Automatic detection of mild cognitive impairment based on deep learning and radiomics of MR imaging.

Purpose: Early and rapid diagnosis of mild cognitive impairment (MCI) has important clinical value in improving the prognosis of Alzheimer's disease (AD). The hippocampus and parahippocampal gyrus play crucial roles in the occurrence of cognitive function decline. In this study, deep learning and radiomics techniques were used to automatically detect MCI from healthy controls (HCs). Method: This study included 115 MCI patients and 133 normal individuals with 3D-T1 weighted MR structural images from the ADNI database. The identification and segmentation of the hippocampus and parahippocampal gyrus were automatically performed with a VB-net, and radiomics features were extracted. Relief, Minimum Redundancy Maximum Correlation, Recursive Feature Elimination and the minimum absolute shrinkage and selection operator (LASSO) were used to reduce the dimensionality and select the optimal features. Five independent machine learning classifiers including Support Vector Machine (SVM), Random forest (RF), Logistic Regression (LR), Bagging Decision Tree (BDT), and Gaussian Process (GP) were trained on the training set, and validated on the testing set to detect the MCI. The Delong test was used to assess the performance of different models. Result: Our VB-net could automatically identify and segment the bilateral hippocampus and parahippocampal gyrus. After four steps of feature dimensionality reduction, the GP models based on combined features (11 features from the hippocampus, and 4 features from the parahippocampal gyrus) showed the best performance for the MCI and normal control subject discrimination. The AUC of the training set and test set were 0.954 (95% CI: 0.929-0.979) and 0.866 (95% CI: 0.757-0.976), respectively. Decision curve analysis showed that the clinical benefit of the line graph model was high. Conclusion: The GP classifier based on 15 radiomics features of bilateral hippocampal and parahippocampal gyrus could detect MCI from normal controls with high accuracy based on conventional MR images. Our fully automatic model could rapidly process the MRI data and give results in 1 minute, which provided important clinical value in assisted diagnosis.

Response of sediment transport capacity to soil properties and hydraulic parameters in the typical agricultural regions of the Loess Plateau.

The sediment transport capacity by overland flow (Tc) is a key parameter in process-based soil erosion models and Tc variation is sensitive to changes in soil properties. This study was undertaken to investigate Tc variations with respect to soil properties and establish a universal relationship to predict Tc. The test soils were collected from typical agricultural regions (Guanzhong basin-Yangling (YL), Weibei Dry plateau-Chunhua (CH), Hilly and gully region-Ansai (AS), Ago-pastoral transition zone along the Great Wall-Yuyang (YY), and Weiriver floodplain-Weicheng (WC)) of the Loess Plateau, and subjected to 36 different combinations of slope gradients (S, 5.24-44.52 %) and flow discharge (q, 0.00033-0.00125 m2 s-1) in a hydraulic flume. The results showed that the mean Tc values for WC were 2.15, 1.38, 1.32, and 1.16 times greater than those for YL, CH, AS, and YY, respectively. Tc significantly decreased with clay content (C), mean weight diameter (MWD), and soil organic matter content (SOM). Tc for different soil types increased with S and q as a binary power function, and Tc variation was more sensitive to S than to q. Stream power (w) was the most appropriate hydraulic variable to express Tc for different soils. Tc for different soil types could be satisfactorily simulated using a quaternary power function of S, q, C, and MWD (R2 = 0.94; NSE = 0.94) or a ternary power function of w, C, and MWD (R2 = 0.94; NSE = 0.94). The new Tc equation can reflect the effect of soil properties on it and facilitate the development of a process-based soil erosion model.

Automatic diagnosis of Parkinson's disease using artificial intelligence base on routine T1-weighted MRI.

Background: Parkinson's disease (PD) is the second most common neurodegenerative disease. An objective diagnosis method is urgently needed in clinical practice. In this study, deep learning and radiomics techniques were studied to automatically diagnose PD from healthy controls (HCs). Methods: 155 PD patients and 154 HCs were randomly divided into a training set (246 patients) and a testing set (63 patients). The brain subregions identification and segmentation were automatically performed with a VB-net, and radiomics features of billateral thalamus, caudatum, putamen and pallidum were extracted. Five independent machine learning classifiers [Support Vector Machine (SVM), Stochastic gradient descent (SGD), random forest (RF), quadratic discriminant analysis (QDA) and decision tree (DT)] were trained on the training set, and validated on the testing. Delong test was used to compare the performance of different models. Results: Our VB-net could automatically identify and segment the brain into 109 regions. 2,264 radiomics features were automatically extracted from the billateral thalamus, caudatum, putamen or pallidum of each patient. After four step of features dimensionality reduction, Delong tests showed that the SVM model based on combined features had the best performance, with AUCs of 0.988 (95% CI: 0.979 ~ 0.998, specificity = 91.1%, sensitivity =100%, accuracy = 89.4% and precision = 88.2%) and 0.976 (95% CI: 0.942 ~ 1.000, specificity = 100%, sensitivity = 87.1%, accuracy = 93.5% and precision = 88.6%) in the training set and testing set, respectively. Decision curve analysis showed that the clinical benefit of the line graph model was high. Conclusion: The SVM model based on combined features could be used to diagnose PD with high accuracy. Our fully automatic model could rapidly process the MRI data and distinguish PD and HCs in one minute. It greatly improved the diagnostic efficiency and has a great potential value in clinical practice to help the early diagnosis of PD.

Topological Alterations of the Brain Functional Network in Type 2 Diabetes Mellitus Patients With and Without Mild Cognitive Impairment.

The aim of this study was to explore the topological alterations of the brain functional network in type 2 diabetes mellitus (T2DM) patients with and without mild cognitive impairment (MCI) using resting-state functional magnetic resonance imaging (rs-fMRI) and graph theory approaches. In total, 27 T2DM patients with MCI, 27 T2DM patients without MCI, and 27 healthy controls (HCs) underwent rs-fMRI scanning. The whole-brain functional network was constructed by thresholding the Pearson's correlation matrices of 90 brain regions. The topological organization of the constructed networks was analyzed by using graph theory approaches. The global and nodal properties of the participants in the three groups were compared by using one-way ANOVA as well as post hoc Tukey's t-tests. The relationships between the altered topological properties and clinical features or scores of neuropsychological tests were analyzed in T2DM patients with MCI. At the global level, the global and local efficiency of the patients in the T2DM with MCI group were significantly higher than that of participants in the HCs group, and the length of the characteristic path was significantly lower than that of the participants in the HCs group (p < 0.05). No significant difference was found among the other groups. At the nodal level, when compared with T2DM patients without MCI, T2DM patients with MCI showed significantly increased nodal centrality in four brain regions, which were mainly located in the orbitofrontal lobe and anterior cingulate gyrus (ACG) (p < 0.05). No significant difference was found between the T2DM patients without MCI and HCs. Moreover, nodal degree related coefficient (r = -0381, p = 0.050) and nodal efficiency (r = -0.405, P = 0.036) of the ACG showed a significant closed correlation with the scores of the digit span backward test in the T2DM patients with MCI. Our results suggested that the increased nodal properties in brain regions of the orbitofrontal lobe and ACG were biomarkers of cognitive impairment in T2DM patients and could be used for its early diagnosis. The global topological alterations may be related to the combination of MCI and T2DM, rather than any of them.

Differentiating Parkinson's disease motor subtypes: A radiomics analysis based on deep gray nuclear lesion and white matter.

OBJECTIVE: To investigate the feasibility of radiomics analysis of brain MR images to differentiate Parkinson's disease motor subtypes. METHODS: 42 postural instability gait difficulty (PIGD) patients, 92 tremor-dominant (TD) patients and 96 healthy controls were included from the Parkinson's Progressive Marker Initiative public database. For each subject, 4850 radiomic features from 148 cortical and 14 subcortical brain regions were extracted. The variance threshold and the least absolute shrinkage and selection operator were used to select the optimal features. Classification models based on Support Vector Machine, Logistic Regrcession, and Multi-Layer Perceptron were constructed to assess the performance of optimal features in the discrimination of the two subtypes. Correlations between radiomic features and clinical scores of the two subtypes were estimated. RESULTS: The Support Vector Machine demonstrated the best performance in discriminating between the two subtypes, and the mean area under the curve was 0.833 (specificity = 83.3%, sensitivity = 75.0%, and accuracy = 80.7%). For the postural instability gait difficulty patients, these optimal features in the hippocampal showed closed correlations with the Montreal Cognitive Assessment scores (P < 0.05). CONCLUSION: The results of our study provide preliminary evidence that radiomics analysis of brain MR images could allow discrimination between patients with TD, PIGD and control subjects and has great potential value in the clinical practice.

Individualized Prediction of Early Alzheimer's Disease Based on Magnetic Resonance Imaging Radiomics, Clinical, and Laboratory Examinations: A 60-Month Follow-Up Study.

BACKGROUND: Accurately predicting whether and when mild cognitive impairment (MCI) will progress to Alzheimer's disease (AD) is of vital importance to help developing individualized treatment plans to defer the occurrence of irreversible dementia. PURPOSE: To develop and validate radiomics models and multipredictor nomogram for predicting the time to progression (TTP) from MCI to AD. STUDY TYPE: Retrospective. POPULATION: One hundred sixty-two MCI patients (96 men and 66 women [median age, 72; age range, 56-88 years]) were included from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. FIELD STRENGTH/SEQUENCE: T1 -weighted imaging and T2 -weighted fluid-attenuation inversion recovery imaging acquired at 3.0 T. ASSESSMENT: During the 5-year follow-up, 68 patients converted to AD and 94 remained stable. Patients were randomly divided into the training (n = 112) and validation datasets (n = 50). Radiomic features were extracted from the whole cerebral cortex and subcortical nucleus of MR images. A radiomics model was established using least absolute shrinkage and selection operator (LASSO) Cox regression. The clinical-laboratory model and radiomics-clinical-laboratory model were developed by multivariate Cox proportional hazard model. The performance of each model was assessed by the concordance index (C-index). A multipredictor nomogram derived from the radiomics-clinical-laboratory model was constructed for individualized TTP estimation. STATISTICAL TESTS: LASSO cox regression, univariate and multivariate Cox regression, Kaplan-Meier analysis and Student's t test were performed. RESULTS: The C-index of the radiomics, clinical-laboratory and radiomics-clinical-laboratory models were 0.924 (95% confidence interval [CI]: 0.894-0.952), 0.903 (0.868-0.938), 0.950 (0.929-0.971) in the training cohort and 0.811 (0.707-0.914), 0.901 (0824-0.977), 0.907 (0.836-0.979) in the validation cohort, respectively. A multipredictor nomogram with 15 predictors was established, which had high accuracy for individual TTP prediction with the C-index of 0.950 (0.929-0.971). DATA CONCLUSION: The prediction of individual TTP from MCI to AD could be accurately conducted using the radiomics-clinical-laboratory model and multipredictor nomogram. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: 2.

The Clinical and Chest CT Features Associated With Severe and Critical COVID-19 Pneumonia.

OBJECTIVE: The aim of this study was to investigate the clinical and computed tomography (CT) features associated with severe and critical coronavirus disease 2019 (COVID-19) pneumonia. MATERIALS AND METHODS: Eighty-three patients with COVID-19 pneumonia including 25 severe/critical cases and 58 ordinary cases were enrolled. The chest CT images and clinical data of them were reviewed and compared. The risk factors associated with disease severity were analyzed. RESULTS: Compared with the ordinary patients, the severe/critical patients had older ages, higher incidence of comorbidities, cough, expectoration, chest pain, and dyspnea. The incidences of consolidation, linear opacities, crazy-paving pattern, and bronchial wall thickening in severe/critical patients were significantly higher than those of the ordinary patients. Besides, severe/critical patients showed higher incidences of lymph node enlargement, pericardial effusion, and pleural effusion than the ordinary patients. The CT scores of severe/critical patients were significantly higher than those of the ordinary patients (P < 0.001). Receiver operating characteristic curve showed that the sensitivity and specificity of CT score were 80.0% and 82.8%, respectively, for the discrimination of the 2 types. The clinical factors of age older than 50 years, comorbidities, dyspnea, chest pain, cough, expectoration, decreased lymphocytes, and increased inflammation indicators were risk factors for severe/critical COVID-19 pneumonia. Computed tomography findings of consolidation, linear opacities, crazy-paving pattern, bronchial wall thickening, high CT scores, and extrapulmonary lesions were features of severe/critical COVID-19 pneumonia. CONCLUSIONS: There are significant differences in clinical symptoms, laboratory examinations, and CT manifestations between the ordinary patients and the severe/critical patients. Many factors are related to the severity of the disease, which can help clinicians to judge the severity of the patient and evaluate the prognosis.

Pollution characteristics and ecological risk assessment of heavy metals in three land-use types on the southern Loess Plateau, China.

The accumulation of heavy metals in agricultural soils has been the subject of great concern because these metals have the potential to be transferred to soil solutions and subsequently accumulate in the food chain. To study the persistence of trace metals in crop and orchard soils, representative surface soil samples were collected from terrace farmland that had been cultivated for various numbers of years (3, 8, 12, 15, and >20 years), terrace orchard land that had been cultivated for various numbers of years (4, 7, 10, 12, 15, 18, 25, and >30 years), and slope farmland with various gradients (3°, 5°, 8°, 12°, 15°, and 25°) and analyzed for heavy metals (As, Cr, Cu, Hg, Ni, and Zn). These samples were collected from Nihegou catchment of Chunhua county in the southern Loess Plateau of China. The six heavy metals demonstrated different trends with time or gradient in the three land-use types. The Cu and Zn contents of the soil were higher than the referee background values of the loessal soil, and the contents of Cr and Ni, and especially those of As and Hg, were lower. Cu was the only heavy metal that just met the Grade III Environmental Quality Standard for Soils of China, while the others reached grade I. Cu and Hg were considered contaminant factors and Hg was a moderate potential ecological risk factor in the catchment. Of the sites investigated, 89.5% fell into the category with a low degree of contamination (C d ) and rest were moderate, while all three land-use types had low potential ecological risk (RI). Changes of C d and RI were consistent with the cultivated time in the terrace farmland and terrace orchard land. Values of RI increased while C d decreased with the increasing of slope gradient in the slope farmland. Evaluating the ecological risk posed by heavy metals using more soil samples in a larger study area is necessary on the Loess Plateau of China.

Simulation of Runoff Hydrograph on Soil Surfaces with Different Microtopography Using a Travel Time Method at the Plot Scale.

In this study, a simple travel time-based runoff model was proposed to simulate a runoff hydrograph on soil surfaces with different microtopographies. Three main parameters, i.e., rainfall intensity (I), mean flow velocity (vm) and ponding time of depression (tp), were inputted into this model. The soil surface was divided into numerous grid cells, and the flow length of each grid cell (li) was then calculated from a digital elevation model (DEM). The flow velocity in each grid cell (vi) was derived from the upstream flow accumulation area using vm. The total flow travel time through each grid cell to the surface outlet was the sum of the sum of flow travel times along the flow path (i.e., the sum of li/vi) and tp. The runoff rate at the slope outlet for each respective travel time was estimated by finding the sum of the rain rate from all contributing cells for all time intervals. The results show positive agreement between the measured and predicted runoff hydrographs.

Effects of crop canopies on rain splash detachment.

Crops are one of the main factors affecting soil erosion in sloping fields. To determine the characteristics of splash erosion under crop canopies, corn, soybean, millet, and winter wheat were collected, and the relationship among splash erosion, rainfall intensity, and throughfall intensity under different crop canopies was analyzed through artificial rainfall experiments. The results showed that, the mean splash detachment rate on the ground surface was 390.12 g/m2 · h, which was lower by 67.81% than that on bare land. The inhibiting effects of crops on splash erosion increased as the crops grew, and the ability of the four crops to inhibit splash erosion was in the order of winter wheat>corn>soybeans>millet. An increase in rainfall intensity could significantly enhance the occurrence of splash erosion, but the ability of crops to inhibit splash erosion was 13% greater in cases of higher rainfall intensity. The throughfall intensity under crop canopies was positively related to the splash detachment rate, and this relationship was more significant when the rainfall intensity was 40 mm/h. Splash erosion tended to occur intensively in the central row of croplands as the crop grew, and the non-uniformity of splash erosion was substantial, with splash erosion occurring mainly between the rows and in the region directly under the leaf margin. This study has provided a theoretical basis for describing the erosion mechanisms of cropland and for assisting soil erosion prediction as well as irrigation and fertilizer management in cultivated fields.

[Energy flow characteristics of the compound agriculture-fruit farming system in Xipo Village, Shaanxi, Northwest China].

Taking the crop-fruit farming system in Xipo Village in Chunhua, Shaanxi Province as a case, the energy flow path, input and output structure, and the indices of energy cycle for the agriculture, fruit, stockbreeding and human subsystems were compared between 2008 and 2010. Results showed that during the study period the total investment to the agriculture-fruit farming system (CAF) decreased by 1.6%, while the total output increased by 56.7%, which led to a 59.4% increase of the output/input ratio. Energy output/input ratio of the agriculture, fruit, stockbreeding, human subsystems increased by 36.6%, 21.0%, 10.0% and 3.8%, respectively. The Xipo Village still needed to stabilize the agriculture, develop stockbreeding and strengthen fruit to upgrade the compound agriculture-fruit farming system.

[Coupling coordinated development of ecological-economic system in Loess Plateau].

Based on system theory, a coupling coordinated development model of ecological-economic system in Loess Plateau was established, and the evaluation criteria and basic types of the coordinated development of the ecological-economic system were proposed. The county-level coupling coordinated development of the ecological-economic system was also discussed, based on the local characteristics. The interactions between the ecological and economic systems in Loess Plateau could be divided into four stages, i.e., seriously disordered development stage, mild-disordered development stage, low-level coordinated development stage, and high level well-coordinated development stage. At each stage, there existed a cyclic process of profit and loss-antagonist-running-dominant-synchronous development. The coupling development degree of the ecological-economic system in Loess Plateau was overall at a lower level, being about 62.7% of the counties at serious disorder, 30.1% of the counties at mild disorder, and 7.1% of the counties at low but coordinated level. The coupling development degree based on the model established in this study could better reflect the current social-economic and ecological environment situations, especially the status of coordination. To fully understand the coupling of ecological-economic system and to adopt appropriate development mode would be of significance to promote the county-level coordinated development in Loess Plateau.

[Profit and loss of nutrients in crop-fruit ecosystems].

The analysis on the balance of nutrient's profit and loss in crop-fruit ecosystems of three villges in middle-south Loess Plateau showed that fertilization was the main input of nutrients, of which, chemical fertilizer occupied a larger proportion, while organic manure was relatively insufficient. The quantity of N, P and K spent in interior circulation was far less than that exported from the ecosystem. The ratio of nutrient's output and input in three test villages was differed, but all at a low level.

[Productivity of crop-fruit ecological agriculture in middle-south Loess Plateau].

With the Xipo, Feimahe and Nangou villages as test objects, the productive characteristics of crop-fruit ecological agriculture in the middle-south Loess Plateau were investigated. The results showed that the biomass productivity of a plant was different with its organs, the highest for grain or fruit, and followed by stem, leaf and root. In the crop-fruit ecological agriculture, the higher ratio the crop subsystem, the higher productivity its biomass, the lower economic productivity and lower economic value was; while the fruit subsystem was on the contrary. The productivity of livestock farming subsystem was in a lower level, which restricted the increase of the total ecosystem's productivity. Based on these results, a countermeasure of increasing the ecosystem productivity was put forward.