Using Manifold Learning for Atlas Selection in Multi-Atlas Segmentation

Multi-atlas segmentation has been widely used to segment various anatomical structures. The success of this technique partly relies on the selection of atlases that are best mapped to a new target image after registration. Recently, manifold learning has been proposed as a method for atlas selection. Each manifold learning technique seeks to optimize a unique objective function. Therefore, different techniques produce different embeddings even when applied to the same data set. Previous studies used a single technique in their method and gave no reason for the choice of the manifold learning technique employed nor the theoretical grounds for the choice of the manifold parameters. In this study, we compare side-by-side the results given by 3 manifold learning techniques (Isomap, Laplacian Eigenmaps and Locally Linear Embedding) on the same data set. We assess the ability of those 3 different techniques to select the best atlases to combine in the framework of multi-atlas segmentation. First, a leave-one-out experiment is used to optimize our method on a set of 110 manually segmented atlases of hippocampi and find the manifold learning technique and associated manifold parameters that give the best segmentation accuracy. Then, the optimal parameters are used to automatically segment 30 subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). For our dataset, the selection of atlases with Locally Linear Embedding gives the best results. Our findings show that selection of atlases with manifold learning leads to segmentation accuracy close to or significantly higher than the state-of-the-art method and that accuracy can be increased by fine tuning the manifold learning process.     

Localization of Microfibrillar-Associated Protein 4 (MFAP4) in Human Tissues: Clinical Evaluation of Serum MFAP4 and Its Association with Various Cardiovascular Conditions

Microfibrillar-associated protein 4 (MFAP4) is located in the extracellular matrix (ECM). We sought to identify tissues with high levels of MFAP4 mRNA and MFAP4 protein expression. Moreover, we aimed to evaluate the significance of MFAP4 as a marker of cardiovascular disease (CVD) and to correlate MFAP4 with other known ECM markers, such as fibulin-1, osteoprotegerin (OPG), and osteopontin (OPN). Quantitative real-time PCR demonstrated that MFAP4 mRNA was more highly expressed in the heart, lung, and intestine than in other elastic tissues. Immunohistochemical studies demonstrated high levels of MFAP4 protein mainly at sites rich in elastic fibers and within blood vessels in all tissues investigated. The AlphaLISA technique was used to determine serum MFAP4 levels in a clinical cohort of 172 patients consisting of 5 matched groups with varying degrees of CVD: 1: patients with ST elevation myocardial infarction (STEMI), 2: patients with non-STEMI, 3: patients destined for vascular surgery because of various atherosclerotic diseases (stable atherosclerotic disease), 4: apparently healthy individuals with documented coronary artery calcification (CAC-positive), and 5: apparently healthy individuals without signs of coronary artery calcification (CAC-negative). Serum MFAP4 levels were significantly lower in patients with stable atherosclerotic disease than CAC-negative individuals (p<0.05). Furthermore, lower serum MFAP4 levels were present in patients with stable atherosclerotic disease compared with STEMI and non-STEMI patients (p<0.05). In patients with stable atherosclerotic disease, positive correlations between MFAP4 and both fibulin-1 (ρ = 0.50; p = 0.0244) and OPG (ρ = 0.62; p = 0.0014) were found. Together, these results indicate that MFAP4 is mainly located in elastic fibers and is highly expressed in blood vessels. The present study suggests that serum MFAP4 varies in groups of patients with different cardiovascular conditions. Further studies are warranted to describe the role of serum MFAP4 as a biomarker of stable atherosclerotic disease.     

The Wilms Tumor Gene,Wt1, Is Critical for Mouse Spermatogenesis via Regulation of Sertoli Cell Polarity and Is Associated with Non-Obstructive Azoospermia in Humans

Azoospermia is one of the major reproductive disorders which cause male infertility in humans; however, the etiology of this disease is largely unknown. In the present study, six missense mutations of WT1 gene were detected in 529 human patients with non-obstructive azoospermia (NOA), indicating a strong association between WT1 mutation and NOA. The Wilms tumor gene, Wt1, is specifically expressed in Sertoli cells (SCs) which support spermatogenesis. To examine the functions of this gene in spermatogenesis, Wt1 was deleted in adult testis using Wt1^flox and Cre-ER^TM mice strains. We found that inactivation of Wt1 resulted in massive germ cell death and only SCs were present in most of the seminiferous tubules which was very similar to NOA in humans. In investigating the potential mechanism for this, histological studies revealed that the blood–testis barrier (BTB) was disrupted in Wt1 deficient testes. In vitro studies demonstrated that Wt1 was essential for cell polarity maintenance in SCs. Further studies found that the expression of cell polarity associated genes (Par6b and E-cadherin) and Wnt signaling genes (Wnt4, Wnt11) were downregulated in Wt1 deficient SCs, and that the expression of Par6b and E-cadherin was regulated by Wnt4. Our findings suggest that Wt1 is important in spermatogenesis by regulating the polarity of SCs via Wnt signaling pathway and that WT1 mutation is one of the genetic causes of NOA in humans.     

Coal Fly Ash Impairs Airway Antimicrobial Peptides and Increases Bacterial Growth

Air pollution is a risk factor for respiratory infections, and one of its main components is particulate matter (PM), which is comprised of a number of particles that contain iron, such as coal fly ash (CFA). Since free iron concentrations are extremely low in airway surface liquid (ASL), we hypothesize that CFA impairs antimicrobial peptides (AMP) function and can be a source of iron to bacteria. We tested this hypothesis in vivo by instilling mice with Pseudomonas aeruginosa (PA01) and CFA and determine the percentage of bacterial clearance. In addition, we tested bacterial clearance in cell culture by exposing primary human airway epithelial cells to PA01 and CFA and determining the AMP activity and bacterial growth in vitro. We report that CFA is a bioavailable source of iron for bacteria. We show that CFA interferes with bacterial clearance in vivo and in primary human airway epithelial cultures. Also, we demonstrate that CFA inhibits AMP activity in vitro, which we propose as a mechanism of our cell culture and in vivo results. Furthermore, PA01 uses CFA as an iron source with a direct correlation between CFA iron dissolution and bacterial growth. CFA concentrations used are very relevant to human daily exposures, thus posing a potential public health risk for susceptible subjects. Although CFA provides a source of bioavailable iron for bacteria, not all CFA particles have the same biological effects, and their propensity for iron dissolution is an important factor. CFA impairs lung innate immune mechanisms of bacterial clearance, specifically AMP activity. We expect that identifying the PM mechanisms of respiratory infections will translate into public health policies aimed at controlling, not only concentration of PM exposure, but physicochemical characteristics that will potentially cause respiratory infections in susceptible individuals and populations.     

Changing Emergence of Shigella Sero-Groups in Bangladesh: Observation from Four Different Diarrheal Disease Hospitals

Background

Shigellosis continues to be a public health challenge for developing countries, including Bangladesh. The aim of the study is to demonstrate recent changes in Shigella sero-groups and their geographical diversity.

Methods

Data were extracted from data archive of four diarrheal disease surveillance systems. A 2% sub sample from urban Dhaka Hospital (2008–2011; n = 10,650), and 10% from urban Mirpur Treatment Centre (2009–2011; n = 3,585), were enrolled systematically; whereas, all patients coming from the Health and Demographic Surveillance System area in rural Matlab (2008–2011; n = 6,399) and rural Mirzapur (2010–2011; n = 2,812) were included irrespective of age, sex, and disease severity. A fresh stool specimen was collected for identification of Shigella spp. Of them, 315 (3%) were positive for Shigella in Dhaka, 490 (8%) from Matlab, 109 (3%) from Mirpur and 369 (13%) from Mirzapur and considered as analyzable sample size.

Results

Among all Shigella isolates regardless of age, significant decreases in percentage of S. flexneri over time was observed in Mirpur (55→29%; p value of χ²-for trend = 0.019) and Mirzapur (59→47%; p = 0.025). A non-significant decrease was also seen in Dhaka (58→48%), while in Matlab there was a non-significant increase (73→81%). Similar patterns were observed among under-5 children at all sites. Emergence of S. sonnei was found in Dhaka (8→25%; p<0.001) and Mirpur (10→33%; p = 0.015), whereas it decreased in Mirzapur (32→23%; p = 0.056). The emergence of S. boydii was seen in all ages in Mirzapur [(3→28%; p<0.001); (3→27%; p<0.001)]. On the other hand, we saw non-significant percent reductions in S. boydii in Dhaka [overall (25→16%); under-5 (16→9%)]. Decreasing rates of Shigella dysenteriae were observed in Matlab, Mirpur and Mirzapur; whereas, in Dhaka it remained unchanged.

Conclusion and Significance

Emergence of S. sonnei and S. boydii as important infectious diarrhea etiologies and variations in geographical diversity underscore the need for monitoring, with possible implications for vaccine development.     

Moderate Alcohol Use and Cardiovascular Disease from Mendelian Randomization

Background

Observational studies show moderate alcohol use negatively associated with ischemic heart disease (IHD) and cardiovascular disease (CVD). However, healthier attributes among moderate users compared to never users may confound the apparent association. A potentially less biased way to examine the association is Mendelian randomization, using alcohol metabolizing genes which influence alcohol use.

Methods

We used instrumental variable analysis with aldehyde dehydrogenase 2 (ALDH2) genotypes (AA/GA/GG) as instrumental variables for alcohol use to examine the association of alcohol use (10 g ethanol/day) with CVD risk factors (blood pressure, lipids and glucose) and morbidity (self-reported IHD and CVD) among men in the Guangzhou Biobank Cohort Study.

Results

ALDH2 genotypes were a credible instrument for alcohol use (F-statistic 74.6). Alcohol was positively associated with HDL-cholesterol (0.05 mmol/L per alcohol unit, 95% confidence interval (CI) 0.02 to 0.08) and diastolic blood pressure (1.15 mmHg, 95% CI 0.23 to 2.07) but not with systolic blood pressure (1.00 mmHg, 95% CI -0.74 to 2.74), LDL-cholesterol (0.03 mmol/L, 95% CI -0.03 to 0.08), log transformed triglycerides (0.03 mmol/L, 95% CI -0.01 to 0.08) or log transformed fasting glucose (0.01 mmol/L, 95% CI -0.006 to 0.03), self-reported CVD (odds ratio (OR) 0.98, 95% CI 0.76 to 1.27) or self-reported IHD (OR 1.10, 95% CI 0.83 to 1.45).

Conclusion

Low to moderate alcohol use among men had the expected effects on most CVD risk factors but not fasting glucose. Larger studies are needed to confirm the null associations with IHD, CVD and fasting glucose.     

Association of polymorphisms in the Chr18q11.2 locus with tuberculosis in Chinese population

A GWAS study has reported that two single nucleotide polymorphisms (SNPs) were associated with predisposition to tuberculosis (TB) in African populations. These two loci represented the long-waited GWAS hits for TB susceptibility. To determine whether these two SNPs are associated with TB in Chinese population, we attempted an replication in a cohort of over one thousand Chinese TB patients and 1,280 healthy controls using melting temperature shift allele-specific genotyping analysis. We found that only SNP rs4331426 was significantly associated with TB in Chinese population (p = 0.011). However, the effect was opposite. The G allele of the SNP in Chinese population is a protective allele (OR = 0.62, 95 % CI 0.44–0.87), while it was the risk allele for African population (OR = 1.19, 95 % CI 1.12–1.26). No significance was found for SNP rs2335704. The results provided an independent support for a role in susceptibility to TB for SNP rs4331426. However, it also indicated that direct predisposition element to TB and the association effects may vary across ethnic groups.