Development and Validation of a Deep Learning-Based Automatic Brain Segmentation and Classification Algorithm for Alzheimer Disease Using 3D T1-Weighted Volumetric Images.

BACKGROUND AND PURPOSE: Limited evidence has suggested that a deep learning automatic brain segmentation and classification method, based on T1-weighted brain MR images, can predict Alzheimer disease. Our aim was to develop and validate a deep learning-based automatic brain segmentation and classification algorithm for the diagnosis of Alzheimer disease using 3D T1-weighted brain MR images. MATERIALS AND METHODS: A deep learning-based algorithm was developed using a dataset of T1-weighted brain MR images in consecutive patients with Alzheimer disease and mild cognitive impairment. We developed a 2-step algorithm using a convolutional neural network to perform brain parcellation followed by 3 classifier techniques including XGBoost for disease prediction. All classification experiments were performed using 5-fold cross-validation. The diagnostic performance of the XGBoost method was compared with logistic regression and a linear Support Vector Machine by calculating their areas under the curve for differentiating Alzheimer disease from mild cognitive impairment and mild cognitive impairment from healthy controls. RESULTS: In a total of 4 datasets, 1099, 212, 711, and 705 eligible patients were included. Compared with the linear Support Vector Machine and logistic regression, XGBoost significantly improved the prediction of Alzheimer disease (P < .001). In terms of differentiating Alzheimer disease from mild cognitive impairment, the 3 algorithms resulted in areas under the curve of 0.758-0.825. XGBoost had a sensitivity of 68% and a specificity of 70%. In terms of differentiating mild cognitive impairment from the healthy control group, the 3 algorithms resulted in areas under the curve of 0.668-0.870. XGBoost had a sensitivity of 79% and a specificity of 80%. CONCLUSIONS: The deep learning-based automatic brain segmentation and classification algorithm allowed an accurate diagnosis of Alzheimer disease using T1-weighted brain MR images. The widespread availability of T1-weighted brain MR imaging suggests that this algorithm is a promising and widely applicable method for predicting Alzheimer disease.

Efficacy of HVT-IBD vector vaccine compared to attenuated live vaccine using in-ovo vaccination against a Korean very virulent IBDV in commercial broiler chickens.

The production performance, efficacy, and safety of two types of vaccines for infectious bursal disease virus (IBDV) were compared with in-ovo vaccination of Cobb 500 broiler chickens for gross and microscopic examination of the bursa of Fabricius, bursa/body weight (b/B) ratio, flow cytometry, and serologic response to Newcastle disease virus (NDV) vaccination. One vaccine was a recombinant HVT-IBD vector vaccine (HVT as for herpesvirus of turkeys) and the other was an intermediate plus live IBDV vaccine. A significant difference was detected at 21 d. Eight of 10 chickens that received the IBDV live vaccine had severe bursal lesions and a relatively low b/B ratio of 0.95, and an inhibited NDV vaccine response. On the other hand, the HVT-IBD vector vaccine resulted in mild bursal lesions and a b/B ratio of 1.89. Therefore, the live vaccine had lower safety than that of the HVT-IBD vector vaccine. To determine the protective efficacy, chickens were intraocularly challenged at 24 d. Eight of 10 chickens in the IBDV live vaccination group showed gross and histological lesions characterized by hemorrhage, cyst formation, lymphocytic depletion, and a decreased b/B ratio. In contrast, the HVT-IBD vector vaccinated chickens showed mild gross and histological lesions in three of 10 chickens with a b/B ratio of 1.36, which was similar to that of the unchallenged controls. Vaccinated chickens showed a significant increase in IBDV antibody titers, regardless of the type of vaccine used. In addition, significantly better broiler flock performance was observed with the HVT-IBD vector vaccine compared to that of the live vaccine. Our results revealed that the HVT-IBD vector vaccine could be used as an alternative vaccine to increase efficacy, and to have an improved safety profile compared with the IBDV live vaccine using in-ovo vaccination against the Korean very virulent IBDV in commercial broiler chickens.

Effect of binding to carbon black on the dynamics of 1,4-polybutadiene.

The nature of the interactions of polymers at the surface of nanoparticles is crucial to understanding the dynamics and their relation to mechanical properties. The effect of binding (both chemical attachment and physical adsorption) on the local and global dynamics of chain molecules remains a controversial subject. Using neutron scattering and dynamic mechanical spectroscopies, we measured the slow conformational and terminal relaxations, as well as the fast local dynamics, of 1,4-polybutadiene (PBD) containing carbon black (CB) particles. We observed a substantial decrease in the flexibility of bound segments at temperatures through the glass transition temperature, T(g). The longer range motions of the PBD become more suppressed and cooperative as temperature decreases, while the relaxation time of the fast local dynamics is little affected by the CB particles. The mobile fraction of PBD is less sensitive to temperature when bound. Mechanical spectroscopy indicates that both the local segmental dynamics and the global chain modes are slowed by the filler. These results are consistent with transient structural arrest of the slow dynamics of atoms adjacent to the particles.

Detection of Merkel cell polyomavirus in Merkel cell carcinomas and small cell carcinomas by PCR and immunohistochemistry.

Recently, the clonal integration of a new human polyomavirus (Merkel cell polyomavirus or MCPyV) has been reported in Merkel cell carcinoma (MCC). In order to investigate the presence of MCPyV in small cell carcinomas (SCCs) and small round cell tumors (SRCTs), we collected formalin-fixed paraffin-embedded tissue specimens including 14 MCCs, 24 SCCs, 7 Ewing sarcoma/primitive neuroectodermal tumors (ES/PNETs) and 5 neuroblastomas. We also collected specimens of other cancers including 12 malignant melanomas, 10 breast, 10 ovarian and 20 gastric cancers. We used 3 primer sets for which the sequences were previously published (LT1, LT3, and VP1) and 3 newly designed primer sets (LT1-1, LT1-1a, and LT3a). Quantitative real-time PCR was also performed with the LTq primer set. Nested PCR using the LT3a primer set detected more cases of MCPyV infection in MCC. In total, 12 of 14 (85.7%) MCC cases were positive for MCPyV by PCR, which was consistent with published data. Some SCC specimens were also positive for MCPyV (37.5%) by PCR. PCR products from MCC and SCC cases showed premature truncation and frameshift mutation. Furthermore, one case of ES/PNET and one gastric carcinoma showed MCPyV DNA. However, MCPyV DNA and transcript were only detected in MCCs with quantitative real-time PCR analysis. In addition, 11 of 13 (84.6%) MCC cases and 6 of 23 (26.1%) SCC cases showed immunoreactivity with monoclonal antibodies against MCPyV large T-antigen. Considering both PCR and IHC results, MCPyV was detected in all MCCs tested. The presence of MCPyV in all MCC cases tested and in some SCC cases suggests that MCPyV may be involved in the malignant transformation.

Dynamics of biological macromolecules: not a simple slaving by hydration water.

We studied the dynamics of hydrated tRNA using neutron and dielectric spectroscopy techniques. A comparison of our results with earlier data reveals that the dynamics of hydrated tRNA is slower and varies more strongly with temperature than the dynamics of hydrated proteins. At the same time, tRNA appears to have faster dynamics than DNA. We demonstrate that a similar difference appears in the dynamics of hydration water for these biomolecules. The results and analysis contradict the traditional view of slaved dynamics, which assumes that the dynamics of biological macromolecules just follows the dynamics of hydration water. Our results demonstrate that the dynamics of biological macromolecules and their hydration water depends strongly on the chemical and three-dimensional structures of the biomolecules. We conclude that the whole concept of slaving dynamics should be reconsidered, and that the mutual influence of biomolecules and their hydration water must be taken into account.

Dynamics of tRNA at different levels of hydration.

The influence of hydration on the nanosecond timescale dynamics of tRNA is investigated using neutron scattering spectroscopy. Unlike protein dynamics, the dynamics of tRNA is not affected by methyl group rotation. This allows for a simpler analysis of the influence of hydration on the conformational motions in RNA. We find that hydration affects the dynamics of tRNA significantly more than that of lysozyme. Both the characteristic length scale and the timescale of the conformational motions in tRNA depend strongly on hydration. Even the characteristic temperature of the so-called "dynamical transition" appears to be hydration-dependent in tRNA. The amplitude of the conformational motions in fully hydrated tRNA is almost twice as large as in hydrated lysozyme. We ascribe these differences to a more open and flexible structure of hydrated RNA, and to a larger fraction and different nature of hydrophilic sites. The latter leads to a higher density of water that makes the biomolecule more flexible. All-atom molecular-dynamics simulations are used to show that the extent of hydration is greater in tRNA than in lysozyme. We propose that water acts as a "lubricant" in facilitating enhanced motion in solvated RNA molecules.

The pi40 gene for durable resistance to rice blast and molecular analysis of pi40-advanced backcross breeding lines.

Rice blast severely reduces production in both irrigated and water-stressed upland ecosystems of tropical and temperate countries. Nearly 50 blast resistance genes have been identified and some of those are incorporated into several rice cultivars. However, most of the resistance genes break down in a few years because of their race specificity and the rapid change in pathogenicity of the blast fungus (Magnaporthe grisea). The objective of this study was to analyze advanced backcross breeding lines (ABL) possessing the gene Pi40 for durable rice blast resistance. In all, 4 resistant genotypes, 4 japonica cultivars, and 10 monogenic differential rice genotypes with some known resistance genes were bioassayed in the greenhouse using seven sequential plantings and 29 virulent M. grisea isolates of Korea. The genotypes with the Pi40 gene had <3% diseased leaf area, which was significantly below the disease threshold level of 40% considered for durable blast resistance. Moreover, the genotypes with the Pi40 gene expressed compatibility with only two to three virulent M. grisea isolates supporting durability of resistance, in contrast to susceptible cultivars with >50% diseased leaf area and 10 compatible isolates. Of the 10 known resistance genes tested, Piz-t, Piz-5, and Pi9 showed differential reactions to the pathogen isolates in seven plantings. Genotyping of the ABL with 260 simple sequence repeat (SSR) markers revealed rapid conversion toward recurrent parent genotypes with fewer donor chromosomal segments (5.3 to 14.5%). Our study based on a sequential testing and background selection of breeding lines with the resistance gene Pi40 provided valuable information for durable blast resistance breeding in rice.

The relationship of pain and health-related quality of life in Korean patients with Parkinson's disease.

BACKGROUND: Parkinson's disease (PD) is a chronic progressive neurodegenerative disorder. Increasing attention has been focused on the pain and health-related quality of life (HrQOL) in patients with PD. Objective - To evaluate the relationship between pain and the HrQOL in patients with PD. METHODS: Eighty-two patients with PD were included and classified into two groups according to the presence of pain. The Hoehn and Yahr scale, the Unified Parkinson's Disease Rating Scale (UPDRS), the Modified Somatic Perception Questionnaire (MSPQ), the Zung Depression Inventory--Self-rating Depression Scale (SDS), the Visual Analogue Scale and the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36) were administered. The factors influencing the pain, HrQOL and parkinsonian manifestations were evaluated. RESULTS: The PD with pain group had higher UPDRS part III scores, lower SF-36 scores, higher SDS scores and higher MSPQ scores than the PD without pain group. The presence of pain, high Hoehn and Yahr stage, advanced age and somatic perception were the factors that had a negative effect on the physical component of the HrQOL. Depression and somatic perception were the most important predictive factors for the mental component of the HrQOL. Depression and poor parkinsonian motor abilities were the leading factors contributing to pain. CONCLUSION: Pain and depression were major detrimental factors affecting the physical and mental aspects of the HrQOL respectively. Therefore, the treatment of pain and depression can be important to improve the HrQOL.

The origin of the dynamic transition in proteins.

Despite extensive efforts in experimental and computational studies, the microscopic understanding of dynamics of biological macromolecules remains a great challenge. It is known that hydrated proteins, DNA and RNA, exhibit a so-called "dynamic transition." It appears as a sharp rise of their mean-squared atomic displacements r2 at temperatures above 200-230 K. Even after a long history of studies, this sudden activation of biomolecular dynamics remains a puzzle and many contradicting models have been proposed. By combining neutron and dielectric spectroscopy data, we were able to follow protein dynamics over an extremely broad frequency range. Our results show that there is no sudden change in the dynamics of the protein at temperatures around approximately 200-230 K. The protein's relaxation time exhibits a smooth temperature variation over the temperature range of 180-300 K. Thus the experimentally observed sharp rise in r2 is just a result of the protein's structural relaxation reaching the limit of the experimental frequency window. The microscopic mechanism of the protein's structural relaxation remains unclear.

Influence of hydration on the dynamics of lysozyme.

Quasielastic neutron and light-scattering techniques along with molecular dynamics simulations were employed to study the influence of hydration on the internal dynamics of lysozyme. We identified three major relaxation processes that contribute to the observed dynamics in the picosecond to nanosecond time range: 1), fluctuations of methyl groups; 2), fast picosecond relaxation; and 3), a slow relaxation process. A low-temperature onset of anharmonicity at T approximately 100 K is ascribed to methyl-group dynamics that is not sensitive to hydration level. The increase of hydration level seems to first increase the fast relaxation process and then activate the slow relaxation process at h approximately 0.2. The quasielastic scattering intensity associated with the slow process increases sharply with an increase of hydration to above h approximately 0.2. Activation of the slow process is responsible for the dynamical transition at T approximately 200 K. The dependence of the slow process on hydration correlates with the hydration dependence of the enzymatic activity of lysozyme, whereas the dependence of the fast process seems to correlate with the hydration dependence of hydrogen exchange of lysozyme.

Onsets of anharmonicity in protein dynamics.

Two onsets of anharmonicity are observed in the dynamics of the protein lysozyme. One at T approximately 100 K appears in all samples regardless of hydration level and is consistent with methyl group rotation. The second, the well-known dynamical transition at T approximately 200-230 K, is only observed at a hydration level h greater than approximately 0.2 and is ascribed to the activation of an additional relaxation process. Its variation with hydration correlates well with variations of catalytic activity suggesting that the relaxation process is directly related to the activation of modes required for protein function.

Interacting regulatory networks in the facultative photosynthetic bacterium, Rhodobacter sphaeroides 2.4.1.

Regulation of photosynthetic membrane synthesis in Rhodobacter sphaeroides 2.4.1 is dependent on the interactions of numerous regulatory elements, with two of the most important being the cbb(3) terminal oxidase and the PrrBAC two-component regulatory system. Here, we reveal that the cbb(3) terminal oxidase possesses extensive, additional regulatory activities under anaerobic conditions, and that the PrrBAC system is further involved in the regulation of the expression of more than 20% of the R. sphaeroides genome under anaerobic conditions, extending well beyond functions related to redox gene expression.

Protein and solvent dynamics: how strongly are they coupled?

Analysis of Raman and neutron scattering spectra of lysozyme demonstrates that the protein dynamics follow the dynamics of the solvents glycerol and trehalose over the entire temperature range measured 100-350 K. The protein's fast conformational fluctuations and low-frequency vibrations and their temperature variations are very sensitive to behavior of the solvents. Our results give insight into previous counterintuitive observations that protein relaxation is stronger in solid trehalose than in liquid glycerol. They also provide insight into the effectiveness of glycerol as a biological cryopreservant.

Comparative study of angiostatic and anti-invasive gene expressions as prognostic factors in gastric cancer.

Genes involving angiogenesis and metastasis play an important role in the progression and infiltration of cancer. We examined the expressions of various angiostatic and potential invasion/metastasis suppressor genes through RT-PCR analyses in 32 gastric cancer specimens with or without distant metastasis. The expressions of the invasion/metastasis suppressor, nm23 and E-cadherin increased much more in the cancer tissue (CT) and metastatic lymph node (MLN) than in the extraneoplastic mucosa (EM) and non-metastatic lymph node (NLN), respectively. The expressions of the angiostatic factor, angiopoietin 2 and thrombospondin 2 increased in the CT and MLN as compared with the EM and NLN, respectively. The newly cloned angiostatic factor, brain-specific angiogenesis inhibitor 1 (BAI1) decreased much more in the CT and MLN than the EM and NLN, respectively. However, BAI1 increased in the CT compared with the EM among the patients with poor prognosis and distant metastasis, such as liver or peritoneum. The expressions of the invasive factor, matrix metalloproteinase-2 and its suppressor, tissue inhibitor metalloproteinase-2 (TIMP-2) increased in the CM as compared with the EM, but the increased expression pattern of these genes in the CT became blunted among the patients with good prognosis. Our results indicate that BAI1 and TIMP-2 expressions in the extraneoplastic mucosa and non-metastatic lymph nodes were not suppressed in the patients with good prognosis, but increased expressions of angiopoietin 2, thrombospondin 2, TIMP-2, nm23 and E-cadherin in the tumor tissue did not lead to a long survival after operation. It is suggested that the extent of BAI1 and TIMP-2 expression in the gastric mucosa may be an important prognostic factor for predicting survival in gastric cancer.