A multimodal deep learning approach for the prediction of cognitive decline and its effectiveness in clinical trials for Alzheimer's disease.

Alzheimer's disease is one of the most important health-care challenges in the world. For decades, numerous efforts have been made to develop therapeutics for Alzheimer's disease, but most clinical trials have failed to show significant treatment effects on slowing or halting cognitive decline. Among several challenges in such trials, one recently noticed but unsolved is biased allocation of fast and slow cognitive decliners to treatment and placebo groups during randomization caused by the large individual variation in the speed of cognitive decline. This allocation bias directly results in either over- or underestimation of the treatment effect from the outcome of the trial. In this study, we propose a stratified randomization method using the degree of cognitive decline predicted by an artificial intelligence model as a stratification index to suppress the allocation bias in randomization and evaluate its effectiveness by simulation using ADNI data set.

Visible light-assisted hydrogen generation from ammonia borane over Z-Scheme NiO-CuO heterostructures.

The design and fabrication of cheap and high-efficiency catalysts for ammonia borane (AB) hydrolysis for hydrogen production is crucial for its commercial applications. Improvement of the catalytic performance of the catalysts with the assistance of sunlight, a costless resource, is extremely attractive. Herein, we have constructed Z-scheme heterostructured VO-NiO-CuO catalysts with strong interfacial electronic interactions and abundant oxygen vacancies to enhance hydrogen production from NH3BH3 solution under visible light illumination. The as-prepared VO-NiO-CuO catalysts exhibit excellent catalytic activity with a high turnover frequency (TOF) of 35.3 molH2 molcat.-1 min-1 toward AB hydrolysis under visible light. It is demonstrated that excellent catalytic performance is highly related to the effective separation and migration of charge on the catalyst surface. As a result, dual active sites were created, making it easier for various reactants to be adsorbed and activated on the catalyst surface. Furthermore, the density functional theory (DFT) calculations indicate that the adsorption and activation of H2O occurred mainly at the Ni site of VO-NiO-CuO. When the VO-NiO-CuO is irradiated with visible light, the photogenerated electrons assembled on the conduction band were transferred to the O atom through the Ni-O bond, which made the bond length of H2O molecules longer and OH bonds more prone to breaking, thus facilitating AB hydrolysis under illumination. The findings in this work pave the way to design novel and efficient heterostructured catalysts for fast hydrogen release from NH3BH3 under visible light irradiation.

A high-generalizability machine learning framework for predicting the progression of Alzheimer's disease using limited data.

Alzheimer's disease is a neurodegenerative disease that imposes a substantial financial burden on society. A number of machine learning studies have been conducted to predict the speed of its progression, which varies widely among different individuals, for recruiting fast progressors in future clinical trials. However, because the data in this field are very limited, two problems have yet to be solved: the first is that models built on limited data tend to induce overfitting and have low generalizability, and the second is that no cross-cohort evaluations have been done. Here, to suppress the overfitting caused by limited data, we propose a hybrid machine learning framework consisting of multiple convolutional neural networks that automatically extract image features from the point of view of brain segments, which are relevant to cognitive decline according to clinical findings, and a linear support vector classifier that uses extracted image features together with non-image information to make robust final predictions. The experimental results indicate that our model achieves superior performance (accuracy: 0.88, area under the curve [AUC]: 0.95) compared with other state-of-the-art methods. Moreover, our framework demonstrates high generalizability as a result of evaluations using a completely different cohort dataset (accuracy: 0.84, AUC: 0.91) collected from a different population than that used for training.

[Clinical research of Dahuang Fuzi decoction in auxiliary treatment of severe acute pancreatitis: a multi-center observation in 206 patients].

OBJECTIVE: To observe the efficacy of Dahuang Fuzi decoction in patients with severe acute pancreatitis (SAP), and to provide valuable medical evidences for a treatment of SAP with combined traditional Chinese and western medicine. METHODS: A prospective, multi-center, randomized, controlled clinical trial was designed. Two hundred and six adult patients with SAP admitted to intensive care unit (ICU) in three tertiary university teaching hospitals in Dalian from January 2007 to February 2010 were randomly divided into two groups: soapsuds enema control group (control group, n=101) and Dahuang Fuzi decoction enema study group (study group, n=105). The levels of serum amylase, albumin (Alb), D-lactic acid, endotoxin and diamine oxidase (DAO), high-sensitive C-reactive protein (hs-CRP), immunoglobulin (IgG, IgA, IgM), complements (C3, C4), tumour necrosis factor-α (TNF-α), interleukins (IL-6, IL-8) were determined before and after treatment for 2, 4, 7 days. The bowel sound, gastrointestinal function score, the acute physiology and chronic health evaluation II (APACHEII) score and the length of mechanical ventilation (MV), the length of stay in ICU, the mortality rate and average hospital expenses within 28 days were compared. RESULTS: Compared with control group, in the study group the levels of serum amylase, DAO, D-lactic acid and endotoxin were lowered, the Alb was increased, the levels of TNF-α, IL-6, IL-8, hs-CRP were decreased, the function of body immunity was enhanced, intestinal peristalsis was enhanced, gastrointestinal function score and APACHEII score were improved, the length of MV was reduced, the length of stay in ICU was diminished, the 28-day mortality and average hospital expenses were lowered [4 days amylase (U/L): 357.35±137.54 vs. 492.95±189.42, 2 days DAO (kU/L) : 5.20±0.59 vs. 5.45±0.72, 4 days D-lactic acid (mmol/L): 3.31±0.48 vs. 4.15±0.55, 2 days endotoxin (kEU/L): 0.29±0.11 vs. 0.34±0.14, 4 days Alb (g/L): 34.75±3.56 vs. 32.53±3.44, 2 days TNF-α (ng/L): 3.08±0.45 vs. 3.36±1.11, 2 days IL-6 (ng/L): 298.54±67.82 vs. 313.56±73.91, 4 days IL-8 (ng/L): 30.48±8.56 vs. 45.16±10.81, 2 days hs-CRP (mg/L): 32.56±11.83 vs. 40.42±15.10, 4 days IgG (g/L): 7.05±2.56 vs. 9.53±2.94, 2 days IgA (mg/L): 1 600±170 vs. 1 400±140, 4 days IgM (mg/L): 1 310±280 vs. 1 650±290, 4 days C3 (g/L): 1.11±0.09 vs. 1.50±0.15, 4 days C4 (g/L) : 0.32±0.11 vs. 0.41±0.10, 2 days bowel sound (times/min): 1.26±0.45 vs. 1.15±0.41, 2 days gastrointestinal function score: 2.24±0.98 vs. 2.42±1.05, 4 days APACHEII score: 16.4±6.8 vs. 20.1±7.1, the length of MV (days): 6.5±3.1 vs. 10.1±4.6, the length of stay in ICU (days): 11.3±6.3 vs. 13.8±7.5, mortality: 8.6% vs. 16.8%, average hospital expenses (yuan): 72 thousands vs. 86 thousands, P<0.05 or P<0.01]. CONCLUSION: Dahuang Fuzi decoction may enhance the intestinal peristalsis, protect the gastrointestinal barrier function, reduce the bacteria and endotoxin translocation and the releasing of inflammation mediators, protect the function of body immunity, reduce the length of MV, the length of stay in ICU, and lower the 28-day mortality and average hospital expenses, and it can improve the prognosis of patients with SAP.

Quantitative Follow-Up Assessment of Patients with Interstitial Lung Disease by 3D-Curved High-Resolution CT Imaging Parallel to the Chest Wall.

We evaluated the progression of interstitial lung disease (ILD) by three-dimensional curved high-resolution computed tomography (3D-cHRCT) at a constant depth from the chest wall and compare the results to pulmonary function test (PFT) results on a follow-up assessment. We reviewed the patients with ILD who underwent HRCT and concurrent PFTs at least twice from April 2008 to December 2014. Forty-five patients with ILD were enrolled. 3D-cHRCT images of the lung at various depths from the chest wall were reconstructed, and total area (TA), high-attenuation area (HAA) >-500 HU, and %HAA ([HAA/TA] × 100) were calculated. The TA, HAA, and %HAA ratios (follow-up to baseline) were assessed for use in the diagnosis of physiologically progressive ILD (defined as; forced vital capacity [FVC] ratio <0.9 or %diffusing capacity of the lung for carbon monoxide [%DLCO] ratio <0.85 [follow-up to baseline]). Of all ratios obtained from 3D-cHRCT images at 5-30mm depths, the %HAA ratio at 20-mm had the largest area under the receiver operating characteristic curve (0.815, 95 % confidence interval 0.677-0.953). By univariate logistic regression analysis, TA, HAA, and %HAA ratios at 20-mm showed significant correlations with physiologically progressive ILD. 3D-cHRCT imaging performed in parallel with the chest wall offers novel quantitative parameters that are useful for following ILD.

LncRNA HOTTIP mediated DKK1 downregulation confers metastasis and invasion in colorectal cancer cells.

Recent studies highlight long non-coding RNAs (lncRNAs) as key regulators of cancer biology that contribute to carcinogenesis. The lncRNA HOXA transcript at the distal tip (HOTTIP) is involved in the development of several cancers. Previous studies demonstrated that HOTTIP could promote colorectal cancer (CRC) cell proliferation via silencing of p21 expression. However, the potential role of HOTTIP in CRC metastasis has not yet been discussed. Here, we found that HOTTIP level was significantly higher in CRC than in corresponding adjacent normal tissues, and patients with a larger tumor size, advanced pathological stage, or distant metastasis had higher HOTTIP expression. Moreover, silencing HOTTIP expression by siRNA or shRNA could inhibit CRC cell migration and invasion in vitro and in vivo, whereas HOTTIP overexpression promoted cell metastasis, as documented in the SW480 cell lines. Mechanistic analyses indicated that HOTTIP regulates CRC cell metastasis partly through the downregulation of tumor suppressor DKK1 expression. Collectively, our results suggest that tumor expression of lncRNA HOTTIP plays an important role in CRC metastasis. HOTTIP may serve as a candidate biomarker in this disease.

Prediction of Liver Fibrosis Using CT Under Respiratory Control: New Attempt Using Deformation Vectors Obtained by Non-rigid Registration Technique.

AIM: To investigate whether liver fibrosis can be predicted by quantifying the deformity of the liver obtained based on computed tomographic (CT) images scanned under respiratory control. MATERIALS AND METHODS: For dynamic CT of 47 patients, portal venous and equilibrium phases were scanned during inspiration and expiration, respectively. After rigid registration of the two images, non-rigid registration of the liver was performed, and the amount and direction of each voxel's shift during non-rigid registration was defined as the deformation vector. The correlation of each CT parameter for the obtained deformation vectors with the pathologically-proven degree of liver fibrosis was assessed using Spearman's rank correlation test. Receiver operating characteristic curve analysis was conducted for prediction of liver fibrosis. RESULTS: The standard deviation, coefficient of variance (CV) and skewness were significantly negatively correlated with the degree of liver fibrosis (p=0.030, 0.009 and 0.037, respectively). Of these measures, CV was best correlated and significantly decreased as liver fibrosis progressed (rho=-0.376). CV showed accuracies of 66.0-70.2%, and the areas under curves were 0.654-0.727 for prediction of fibrosis of grade F1 or greater, F2 or greater, F3 or greater and F4 fibrosis. CONCLUSION: The deformation vector is a potential CT parameter for evaluating liver fibrosis.

Quantitative evaluation of interstitial pneumonia using 3D-curved high-resolution CT imaging parallel to the chest wall: A pilot study.

OBJECTIVES: To quantify the imaging findings of patients with interstitial pneumonia (IP) and emphysema using three-dimensional curved high-resolution computed tomography (3D-cHRCT) at a constant depth from the chest wall, and compare the results to visual assessment of IP and each patient's diffusing capacity of the lungs for carbon monoxide (DLco). METHODS: We retrospectively reviewed the axial CT findings and pulmonary function test results of 95 patients with lung cancer (72 men and 23 women, aged 45-84 years) with or without IP, as follows: non-IP (n = 47), mild IP (n = 31), and moderate IP (n = 17). The 3D-cHRCT images of the lung at a 1-cm depth from the chest wall were reconstructed automatically using original software; total area (TA), high-attenuation area (HAA) >-500 HU, and low-attenuation area (LAA) <-950 HU were calculated on a workstation. The %HAA and %LAA were calculated as follows: [Formula: see text], and [Formula: see text]. RESULTS: The %HAA and %LAA respective values were 3.2±0.9 and 27.7±8.2, 3.9±1.2 and 27.6±5.9, and 6.9±2.2 and 25.4±8.7 in non-IP, mild IP, and moderate IP patients, respectively. There were significant differences in %HAA between the 3 groups of patients (P<0.001), but no differences in %LAA (P = 0.558). Multiple linear regression analysis revealed that %HAA and %LAA were negatively correlated with predicted DLco (standard partial regression coefficient [b*] = -0.453, P<0.001; b* = -0.447, P<0.001, respectively). CONCLUSIONS: The %HAA and %LAA values computed using 3D-cHRCT were significantly correlated with DLco and may be important quantitative parameters for both IP and emphysema.

Thoracic Temporal Subtraction Three Dimensional Computed Tomography (3D-CT): Screening for Vertebral Metastases of Primary Lung Cancers.

PURPOSE: We developed an original, computer-aided diagnosis (CAD) software that subtracts the initial thoracic vertebral three-dimensional computed tomography (3D-CT) image from the follow-up 3D-CT image. The aim of this study was to investigate the efficacy of this CAD software during screening for vertebral metastases on follow-up CT images of primary lung cancer patients. MATERIALS AND METHODS: The interpretation experiment included 30 sets of follow-up CT scans in primary lung cancer patients and was performed by two readers (readers A and B), who each had 2.5 years' experience reading CT images. In 395 vertebrae from C6 to L3, 46 vertebral metastases were identified as follows: osteolytic metastases (n = 17), osteoblastic metastases (n = 14), combined osteolytic and osteoblastic metastases (n = 6), and pathological fractures (n = 9). Thirty-six lesions were in the anterior component (vertebral body), and 10 lesions were in the posterior component (vertebral arch, transverse process, and spinous process). The area under the curve (AUC) by receiver operating characteristic (ROC) curve analysis and the sensitivity and specificity for detecting vertebral metastases were compared with and without CAD for each observer. RESULTS: Reader A detected 47 abnormalities on CT images without CAD, and 33 of them were true-positive metastatic lesions. Using CAD, reader A detected 57 abnormalities, and 38 were true positives. The sensitivity increased from 0.717 to 0.826, and on ROC curve analysis, AUC with CAD was significantly higher than that without CAD (0.849 vs. 0.902, p = 0.021). Reader B detected 40 abnormalities on CT images without CAD, and 36 of them were true-positive metastatic lesions. Using CAD, reader B detected 44 abnormalities, and 39 were true positives. The sensitivity increased from 0.783 to 0.848, and AUC with CAD was nonsignificantly higher than that without CAD (0.889 vs. 0.910, p = 0.341). Both readers detected more osteolytic and osteoblastic metastases with CAD than without CAD. CONCLUSION: Our temporal 3D-CT subtraction CAD software easily detected vertebral metastases on the follow-up CT images of lung cancer patients regardless of the osteolytic or osteoblastic nature of the lesions.

Effect of probiotics on the incidence of ventilator-associated pneumonia in critically ill patients: a randomized controlled multicenter trial.

PURPOSE: To evaluate the potential preventive effect of probiotics on ventilator-associated pneumonia (VAP). METHODS: This was an open-label, randomized, controlled multicenter trial involving 235 critically ill adult patients who were expected to receive mechanical ventilation for ≥48 h. The patients were randomized to receive (1) a probiotics capsule containing live Bacillus subtilis and Enterococcus faecalis (Medilac-S) 0.5 g three times daily through a nasogastric feeding tube plus standard preventive strategies or (2) standard preventive strategies alone, for a maximum of 14 days. The development of VAP was evaluated daily, and throat swabs and gastric aspirate were cultured at baseline and once or twice weekly thereafter. RESULTS: The incidence of microbiologically confirmed VAP in the probiotics group was significantly lower than that in the control patients (36.4 vs. 50.4 %, respectively; P = 0.031). The mean time to develop VAP was significantly longer in the probiotics group than in the control group (10.4 vs. 7.5 days, respectively; P = 0.022). The proportion of patients with acquisition of gastric colonization of potentially pathogenic microorganisms (PPMOs) was lower in the probiotics group (24 %) than the control group (44 %) (P = 0.004). However, the proportion of patients with eradication PPMO colonization on both sites of the oropharynx and stomach were not significantly different between the two groups. The administration of probiotics did not result in any improvement in the incidence of clinically suspected VAP, antimicrobial consumption, duration of mechanical ventilation, mortality and length of hospital stay. CONCLUSION: Therapy with the probiotic bacteria B. Subtilis and E. faecalis are an effective and safe means for preventing VAP and the acquisition of PPMO colonization in the stomach.

Coronary lumen and plaque segmentation from CTA using higher-order shape prior.

We propose a novel segmentation method based on multi-label graph cuts utilizing higher-order potentials to impose shape priors. Each higher-order potential is defined with respect to a candidate shape, and takes a low value if and only if most of the voxels inside the shape are foreground and most of those outside are background. We apply this technique to coronary lumen and plaque segmentation in CT angiography, exploiting the prior knowledge that the vessel walls tend to be tubular, whereas calcified plaques are more likely globular. We use the Hessian analysis to detect the candidate shapes and introduce corresponding higher-order terms into the energy. Since each higher-order term has any effect only when its highly specific condition is met, we can add many of them at possible locations and sizes without severe side effects. We show the effectiveness of the method by testing it on the standardized evaluation framework presented at MICCAI segmentation challenge 2012. The method achieved values comparable to the best in each of the sensitivity and positive predictive value, placing it at the top in average rank.

Pulmonary lobar volumetry using novel volumetric computer-aided diagnosis and computed tomography.

OBJECTIVES: To compare the accuracy of pulmonary lobar volumetry using the conventional number of segments method and novel volumetric computer-aided diagnosis using 3D computed tomography images. METHODS: We acquired 50 consecutive preoperative 3D computed tomography examinations for lung tumours reconstructed at 1-mm slice thicknesses. We calculated the lobar volume and the emphysematous lobar volume < -950 HU of each lobe using (i) the slice-by-slice method (reference standard), (ii) number of segments method, and (iii) semi-automatic and (iv) automatic computer-aided diagnosis. We determined Pearson correlation coefficients between the reference standard and the three other methods for lobar volumes and emphysematous lobar volumes. We also compared the relative errors among the three measurement methods. RESULTS: Both semi-automatic and automatic computer-aided diagnosis results were more strongly correlated with the reference standard than the number of segments method. The correlation coefficients for automatic computer-aided diagnosis were slightly lower than those for semi-automatic computer-aided diagnosis because there was one outlier among 50 cases (2%) in the right upper lobe and two outliers among 50 cases (4%) in the other lobes. The number of segments method relative error was significantly greater than those for semi-automatic and automatic computer-aided diagnosis (P < 0.001). The computational time for automatic computer-aided diagnosis was 1/2 to 2/3 than that of semi-automatic computer-aided diagnosis. CONCLUSIONS: A novel lobar volumetry computer-aided diagnosis system could more precisely measure lobar volumes than the conventional number of segments method. Because semi-automatic computer-aided diagnosis and automatic computer-aided diagnosis were complementary, in clinical use, it would be more practical to first measure volumes by automatic computer-aided diagnosis, and then use semi-automatic measurements if automatic computer-aided diagnosis failed.