plaques, and the phenotype of plaque macrophages shifts dynamically to adapt to changes in the plaque microenvironment. The aerobic microenvironment of early atherosclerotic plaques promotes the transformation of M2/alternatively activated macrophages mainly through oxidative phosphorylation; the anoxic microenvironment of advanced atherosclerotic plaques mainly promotes the formation of M1/classically activated macrophages through anaerobic glycolysis; and the adventitia angiogenesis of aged atherosclerotic plaques leads to an increase in the proportion of M2/M1 macrophages. Therefore, this review deeply elucidates the dynamic change mechanism of plaque macrophages and the regulation of plaque oxygen content and immune metabolism to find new targets for the treatment of As.

Computed tomography perfusion deficit volume predicts the functional outcome of endovascular therapy for basilar artery occlusion.

OBJECTIVES: To investigate the relationship between baseline computed tomography perfusion deficit volumes and functional outcomes in patients with basilar artery occlusion (BAO) undergoing endovascular therapy. METHODS: This was a single-center study in which the data of 64 patients with BAO who underwent endovascular therapy were retrospectively analyzed. All the patients underwent multi-model computed tomography on admission. The posterior-circulation Acute Stroke Prognosis Early Computed Tomography Score was applied to assess the ischemic changes. Perfusion deficit volumes were obtained using Syngo.via software. The primary outcome of the analysis was a good functional outcome (90-day modified Rankin Scale score ≤ 3). Logistic regression and receiver operating characteristic curves were used to explore predictors of functional outcome. RESULTS: A total of 64 patients (median age, 68 years; 72 % male) were recruited, of whom 26 (41 %) patients achieved good functional outcomes, while 38 (59 %) had poor functional outcomes. Tmax > 10 s, Tmax > 6 s, and rCBF < 30 % volume were independent predictors of good functional outcomes (odds ratio range, 1.0-1.2; 95 % confidence interval [CI], 1.0-1.4]) and performed well in the receiver operating characteristic curve analyses, exhibiting positive prognostic value; the areas under the curve values were 0.85 (95 % CI, 0.75-0.94), 0.81 (95 % CI, 0.70-0.90), and 0.78 (95 % CI, 0.67-0.89). CONCLUSION: Computed tomography perfusion deficit volume represents a valuable tool in predicting high risk of disability and mortality in patients with BAO after endovascular treatment.

[Application values of 64-row computed tomographic angiography in diagnosing liver VX2 tumor in rabbits].

OBJECTIVE: To evaluate the diagnostic values of computed tomographic angiography (CTA) for liver VX2 tumor in rabbits. METHODS: A total of 40 New Zealand white rabbits were employed by implanting tumor fragment into left liver by computed tomography (CT)-guided percutaneous puncture. And successful modeling was established in 38 rabbits. Plain and three-phase enhanced CT scan and its CTA reconstruction were performed in different stages (14, 21 and 28 days) after tumor implantation, followed by digital subtraction angiography (DSA) via femoral artery and hepatic artery. Dynamic observations of tumor growth characteristics were recorded by plain and enhanced CT and comparing CTA and DSA images. And comparisons of CTA and DSA were made for identifying the origin, course and blood supply of transplanted tumors. RESULTS: (1) Multi-slice computed tomography (MSCT) could dynamically monitor the growth transfer characteristics. And all tumors became strengthened obviously in arterial phase and washed out in venous and delayed phases. (2) CTA showed that the feeding artery of transplanted tumor was hepatic artery. (3) No obvious differences existed in the display of feeding artery of transplanted tumor between CTA and DSA.However CTA was superior to DSA in displaying three-dimensional structure. CONCLUSION: Liver VX2 tumors are hypervascular in rabbits. CT enhanced scanning may dynamically monitor the growth patterns of tumors and CTA is a better method than DSA in visualizing the feeding arteries of tumors. Before and after interventions, CTA may replace DSA in evaluating liver VX2 tumors in rabbits.

Non-invasive prediction model of axillary lymph node status in patients with early-stage breast cancer: a feasibility study based on dynamic contrast-enhanced-MRI radiomics.

OBJECTIVES: Accurate axillary evaluation plays an important role in prognosis and treatment planning for breast cancer. This study aimed to develop and validate a dynamic contrast-enhanced (DCE)-MRI-based radiomics model for preoperative evaluation of axillary lymph node (ALN) status in early-stage breast cancer. METHODS: A total of 410 patients with pathologically confirmed early-stage invasive breast cancer (training cohort, N = 286; validation cohort, N = 124) from June 2018 to August 2022 were retrospectively recruited. Radiomics features were derived from the second phase of DCE-MRI images for each patient. ALN status-related features were obtained, and a radiomics signature was constructed using SelectKBest and least absolute shrinkage and selection operator regression. Logistic regression was applied to build a combined model and corresponding nomogram incorporating the radiomics score (Rad-score) with clinical predictors. The predictive performance of the nomogram was evaluated using receiver operator characteristic (ROC) curve analysis and calibration curves. RESULTS: Fourteen radiomic features were selected to construct the radiomics signature. The Rad-score, MRI-reported ALN status, BI-RADS category, and tumour size were independent predictors of ALN status and were incorporated into the combined model. The nomogram showed good calibration and favourable performance for discriminating metastatic ALNs (N + (≥1)) from non-metastatic ALNs (N0) and metastatic ALNs with heavy burden (N + (≥3)) from low burden (N + (1-2)), with the area under the ROC curve values of 0.877 and 0.879 in the training cohort and 0.859 and 0.881 in the validation cohort, respectively. CONCLUSIONS: The DCE-MRI-based radiomics nomogram could serve as a potential non-invasive technique for accurate preoperative evaluation of ALN burden, thereby assisting physicians in the personalized axillary treatment for early-stage breast cancer patients. ADVANCES IN KNOWLEDGE: This study developed a potential surrogate of preoperative accurate evaluation of ALN status, which is non-invasive and easy-to-use.

Development and Validation of a Dual-Energy CT-Based Model for Predicting the Number of Central Lymph Node Metastases in Clinically Node-Negative Papillary Thyroid Carcinoma.

RATIONALE AND OBJECTIVES: This study aimed to develop and validate a dual-energy CT (DECT)-based model for preoperative prediction of the number of central lymph node metastases (CLNMs) in clinically node-negative (cN0) papillary thyroid carcinoma (PTC) patients. MATERIALS AND METHODS: Between January 2016 and January 2021, 490 patients who underwent lobectomy or thyroidectomy, CLN dissection, and preoperative DECT examinations were enrolled and randomly allocated into the training (N = 345) and validation cohorts (N = 145). The patients' clinical characteristics and quantitative DECT parameters obtained on primary tumors were collected. Independent predictors of> 5 CLNMs were identified and integrated to construct a DECT-based prediction model, for which the area under the curve (AUC), calibration, and clinical usefulness were assessed. Risk group stratification was performed to distinguish patients with different recurrence risks. RESULTS: More than 5 CLNMs were found in 75 (15.3%) cN0 PTC patients. Age, tumor size, normalized iodine concentration (NIC), normalized effective atomic number (nZeff) and the slope of the spectral Hounsfield unit curve (λHu) in the arterial phase were independently associated with> 5 CLNMs. The DECT-based nomogram that incorporated predictors demonstrated favorable performance in both cohorts (AUC: 0.842 and 0.848) and significantly outperformed the clinical model (AUC: 0.688 and 0.694). The nomogram showed good calibration and added clinical benefit for predicting> 5 CLNMs. The KaplanMeier curves for recurrence-free survival showed that the high- and low-risk groups stratified by the nomogram were significantly different. CONCLUSION: The nomogram based on DECT parameters and clinical factors could facilitate preoperative prediction of the number of CLNMs in cN0 PTC patients.

The use of SNAP and T1-weighted VISTA in cervical artery dissection.

BACKGROUND AND PURPOSE: Some cervical artery dissection (CAD) can't be easily confirmed by commonly used angiography techniques in clinical practice. We aimed to compare the abilities of the vessel wall magnetic resonance imaging (MRI) techniques including simultaneous noncontrast angiography and intraplaque hemorrhage (SNAP) sequence and T1-weighted volumetric isotropic turbo spin echo acquisition (T1-w VISTA) sequence alone for evaluating CAD. MATERIALS AND METHODS: From July 2017 to October 2020, 59 patients underwent MRI examinations including SNAP and T1-w VISTA sequences for cervical artery pathologies. SNAP and T1-w VISTA images were retrospectively and independently reviewed to evaluate their diagnostic performances of CAD by using the final diagnosis as the reference standard which was established by clinical history, physical examination, and all available images. The agreement between T1-w VISTA and SNAP in the identification of the imaging features of CAD, including intramural hematoma (IMH), intimal flap, and double lumen, were compared. The IMH-wall contrasts by T1-w VISTA and SNAP were also compared. RESULTS: CAD was confirmed in 43 of the 59 patients. T1-w VISTA and SNAP showed the same diagnostic performance, and their consistencies with the final diagnosis were good (κ = 0.776, p < 0.001). The sensitivity and specificity in CAD diagnosis were 0.978 and 0.750 for T1-w VISTA and SNAP. The IMH, intimal flap, and double lumen observed on SNAP were also determined by T1-w VISTA (κ = 1.000, p < 0.001 for all). The SNAP sequence showed higher IMH-wall contrast than T1-w VISTA (7.34 ± 4.56 vs. 3.12 ± 1.17, p < 0.001). CONCLUSIONS: SNAP and T1-w VISTA sequences had the same performance in CAD diagnosis, thus they were both recommended. In addition, SNAP showed better IMH-wall contrast than T1-w VISTA.

The value of computed tomography perfusion deficit volumes in acute isolated brainstem infarction.

Purpose: Diagnosis of acute isolated brainstem infarction is challenging owing to non-specific, variable symptoms, and the effectiveness of non-contrast computed tomography (NCCT) is poor owing to limited spatial resolution and artifacts. Computed tomography perfusion (CTP) imaging parameters are significantly associated with functional outcomes in posterior circulation acute ischemic stroke; however, the role of CTP in isolated brainstem infarction remains unclear. We aimed to determine the value of CTP imaging parameters in predicting functional outcomes for affected patients. Methods: In total, 116 consecutive patients with isolated pontine/midbrain hypoperfusion who underwent CTP and follow-up by magnetic resonance imaging (MRI) between January 2018 and March 2022, were retrospectively analyzed. Perfusion deficit volumes on all maps, and the final infarction volume (FIV) on MRI were quantified. "Good" functional outcome was defined as a 90-day modified Rankin Scale score of 0 and 1. Statistical analysis included uni- and multivariate regression analyses, binary logistic regressions, and receiver operating characteristics (ROC) analyses. Results: In total, 113 patients had confirmed isolated pontine/midbrain infarction on follow-up MRI. Onset-to-scan time, visibility of ischemic lesions on NCCT, the baseline National Institutes of Health Stroke Scale (NIHSS) score, and perfusion deficit volumes on all CTP maps were significantly associated with FIV (p < 0.05). In a multivariate linear regression model, adjusted for age, sex, NIHSS score, onset-to-scan time, and visibility of NCCT, perfusion deficit volumes remained significantly associated with FIV. In binary logistic regression analyses, perfusion deficit volumes on all CTP maps remained independent predictors of a good functional outcome. In ROC analyses, the cerebral blood flow deficit volume showed a slightly higher discriminatory value with the largest area under the curve being 0.683 [(95% CI, 0.587-0.780), p = 0.001]. Conclusion: Perfusion deficit volumes of CTP imaging could reflect the FIV and contain prognostic information on functional outcomes in patients with acute isolated brainstem infarction.

Emerging Mechanisms and Targeted Therapy of Ferroptosis in Neurological Diseases and Neuro-oncology.

Ferroptosis is a novel type of cell death characterized by iron-dependent lipid peroxidation that involves a variety of biological processes, such as iron metabolism, lipid metabolism, and oxidative stress. A growing body of research suggests that ferroptosis is associated with cancer and neurodegenerative diseases, such as glioblastoma, Alzheimer's disease, Parkinson's disease, and stroke. Building on these findings, we can selectively induce ferroptosis for the treatment of certain cancers, or we can treat neurodegenerative diseases by inhibiting ferroptosis. This review summarizes the relevant advances in ferroptosis, the regulatory mechanisms of ferroptosis, the participation of ferroptosis in brain tumors and neurodegenerative diseases, and the corresponding drug therapies to provide new potential targets for its treatment.

Detection of mild cognitive impairment in type 2 diabetes mellitus based on machine learning using privileged information.

Type 2 diabetes mellitus (T2DM) patients may develop into mild cognitive impairment (MCI) or even dementia. However, there is lack of reliable machine learning model for detection MCI in T2DM patients based on machine learning method. In addition, the brain network changes associated with MCI have not been studied. The aim of this study is to develop a machine learning based algorithm to help detect MCI in T2DM. There are 164 participants were included in this study. They were divided into T2DM-MCI (n = 56), T2DM-nonMCI (n = 49), and normal controls (n = 59) according to the neuropsychological evaluation. Functional connectivity of each participant was constructed based on resting-state magnetic resonance imaging (rs-fMRI). Feature selection was used to reduce the feature dimension. Then the selected features were set into the cascaded multi-column random vector functional link network (RVFL) classifier model using privileged information. Finally, the optimal model was trained and the classification performance was obtained using the testing data. The results show that the proposed algorithm has outstanding performance compared with classic methods. The classification accuracy of 73.18 % (T2DM-MCI vs NC) and 79.42 % (T2DM-MCI vs T2DM-nonMCI) were achieved. The functional connectivity related to T2DM-MCI mainly distribute in the frontal lobe, temporal lobe, and central region (motor cortex), which could be used as neuroimaging biomarkers to recognize MCI in T2DM patients. This study provides a machine learning model for diagnosis of MCI in T2DM patients and has potential clinical significance for timely intervention and treatment to delay the development of MCI.

Integrated application of transcriptome and metabolomics reveals potential therapeutic targets for the polarization of atherosclerotic macrophages.

The polarization of macrophages often leads to severe calcification and necrosis in aged atherosclerotic plaques, which eventually leads to poor prognosis of ischaemic cardiovascular and cerebrovascular diseases. More reliable diagnostic methods are urgently needed to discover therapeutic targets of macrophage polarization in aged atherosclerotic plaques. Metabolomics of aged plaques (n = 20) and macrophage polarization transcriptomes (n = 30) were integrated to identify metabolic therapeutic targets of macrophage polarization associated with aged plaque. Finally, metabolic inhibitors were used to verify the reliability of the target genes. Integrated multiomics analysis revealed that 6 metabolic pathways (including 21 genes) regulate macrophage polarization in aged atherosclerosis. Targeted treatment of macrophage polarization with metabolic inhibitors can effectively reduce the adverse risk of aged atherosclerosis. The combination of transcriptomics and metabolomics approaches can identify effective therapeutic targets for macrophage polarization in arteriosclerosis.

Intraplacental T2-hypointense bands may help predict placental invasion depth and postpartum hemorrhage in placenta accrete spectrum disorders in high-risk gravid patients.

INTRODUCTION: Accurate noninvasive prenatal diagnosis of placenta accreta spectrum (PAS) disorder is critical for delivery management. PURPOSE: To evaluate the diagnostic ability of MRI features in predicting the PAS, invasive depth and postpartum hemorrhage (PPH) in high-risk gravid patients. MATERIALS AND METHODS: Between February 2019 and November 2020, women with ultrasound (US)-suspected PAS were enrolled. With the exclusion criteria, 80 women were included in the study. Two experienced genitourinary radiologists reviewed and recorded the MRI features. The chi square test was used to compare the effectiveness of MRI features. Relative risk ratios were computed to test the association of intraplacental T2-hypointense bands with poor outcomes of cesarean section. Receiver operating characteristic (ROC) analyses based on the number and area of intraplacental T2-hypointense bands were used to predict PAS, invasion depth, and PPH. RESULTS: PAS was diagnosed in 56 of 80 women (70%). At delivery, 24 of 80 women (30%) experienced PPH (≥1000 mL). Intraplacental T2-hypointense bands were detected at MRI in 28 of 56 women with PAS (50%). The relative risk ratio of intraplacental T2-hypointense bands was 1.51 for PAS, 3.17 for depth of PAS invasiveness and 4.74 for PPH. The largest areas of intraplacental T2-hypointense bands for predicting PAS, invasion depth and PPH were 0.66 cm2, 1.68 cm2 and 1.99 cm2, respectively. DISCUSSION: The appearance of intraplacental T2-hypointense bands has important diagnostic value for PAS, its invasion depth and PPH. The area of the largest T2-hypointense band in the placenta can predict poor outcomes of cesarean section.

Increased aneurysm wall permeability colocalized with low wall shear stress in unruptured saccular intracranial aneurysm.

Aneurysm wall permeability has recently emerged as an in vivo marker of aneurysm wall remodeling. We sought to study the spatial relationship between hemodynamic forces derived from 4D-flow MRI and aneurysm wall permeability by DCE-MRI in a region-based analysis of unruptured saccular intracranial aneurysms (IAs). We performed 4D-flow MRI and DCE-MRI on patients with unruptured IAs of ≥ 5 mm to measure hemodynamic parameters, including wall shear stress (WSS), oscillatory shear index (OSI), WSS temporal (WSSGt) and spatial (WSSGs) gradient, and aneurysm wall permeability (Ktrans) in different sectors of aneurysm wall defined by evenly distributed radial lines emitted from the aneurysm center. The spatial association between Ktrans and hemodynamic parameters measured at the sector level was evaluated. Thirty-one patients were scanned. Ktrans not only varied between aneurysms but also demonstrated spatial heterogeneity within an aneurysm. Among all 159 sectors, higher Ktrans was associated with lower WSS, which was seen in both Spearman's correlation analysis (rho = - 0.18, p = 0.025) and linear regression analysis using generalized estimating equation to account for correlations between multiple sectors of the same aneurysm (regression coefficient = - 0.33, p = 0.006). Aneurysm wall permeability by DCE-MRI was shown to be spatially heterogenous in unruptured saccular IAs and associated with local WSS by 4D-flow MRI.