An automatic and accurate deep learning-based neuroimaging pipeline for the neonatal brain.

BACKGROUND: Accurate segmentation of neonatal brain tissues and structures is crucial for studying normal development and diagnosing early neurodevelopmental disorders. However, there is a lack of an end-to-end pipeline for automated segmentation and imaging analysis of the normal and abnormal neonatal brain. OBJECTIVE: To develop and validate a deep learning-based pipeline for neonatal brain segmentation and analysis of structural magnetic resonance images (MRI). MATERIALS AND METHODS: Two cohorts were enrolled in the study, including cohort 1 (582 neonates from the developing Human Connectome Project) and cohort 2 (37 neonates imaged using a 3.0-tesla MRI scanner in our hospital).We developed a deep leaning-based architecture capable of brain segmentation into 9 tissues and 87 structures. Then, extensive validations were performed for accuracy, effectiveness, robustness and generality of the pipeline. Furthermore, regional volume and cortical surface estimation were measured through in-house bash script implemented in FSL (Oxford Centre for Functional MRI of the Brain Software Library) to ensure reliability of the pipeline. Dice similarity score (DSC), the 95th percentile Hausdorff distance (H95) and intraclass correlation coefficient (ICC) were calculated to assess the quality of our pipeline. Finally, we finetuned and validated our pipeline on 2-dimensional thick-slice MRI in cohorts 1 and 2. RESULTS: The deep learning-based model showed excellent performance for neonatal brain tissue and structural segmentation, with the best DSC and the 95th percentile Hausdorff distance (H95) of 0.96 and 0.99 mm, respectively. In terms of regional volume and cortical surface analysis, our model showed good agreement with ground truth. The ICC values for the regional volume were all above 0.80. Considering the thick-slice image pipeline, the same trend was observed for brain segmentation and analysis. The best DSC and H95 were 0.92 and 3.00 mm, respectively. The regional volumes and surface curvature had ICC values just below 0.80. CONCLUSIONS: We propose an automatic, accurate, stable and reliable pipeline for neonatal brain segmentation and analysis from thin and thick structural MRI. The external validation showed very good reproducibility of the pipeline.

Combining quantitative susceptibility mapping to radiomics in diagnosing Parkinson's disease and assessing cognitive impairment.

OBJECTIVE: To explore whether magnetic susceptibility value (MSV) and radiomics features of the nigrostriatal system could be used as imaging markers for diagnosing Parkinson's disease (PD) and its related cognitive impairment (CI). METHODS: A total of 104 PD patients and 45 age-sex-matched healthy controls (HCs) underwent quantitative susceptibility mapping (QSM). The former completed Hoehn-Yahr Stage and Montreal Cognitive Assessment (MoCA). The patients were divided into several subgroups according to disease stages, courses, and MoCA scores. The ROI was subdivided into the substantia nigra (SN), head of caudate nucleus (HCN), and putamen. The MSVs and radiomics features were obtained from QSM. The multivariable logistic regression (MLR) and support vector machine (SVM) models were constructed to diagnose PD. The correlations between MSVs, radiomics features, and MoCA scores were evaluated. RESULTS: The MSVs in bilateral SN pars compacta (SNc) of PD patients were higher than those of the HCs (p < 0.001). There were differences in some radiomics features between the two groups (p < 0.05). The MSVs of the right SNc and the radiomics features of the right SN had the highest area under the curve (AUC), respectively. The comprehensive MLR model (0.90) and SVM model (0.95) revealed better classification performance than MSVs (p < 0.05) in diagnosing PD. The MSVs from the HCN were negatively correlated with MoCA scores in PD subgroups. There were correlations between radiomics features and MoCA scores in PD patients. CONCLUSIONS: Radiomics features and MSVs of the nigrostriatal system from QSM could have crucial role in diagnosing PD and assessing CI. KEY POINTS: • The MLR and the SVM models have excellent diagnostic performance in the diagnosis of PD. • A PD diagnostic nomogram, created based on MSV and the radiomics scores of SVM model, is very convenient for clinical use. • The radiomics features of the nigrostriatal system based on QSM help to evaluate the cognitive impairment in PD patients.

T1-Weighted Imaging-Based Hippocampal Radiomics in the Diagnosis of Alzheimer's Disease.

RATIONALE AND OBJECTIVES: To investigate the potential of T1-weighted imaging (T1WI)-based hippocampal radiomics as imaging markers for the diagnosis of Alzheimer's disease (AD) and their efficacy in discriminating between mild cognitive impairment (MCI) and dementia in AD. METHODS: A total of 126 AD patients underwent T1WI-based magnetic resonance imaging (MRI) examinations, along with 108 age-sex-matched healthy controls (HC). This was a retrospective, single-center study conducted from November 2021 to February 2023. AD patients were categorized into two groups based on disease progression and cognitive function: AD-MCI and dementia (AD-D). T1WI-based radiomics features of the bilateral hippocampi were extracted. To diagnose AD and differentiate between AD-MCI and AD-D, predictive models were developed using random forest (RF), logistic regression (LR), and support vector machine (SVM). We compared radiomics features between the AD and HC groups, as well as within the subgroups of AD-MCI and AD-D. Area under the curve (AUC), accuracy, sensitivity, and specificity were all used to assess model performance. Furthermore, correlations between radiomics features and Mini-Mental State Examination (MMSE) scores, tau protein phosphorylated at threonine 181 (P-tau-181), and amyloid ß peptide1-42 (Aß1-42) were analyzed. RESULTS: The RF model demonstrated superior performance in distinguishing AD from HC (AUC=0.961, accuracy=90.8%, sensitivity=90.7%, specificity=90.9%) and in identifying AD-MCI and AD-D (AUC=0.875, accuracy=80.7%, sensitivity=87.2%, specificity=73.2%) compared to the other models. Additionally, radiomics features were correlated with MMSE scores, P-tau-181, and Aß1-42 levels in AD. CONCLUSION: T1WI-based hippocampal radiomics features are valuable for diagnosing AD and identifying AD-MCI and AD-D.

DL-3-n-butylphthalide alleviates motor disturbance by suppressing ferroptosis in a rat model of Parkinson's disease.

DL-3-n-butylphthalide (NBP)-a compound isolated from Apium graveolens seeds-is protective against brain ischemia via various mechanisms in humans and has been approved for treatment of acute ischemic stroke. NBP has shown recent potential as a treatment for Parkinson's disease. However, the underlying mechanism of action of NBP remains poorly understood. In this study, we established a rat model of Parkinson's disease by intraperitoneal injection of rotenone for 28 successive days, followed by intragastric injection of NBP for 14-28 days. We found that NBP greatly alleviated rotenone-induced motor disturbance in the rat model of Parkinson's disease, inhibited loss of dopaminergic neurons and aggregation of α-synuclein, and reduced iron deposition in the substantia nigra and iron content in serum. These changes were achieved by alterations in the expression of the iron metabolism-related proteins transferrin receptor, ferritin light chain, and transferrin 1. NBP also inhibited oxidative stress in the substantia nigra and protected mitochondria in the rat model of Parkinson's disease. Our findings suggest that NBP alleviates motor disturbance by inhibition of iron deposition, oxidative stress, and ferroptosis in the substantia nigra.

Non-Invasive Estimation of Glioma IDH1 Mutation and VEGF Expression by Histogram Analysis of Dynamic Contrast-Enhanced MRI.

OBJECTIVES: To investigate whether glioma isocitrate dehydrogenase (IDH) 1 mutation and vascular endothelial growth factor (VEGF) expression can be estimated by histogram analysis of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). METHODS: Chinese Glioma Genome Atlas (CGGA) database was wined for differential expression of VEGF in gliomas with different IDH genotypes. The VEGF expression and IDH1 genotypes of 56 glioma samples in our hospital were assessed by immunohistochemistry. Preoperative DCE-MRI data of glioma samples were reviewed. Regions of interest (ROIs) covering tumor parenchyma were delineated. Histogram parameters of volume transfer constant (Ktrans ) and volume of extravascular extracellular space per unit volume of tissue (Ve ) derived from DCE-MRI were obtained. Histogram parameters of Ktrans , Ve and VEGF expression of IDH1 mutant type (IDH1mut ) gliomas were compared with the IDH1 wildtype (IDH1wt ) gliomas. Receiver operating characteristic (ROC) curve analysis was performed to differentiate IDH1mut from IDH1wt gliomas. The correlation coefficients were determined between histogram parameters of Ktrans , Ve and VEGF expression in gliomas. RESULTS: In CGGA database, VEGF expression in IDHmut gliomas was lower as compared to wildtype counterpart. The immunohistochemistry of glioma samples in our hospital also confirmed the results. Comparisons demonstrated statistically significant differences in histogram parameters of Ktrans and Ve [mean, standard deviation (SD), 50th, 75th, 90th. and 95th percentile] between IDH1mut and IDH1wt gliomas (P < 0.05, respectively). ROC curve analysis revealed that 50th percentile of Ktrans (0.019 min-1) and Ve (0.039) provided the perfect combination of sensitivity and specificity in differentiating gliomas with IDH1mut from IDH1wt . Irrespective of IDH1 mutation, histogram parameters of Ktrans and Ve were correlated with VEGF expression in gliomas (P < 0.05, respectively). CONCLUSIONS: VEGF expression is significantly lower in IDH1mut gliomas as compared to the wildtype counterpart, and it is non-invasively predictable with histogram analysis of DCE-MRI.

Volume-based histogram analysis of dynamic contrast-enhanced MRI for estimation of gliomas IDH1 mutation status.

PURPOSE: The study aimed to investigate whether isocitrate dehydrogenase 1 (IDH1) mutation status in gliomas can be estimated by volume-based histogram analysis of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). MATERIALS AND METHODS: Preoperative DCE-MRI data of 85 pathologically confirmed glioma patients including 33 carrying IDH1 mutant type (IDH1mut) and 52 with IDH1 wildtype (IDH1wt) were reviewed in a retrospective approach. Regions of interest (ROI) covering entire tumor volume were manually delineated using O.K. software (OmniKinetics, GE Healthcare, China). Histogram parameters of volume transfer constant (Ktrans) and volume of extravascular /extracellular space per unit volume of tissue (Ve) derived from DCE-MRI were obtained. Mann-Whitney U tests were made to compare the differences in histogram parameters of Ktrans and Ve between IDH1mut and IDH1wt in all gliomas and high-grade gliomas (HGGs, grade III and IV). Receiver operator characteristic (ROC) analysis were implemented to assess the diagnostic performance. RESULTS: In histogram parameters of Ktrans and Ve, pairwise comparisons demonstrated statistically significant differences in mean, standard deviation (SD), 90th and 95th percentiles (90%, 95%) values between IDH1mut and IDH1wt in all cases of gliomas and HGGs (P < 0.05, respectively). The ROC analysis revealed that the cut-off values of 95% value of Ktrans (0.097 min-1) and mean value of Ve (0.099) provided the best combination of sensitivity and specificity to distinguish all gliomas with IDH1mut from IDH1wt. In HGGs, the cut-off values of mean value of Ktrans and Ve (0.044 min-1, 0.099) played similar role. CONCLUSION: Volume-based histogram analysis of DCE-MRI performs well in identification of IDH1mut gliomas.

[Kinematics of the triangular fibrocartilage complex during forearm rotation in vivo].

OBJECTIVE: To investigate three-dimensional kinematics of the superficial and deep portion of triangular fibrocartilage complex (TFCC) in different parts of the forearm rotation. METHODS: Six wrists of 6 volunteers were used to obtain CT scans at different positions of the wrist. The wrists were scanned from 90 degrees of pronation to 90 degrees of supination at an interval of 30 degrees. The 3-dimensional radius and ulna were reconstructed with customized software and changes in length of the superficial and deep portion of TFCC during forearm rotation. RESULTS: In forearm pronation, the superficial dorsal portion and the deep palmar portion of the TFCC were tight. While the superficial palmar portion and the deep dorsal potion of the TFCC were lax. In supination, the changes in length of all these fibers were reverse. CONCLUSIONS: In forearm rotation one portion fibers of dorsal TFCC and one portion fibers of palmar TFCC are tight, and this mechanism controls stability during DRUJ rotation.

Quantitative Assessment of Tumor Cell Proliferation in Brain Gliomas with Dynamic Contrast-Enhanced MRI.

RATIONALE AND OBJECTIVES: This study aimed to investigate whether volume transfer constant (Ktrans) and volume of extravascular extracellular space per unit volume of tissue (Ve) derived from dynamic contrast-enhanced magnetic resonance imaging (DCE MRI) could quantitatively assess the tumor proliferation index (Ki-67) of gliomas noninvasively. MATERIALS AND METHODS: The preoperative DCE MRI data of 69 patients with pathologically confirmed glioma (28, 8, and 33 cases in grades Ⅱ, Ⅲ, and Ⅳ) were retrospectively reviewed. The maximal Ktrans and Ve were measured in the tumor body. The immunohistochemistry was used to detect the expression of Ki-67 proteins in glioma specimens. The Mann-Whitney U test was applied to analyze the differences in Ktrans, Ve, and Ki-67 index across histologically defined glioma grades. Spearman correlation was performed between Ktrans, Ve, and Ki-67 index. The receiver operating characteristic curve analysis was used to determine the cutoff values of Ktrans and Ve in distinguishing different Ki-67 index expression levels. RESULTS: Ktrans, Ve, and Ki-67 index of grade Ⅱ (0.027 min-1, 0.065, 4.04%) were significantly lower than those of grade Ⅲ (0.093 min-1, 0.297, 25.13%) and Ⅳ (0.100 min-1, 0.299, 25.37%). Both Ktrans and Ve significantly correlated with the Ki-67 index in all tumors and high-grade gliomas (HGGs, grade Ⅲ and Ⅳ). The receiver operating characteristic curve analysis revealed that the cutoff values for Ktrans (0.079 min-1) and Ve (0.249) provided the best combination of sensitivity and specificity to distinguish the gliomas with high Ki-67 index from those with low Ki-67 index. CONCLUSION: The DCE MRI-derived parameters were valuable in assessing the tumor cell proliferation in HGG noninvasively.

[Application of apparent diffusion coefficient and fractional anisotropy in identification of tumor component and grading of brain astrocytoma].

OBJECTIVE: To evaluate the value of apparent diffusion coefficient (ADC) and fractional anisotropy (FA) in identification of tumor element and grading of brain astrocytoma. METHODS: Thirty-three patients with histologically confirmed astrocytoma underwent diffusion weighted imaging (DWI), diffusion tensor imaging (DTI), and conventional MRI before operation. The values of ADC and FA of different regions in the same tumor and of astrocytoma of different grades were measured and compared. RESULTS: The ADC values of the tumor parenchyma, necrotic region, peritumoral edema region were (1.28 +/- 0.44), (1.97 +/- 0.53), and (1.74 +/- 0.47) respectively, all significantly higher than that of the corresponding normal brain tissues [(0.80 +/- 0.18), P = 0.009, P = 0.000, P = 0.000] with significantly differences between the tumor parenchyma and necrotic region and peritumoral edema region (both P < 0.05), however, there was not significant difference between the necrotic region and peritumoral edema region. The FA values of the tumor parenchyma, necrotic region, and peritumoral edema region were (0.18 +/- 0.07), (0.14 +/- 0.05), and (0.16 +/- 0.05) respectively, all significantly higher than that of the corresponding normal brain tissues [(0.58 +/- 0.10), all P = 0.000], without significant differences among the former 3 groups. There were no significant differences in the ADC and FA values among the tumors at different grades, however, there was a tendency of ADC to decrease and of FA to increase along the increase of grade of tumor, although not significantly. CONCLUSION: ADC value plays an important part in distinguishing tumor components and determining tumor boundary, and plays a certain role in judging the grade of astrocytomas. FA value is vital to determine the tumor boundary, and has certain value in differentiating high-grade from low-grade astrocytomas.

Comparison between perfusion computed tomography and dynamic contrast-enhanced magnetic resonance imaging in assessing glioblastoma microvasculature.

PURPOSE: Perfusion computed tomography (PCT) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) provide independent measurements of biomarkers related to tumor perfusion. The aim of this study was to compare the two techniques in assessing glioblastoma microvasculature. MATERIALS AND METHODS: Twenty-five patients diagnosed with glioblastoma (14 males and 11 females; 51±11years old, ranging from 33 to 70 years) were includede in this prospective study. All patients underwent both PCT and DCE-MRI. Imaging was performed on a 256-slice CT scanner and a 3-T MRI system. PCT yielded permeability surface-area product (PS) using deconvolution physiological models; meanwhile, DCE-MRI determined volume transfer constant (Ktrans) using the Tofts-Kermode compartment model. All cases were submitted to surgical intervention, and CD105-microvascular density (CD105-MVD) was measured in each glioblastoma specimen. Then, Spearman's correlation coefficients and Bland-Altman plots were obtained for PS, Ktrans and CD105-MVD. P<0.05 was considered statistically significant. RESULTS: Tumor PS and Ktrans values were correlated with CD105-MVD (r=0.644, P<0.001; r=0.683, P<0.001). In addition, PS was correlated with Ktrans in glioblastoma (r=0.931, P<0.001). Finally, Bland-Altman plots showed no significant differences between PS and Ktrans (P=0.063). CONCLUSION: PCT and DCE-MRI measurements of glioblastoma perfusion biomarkers have similar results, suggesting that both techniques may have comparable utility. Therefore, PCT may serve as an alternative modality to DCE-MRI for the in vivo evaluation of glioblastoma microvasculature.

The assessment of immature microvascular density in brain gliomas with dynamic contrast-enhanced magnetic resonance imaging.

PURPOSE: This study was designed to quantitatively evaluate the immature microvascular density (MVD) of brain gliomas using the volume transfer constant (K(trans)) and volume of extravascular extracellular space per unit volume of tissue (Ve) from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) noninvasively. MATERIALS AND METHODS: Fifty-seven patients (35 males, 22 females; age range, 14-70, mean age 46±12 years old) with brain glioma were included in this study. The maximal values of K(trans) and Ve of all patients with brain glioma (grade II 24, III 7 and IV 26) were obtained. The CD105-microvascular density (CD105-MVD) of each tumor was measured in surgical specimen. The differences of K(trans), Ve and CD105-MVD between the different grades of gliomas were analyzed using the Mann-Whitney U-test. The Pearman correlation coefficient was determined between K(trans), Ve and CD105-MVD. A P-value of less than 0.05 was considered statistically significant. RESULTS: The differences in K(trans), Ve and CD105-MVD were statistically significant between low-grade glioma (LGG) and high-grade glioma (HGG) (P=0.001, P<0.001, P<0.001). The K(trans), Ve and CD105-MVD of grade II were significantly lower than those of grade III and IV. K(trans) and Ve were positively correlated with CD105-MVD in HGG (P<0.001, P<0.001). CONCLUSIONS: Our results suggest DCE-MRI plays an important part in noninvasively evaluating the immature MVD of brain gliomas.

Microvascular permeability of brain astrocytoma with contrast-enhanced magnetic resonance imaging: correlation analysis with histopathologic grade.

BACKGROUND: The degree of pathological microvascular proliferation is an important element in evaluation of the astrocytoma grade. This study was aimed to quantitatively assess the microvascular permeability of brain astrocytoma with the volume transfer constant (K(trans)) and volume of extravascular extracellular space per unit volume of tissue (Ve) from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and to evaluate the effectiveness of the K(trans) and Ve in the grading of astrocytoma. METHODS: The highest values of the K(trans) and Ve of 67 patients with astrocytoma (27 with grade II, 12 with grade III, and 28 with grade IV) were obtained. The comparisons of the differences of the K(trans) and Ve between the different grades were conducted using the Mann-Whitney rank-sum tests. Spearman's rank correlation coefficients were determined between K(trans) values, Ve values and astrocytoma grades. Receiver operating characteristic (ROC) curve analyses were performed to determine the cut-off values for the K(trans) and Ve to distinguish between the different grades of astrocytoma. RESULTS: There were significant differences (P < 0.001) between the different grades in the K(trans) values and Ve values, except for grades III and IV. The K(trans) values and Ve values were both correlated with astrocytoma grades (both P < 0.001). The ROC curve analyses showed that the cut-off values for the K(trans) and Ve provided the best combination of sensitivity and specificity in distinguishing between grade II and grade III or IV astrocytomas. CONCLUSIONS: DCE-MRI can play an important role in assessing the microvascular permeability and the grading of brain astrocytoma.