Association of Genetic Polymorphisms with Abdominal Aortic Aneurysm in the Processes of Apoptosis, Inflammation, and Cholesterol Metabolism.

Background and Objectives: This study aims to identify the minor allele of the single nucleotide polymorphisms (SNPs) DAB2IP rs7025486, IL6R rs2228145, CDKN2BAS rs10757278, LPA rs3798220, LRP1 rs1466535, and SORT1 rs599839 in order to assess the risk of abdominal aortic aneurysm (AAA) formation and define the linkage among these SNPs. Materials and Methods: A case-control study with AAA patients (AAA group) and non-AAA controls (control group) was carried out in a study population. DNA was isolated from whole blood samples; the SNPs were amplified using PCR and sequenced. Results: In the AAA group of 148 patients, 87.2% of the patients were male, 64.2% had a history of smoking, and 18.2% had relatives with AAA. The mean ± SD of age, BMI, and aneurysmal diameter in the AAA group were 74.8 ± 8.3 years, 27.6 ± 4.6 kg/m2, and 56.2 ± 11.8 mm, respectively. In comparison with 50 non-AAA patients, there was a significantly elevated presence of the SNPs DAB2IP rs7025486[A], CDKN2BAS rs10757278[G], and SORT1 rs599839[G] in the AAA group (p-values 0.040, 0.024, 0.035, respectively), while LPA rs3798220[C] was significantly higher in the control group (p = 0.049). A haplotype investigation showed that the SNPs DAB2IP, CDKN2BAS, and IL6R rs2228145[C] were significantly elevated in the AAA group (p = 0.037, 0.037, and 0.046) with minor allele frequencies (MAF) of 25.5%, 10.6%, and 15.4%, respectively. Only DAB2IP and CDKN2BAS showed significantly higher occurrences of a mutation (p = 0.028 and 0.047). Except for LPA, all SNPs were associated with a large aortic diameter in AAA (p < 0.001). Linkage disequilibrium detection showed that LPA to DAB2IP, to IL6R, to CDKN2BAS, and to LRP1 rs1466535[T] had D' values of 70.9%, 80.4%, 100%, and 100%, respectively. IL6R to LRP1 and to SORT1 had values for the coefficient of determination (r2) of 3.9% and 2.2%, respectively. Conclusions: In the investigated study population, the SNPs CDKN2BAS rs10757278, LPA rs3798220, SORT1 rs599839, DAB2IP rs7025486, and IL6R rs2228145 were associated with the development of abdominal aortic aneurysms. Individuals with risk factors for atherosclerosis and/or a family history of AAA should be evaluated using genetic analysis.

Snare-Dragging Technique to Target the Hypogastric Artery in an Iliac Bifurcation Dissection.

PURPOSE: To describe snare-assisted vessel targeting to selectively overcome a dissection in the iliac bifurcation and gain antegrade access to the hypogastric artery (HA). TECHNIQUE: The technique is demonstrated in a 64-year-old woman with an asymptomatic Crawford type III thoracoabdominal aneurysm. A 2-stage endovascular repair, consisting of a thoracic endovascular aortic repair (TEVAR) and a branched endovascular aortic repair was planned. In the control angiography after TEVAR, a disrupted plaque with consequent dissection in the right iliac bifurcation was detected. The perfusion of the common iliac artery and external iliac artery resulted impaired. The targeting of the right HA through a contralateral antegrade approach failed, whereas an ipsilateral retrograde approach was possible but unsuitable for therapeutic purposes. Using the catheter of the retrograde ipsilateral access, a snare from a contralateral crossover was cached and dragged into the HA, allowing the targeting of the vessels and further endovascular therapy with angioplasty and stenting. Follow-up 8 months postoperatively demonstrated the patency of the stents and well-preserved perfusion in the right iliac bifurcation. CONCLUSION: The snare-dragging technique can be used to gain access to vessels presenting challenging conformations or dissections. This application may be a valuable support for complex endovascular treatment in a variety of patients. CLINICAL IMPACT: The snare-dragging technique can be used to gain access to vessels presenting challenging conformations or dissections. It allows the catheterization to be establish from the easiest and safest approach and then "transferred" from one access to the other. It avoids the risk of repeated loss of catheterization due to unstable and unfavorable working angles, and it saves time and radiation. It permits different material combinations, adapting to the available resources and materials. We believe that the current technique may increase the strategy spectrum available for endovascular therapy and complex endovascular procedures.

SGLT2 Inhibitors and Peripheral Vascular Events: A Review of the Literature.

Fifty articles comprising 18 randomized controlled trials (RCTs), 16 observational studies, and 16 meta-analyses on the safety and effectiveness of sodium-glucose cotransporter 2 inhibitors were evaluated in the current review. Only one-fourth of the cohorts of recent trials had peripheral arterial disease (PAD), whereas this subgroup was at high risk for amputations. Despite a remarkable heterogeneity of RCTs, only 2 trials on canagliflozin suggested excess amputation rates, whereas several observational studies generated conflicting conclusions and remained short on possible explanations. Preliminary evidence from observational research suggested that patients with PAD may even benefit from SGLT-2 inhibitor treatment due to lower observed heart failure hospitalization rates.

Effectiveness of Intra-Arterial Aneurysm Sac Embolization for Type Ia Endoleak after Endovascular Aneurysm Repair.

PURPOSE: To evaluate the effectiveness and durability of intra-arterial aneurysm sac embolization for the treatment of type Ia endoleak after endovascular aneurysm repair (EVAR). MATERIALS AND METHODS: From February 2011 to December 2016, 22 patients underwent embolization of a type Ia endoleak after EVAR. Four patients (18%) were treated during the index EVAR and 18 (82%) in follow-up. Five patients (23%) were treated urgently and 17 (77%) electively. The embolization was performed with the use of liquid embolic agent, coils, and/or plugs. Adjunctive neck procedures were performed in 55% (n = 12) of the patients. The primary endpoint of this study was freedom from sac enlargement. Key secondary endpoints were technical success and freedom from endoleak-related reinterventions. RESULTS: Technical success was 100%. The 30-day mortality was 5% (n = 1; acute coronary syndrome). At a mean follow-up of 15.4 months (range 0.1-65.4) the freedom from sac enlargement rate was 76% (16 out of 21). Reintervention-free survival rates at 6, 12, and 24 months were 80%, 68% and 68%, respectively. CONCLUSIONS: In patients with persistent type Ia endoleak the embolization of the aneurysm sac with or without adjunctive neck procedures can be safely performed, leading to acceptable clinical and radiologic outcomes.

Novelty Detection using Deep Normative Modeling for IMU-Based Abnormal Movement Monitoring in Parkinson's Disease and Autism Spectrum Disorders.

Detecting and monitoring of abnormal movement behaviors in patients with Parkinson's Disease (PD) and individuals with Autism Spectrum Disorders (ASD) are beneficial for adjusting care and medical treatment in order to improve the patient's quality of life. Supervised methods commonly used in the literature need annotation of data, which is a time-consuming and costly process. In this paper, we propose deep normative modeling as a probabilistic novelty detection method, in which we model the distribution of normal human movements recorded by wearable sensors and try to detect abnormal movements in patients with PD and ASD in a novelty detection framework. In the proposed deep normative model, a movement disorder behavior is treated as an extreme of the normal range or, equivalently, as a deviation from the normal movements. Our experiments on three benchmark datasets indicate the effectiveness of the proposed method, which outperforms one-class SVM and the reconstruction-based novelty detection approaches. Our contribution opens the door toward modeling normal human movements during daily activities using wearable sensors and eventually real-time abnormal movement detection in neuro-developmental and neuro-degenerative disorders.

Surgical Treatment of Endotension after Chimney Endovascular Repair of a Symptomatic Juxtarenal Aneurysm.

Endoleaks and persistent aneurysm sac expansion, also known as endotension, remain the Achilles heel of endovascular aneurysm repair (EVAR). The present article reports on a surgical treatment of 1 patient with endotension after chimney EVAR for a symptomatic juxtarenal aortic aneurysm. The surgical exploration revealed no signs of inflammation or infection and excluded any endoleak, confirming the term "endotension". After 1 year, the absence of sac expansion or endoleak was confirmed, with stable maximum diameter. Regarding therapeutic options, a nonoperative approach is recommended in asymptomatic patients. However, the management strategy for sac expansion without evidence of endoleaks as in case of endotension remains controversial.

Use of Self-Expanding Covered Stents as Bailout Treatment of Spontaneous Rupture of the Femoral Artery Caused by Staphylococcus aureus Infection.

Spontaneous rupture of femoropopliteal arteries caused by infection requires treatment in the urgent setting. The present article reports on the performance of self-expanding covered stents in 3 patients who experienced acute rupture of the femoral artery caused by Staphylococcus aureus infection. After a mean follow-up of 16.4 months, the bailout placement of self-expanding covered devices showed sustained patency and sufficient exclusion of the rupture site. A surgical conversion and bypass reconstruction was not necessary. Long-term follow-up and inclusion of more patients are needed to provide more evidence about the utility of this approach.

Differences in sequencing technologies improve the retrieval of anammox bacterial genome from metagenomes.

BACKGROUND: Sequencing technologies have different biases, in single-genome sequencing and metagenomic sequencing; these can significantly affect ORFs recovery and the population distribution of a metagenome. In this paper we investigate how well different technologies represent information related to a considered organism of interest in a metagenome, and whether it is beneficial to combine information obtained using different technologies. We analyze comparatively three metagenomic datasets acquired from a sample containing the anammox bacterium Candidatus 'Brocadia fulgida' (B. fulgida). These datasets were obtained using Roche 454 FLX and Sanger sequencing with two different libraries (shotgun and fosmid). RESULTS: In each dataset, the abundance of the reads annotated to B. fulgida was much lower than the abundance expected from available cell count information. This was due to the overrepresentation of GC-richer organisms, as shown by GC-content distribution of the reads. Nevertheless, by considering the union of B. fulgida reads over the three datasets, the number of B. fulgida ORFs recovered for at least 80% of their length was twice the amount recovered by the best technology. Indeed, while taxonomic distributions of reads in the three datasets were similar, the respective sets of B. fulgida ORFs recovered for a large part of their length were highly different, and depth of coverage patterns of 454 and Sanger were dissimilar. CONCLUSIONS: Precautions should be sought in order to prevent the overrepresentation of GC-rich microbes in the datasets. This overrepresentation and the consistency of the taxonomic distributions of reads obtained with different sequencing technologies suggests that, in general, abundance biases might be mainly due to other steps of the sequencing protocols. Results show that biases against organisms of interest could be compensated combining different sequencing technologies, due to the differences of their genome-level sequencing biases even if the species was present in not very different abundances in the metagenomes.

Evaluation of Ocular Perfusion in Patients with End-Stage Renal Disease Receiving Hemodialysis Using Optical Coherence Tomography Angiography.

Hemodialysis (HD) is known to affect ocular blood flow. This case-control study aims to evaluate macular and peripapillary vasculature in patients with end-stage renal disease (ESRD) receiving HD in comparison to matched controls. A total of 24 eyes of 24 ESRD patients receiving HD and 24 eyes of 24 healthy, age- and gender-matched control subjects were prospectively included in this study. Optical coherence tomography angiography was used to image the superficial (SCP), deep (DCP), and choriocapillary (CC) macular vascular plexus, as well as the radial peripapillary capillaries (RPC) of the optic disc. In addition, retinal thickness (RT) and retinal volume (RV) were compared between both groups. Flow density (FD) values of each retinal layer and data of parameters related to the foveal avascular zone (FAZ), as well as RT and RV, were analyzed using Mann-Whitney U tests. There was no significant difference in FAZ parameters between the two groups. Whole en face FD of the SCP and CC was noticeably reduced in the HD group in comparison to the control group. FD was negatively correlated with the duration of HD treatment. RT and RV were significantly smaller in the study group than in controls. Retinal microcirculation appears altered in patients with ESRD undergoing HD. Concurrently, the DCP appears more resilient towards hemodynamic changes in comparison to the other microvascular retinal layers. OCTA is a useful, non-invasive tool to investigate retinal microcirculation in ESRD patients.

A methodology for detecting the orthology signal in a PPI network at a functional complex level.

BACKGROUND: Stable evolutionary signal has been observed in a yeast protein-protein interaction (PPI) network. These finding suggests more connected regions of a PPI network to be potential mediators of evolutionary information. Because more connected regions of PPI networks contain functional complexes, we are motivated to exploit the orthology relation for identifying complexes that can be clearly attributed to such evolutionary signal. RESULTS: We proposed a computational methodology for detecting the orthology signal present in a PPI network at a functional complex level. Specifically, we examined highly functionally coherent putative protein complexes as detected by a clustering technique in the complete yeast PPI network, in the yeast sub-network which spans only ortholog proteins as determined by a given second organism, and in yeast sub-networks induced by a set of proteins randomly selected. We proposed a filtering technique for extracting orthology-driven clusters with unique functionalities, that is, neither enriched by clusters identified using the complete yeast PPI network nor identified using random sampling. Moreover, we extracted functional categories that can be clearly attributed to the presence of evolutionary signal as described by these clusters. CONCLUSIONS: Application of the proposed methodology to the yeast PPI network indicated that evolutionary information at a functional complex level can be retrieved from the structure of the network. In particular, we detected protein complexes whose functionality could be uniquely attributed to the evolutionary signal. Moreover, we identified functions that are over-represented in these complexes due the evolutionary signal.

Gaussian interaction profile kernels for predicting drug-target interaction.

MOTIVATION: The in silico prediction of potential interactions between drugs and target proteins is of core importance for the identification of new drugs or novel targets for existing drugs. However, only a tiny portion of all drug-target pairs in current datasets are experimentally validated interactions. This motivates the need for developing computational methods that predict true interaction pairs with high accuracy. RESULTS: We show that a simple machine learning method that uses the drug-target network as the only source of information is capable of predicting true interaction pairs with high accuracy. Specifically, we introduce interaction profiles of drugs (and of targets) in a network, which are binary vectors specifying the presence or absence of interaction with every target (drug) in that network. We define a kernel on these profiles, called the Gaussian Interaction Profile (GIP) kernel, and use a simple classifier, (kernel) Regularized Least Squares (RLS), for prediction drug-target interactions. We test comparatively the effectiveness of RLS with the GIP kernel on four drug-target interaction networks used in previous studies. The proposed algorithm achieves area under the precision-recall curve (AUPR) up to 92.7, significantly improving over results of state-of-the-art methods. Moreover, we show that using also kernels based on chemical and genomic information further increases accuracy, with a neat improvement on small datasets. These results substantiate the relevance of the network topology (in the form of interaction profiles) as source of information for predicting drug-target interactions. AVAILABILITY: Software and Supplementary Material are available at http://cs.ru.nl/~tvanlaarhoven/drugtarget2011/. CONTACT: tvanlaarhoven@cs.ru.nl; elenam@cs.ru.nl. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MTR: taxonomic annotation of short metagenomic reads using clustering at multiple taxonomic ranks.

MOTIVATION: Metagenomics is a recent field of biology that studies microbial communities by analyzing their genomic content directly sequenced from the environment. A metagenomic dataset consists of many short DNA or RNA fragments called reads. One interesting problem in metagenomic data analysis is the discovery of the taxonomic composition of a given dataset. A simple method for this task, called the Lowest Common Ancestor (LCA), is employed in state-of-the-art computational tools for metagenomic data analysis of very short reads (about 100 bp). However LCA has two main drawbacks: it possibly assigns many reads to high taxonomic ranks and it discards a high number of reads. RESULTS: We present MTR, a new method for tackling these drawbacks using clustering at Multiple Taxonomic Ranks. Unlike LCA, which processes the reads one-by-one, MTR exploits information shared by reads. Specifically, MTR consists of two main phases. First, for each taxonomic rank, a collection of potential clusters of reads is generated, and each potential cluster is associated to a taxon at that rank. Next, a small number of clusters is selected at each rank using a combinatorial optimization algorithm. The effectiveness of the resulting method is tested on a large number of simulated and real-life metagenomes. Results of experiments show that MTR improves on LCA by discarding a significantly smaller number of reads and by assigning much more reads at lower taxonomic ranks. Moreover, MTR provides a more faithful taxonomic characterization of the metagenome population distribution. AVAILABILITY: Matlab and C++ source codes of the method available at http://cs.ru.nl/gori/software/MTR.tar.gz.

Embolization for Type Ia Endoleak after EVAR for Abdominal Aortic Aneurysms: A Systematic Review of the Literature.

(1) Successful endovascular repair for abdominal aortic aneurysms is based on the complete exclusion of the aneurysm sac from the systemic circulation. Type Ia endoleak (ELIA) is defined as the persistent perfusion of the aneurysm sac due to incomplete proximal sealing between aorta and endograft, with a consequent risk of rupture and death. Endoleak embolization has been sporadically reported as a viable treatment for ELIA. (2) A systematic literature search in PubMed of all publications in English about ELIA embolization was performed until February 2022. Research methods and reporting were performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. Data regarding patient numbers, technical success (endoleak absence at control angiography), reinterventions, clinical and imaging follow-up, and outcomes were collected and examined by two independent authors. (3) Twenty-one papers (12 original articles, 9 case reports) reported on 126 patients (age range 58-96 years) undergoing ELIA embolization 0-139 months after the index procedure. Indication for embolization was most often founded on unfavorable anatomy and patient comorbidities. Embolic agents used include liquid embolic agents, coils, plugs and combinations thereof. Technical success in this highly selected cohort ranged from 67-100%; the postprocedural complication rate within 30 days was 0-24%. ELIA recurrence was reported as 0-42.8%, with a secondary ELIA-embolization-intervention success rate of 50-100%. At a follow-up at 0-68 months, freedom from sac enlargement amounted to 76-100%, freedom from ELIA to 66.7-100%. (4) Specific literature about ELIA embolization is scant. ELIA embolization is a valuable bailout strategy for no-option patients; the immediate technical success rate is high and midterm and long-term outcomes are acceptable.

Impact of Frozen and Conventional Elephant Trunk on Aortic New-Onset Thrombus and Inflammatory Response.

(1) Aim: The primary endpoint of this study was to evaluate the impact of frozen elephant trunk (FET) and conventional elephant trunk (CET) on aortic mural thrombus. The secondary endpoint was to investigate the incidence of persistent inflammatory response (IR) in the form of post-implantation syndrome (PIS) or persistent fever without infection focus after FET and CET, respectively, as well as the risk factors associated with its occurrence. (2) Methods: A single-center, retrospective, observational study of 57 consecutive patients treated with FET and CET between April 2015 and June 2020 was performed. Demographics, procedural data, perioperative laboratory exams as well as vital parameters were recorded. Pre- and postoperative computer tomography angiography (CTA) scans were analyzed with a dedicated software. IR was defined as the presence of continuous fever (>38°, lasting > 24 h) and leukocytosis (white blood cell count > 12 × 1000/µL) developing after surgery in the absence of an infection focus. (3) Results: Fifty-seven consecutive patients (mean age 58.4 ± 12.6 years, 36.8% females) treated with FET (66.6%) or CET (33.3%) for acute aortic dissection (56.1%), post-dissection-aneurysm (19.2%) or aortic aneurysm (24.5%) were included. The median thrombus volume on CTA preoperatively was 10.1 cm3 (range 2−408 cm3). After surgery, the median new-onset mural thrombus was 9.7 cm3 (range 0.2−376 cm3). Nineteen (33.3%) patients developed IR; patients with IR were significantly younger (p = 0.027), less frequently of female gender (p = 0.003) and more frequently affected from acute dissection (p = 0.002) and stayed in the intensive care unit (ICU) significantly longer (p = 0.033) than those without IR. Postoperatively, the volume of new-onset thrombus was significantly greater in the IR group (84.4 vs. 3.2 cm3, p < 0.001). (4) Conclusions: In the context of CET and FET, the persistent inflammatory response occurred in 33.3% of the patients with persistent fever without infection focus. IR was associated with a higher volume of new-onset thrombus and significantly prolonged ICU stay. Further studies to investigate these observations are needed.

Deep learning for Alzheimer's disease diagnosis: A survey.

Alzheimer's Disease (AD) is an irreversible neurodegenerative disease that results in a progressive decline in cognitive abilities. Since AD starts several years before the onset of the symptoms, its early detection is challenging due to subtle changes in biomarkers mainly detectable in different neuroimaging modalities. Developing computer-aided diagnostic models based on deep learning can provide excellent opportunities for the analysis of different neuroimage modalities along with other non-image biomarkers. In this survey, we perform a comparative analysis of about 100 published papers since 2019 that employ basic deep architectures such as CNN, RNN, and generative models for AD diagnosis. Moreover, about 60 papers that have applied a trending topic or architecture for AD are investigated. Explainable models, normalizing flows, graph-based deep architectures, self-supervised learning, and attention mechanisms are considered. The main challenges in this body of literature have been categorized and explained from data-related, methodology-related, and clinical adoption aspects. We conclude our paper by addressing some future perspectives and providing recommendations to conduct further studies for AD diagnosis.

Volume Changes in the Descending Aorta after Frozen Elephant Trunk and Conventional Hemi-Arch Repair after Acute Type A Aortic Dissection.

The aim of this study was to compare the mortality rates, re-intervention rates, and volumetric changes in aortas following surgery, in terms of the true lumen and false lumen changes, using conventional hemi-arch repair (CET) and frozen elephant trunk (FET) techniques. During the period from 2015 to 2018, 66 patients underwent surgical treatment for acute aortic dissection (Debakey type 1). Demographic and procedure-related data were evaluated. We measured volumetric change before surgical treatment, at discharge, and at 12- and 24-month time points based on computed tomography angiography. The study cohort was divided into two groups (FET vs. CET). The mean age of the patients was 56.9 ± 9.4 years in the FET group versus 63.6 ± 11 years in the CET group (p = 0.063). The mean follow-up time was 24 ± 6 and 25 ± 5 months for the FET and CET groups, respectively. There were no significant differences between the two groups in terms of the medical histories of the cohorts. The results showed a significant increase in true lumen volume after the FET procedure (within 24 months postoperatively; p = 0.005), and no significant changes in total (p = 0.392) or false lumen (p = 0.659) volumes were noted. After the CET procedure, there were significant increases in total and false lumen volumes (p = 0.013, p = 0.042), while no significant change in true lumen was observed (p = 0.219). The volume increase in true lumen after the FET procedure was higher compared to the CET group at all postoperative time points (at discharge, 12 months, and 24 months) without significant evidence (p = 0.416, p = 0.422, p = 0.268). At two years, the volume increase in false lumen was significantly higher among the CET group compared to the FET group (p = 0.02). The Kaplan-Meier curve analysis showed that patients who underwent the CET procedure underwent significantly more re-interventions due to false lumen expansion of the descending aorta (p = 0.047). Present study results indicate that the true and false lumen changes in the aorta following the FET and CET procedures were different. FET led to a significant increase in true lumen volume, while false lumen volume remained stable; however, after the CET procedure, significant false lumen enlargement was noted at mid-term follow-up time points. The re-intervention rate after CET was higher due to false lumen expansion.

Analysis of Spinal Ischemia after Frozen Elephant Trunk for Acute Aortic Dissection: An Observational, Single-Center Study.

BACKGROUND: This observational study aimed to evaluate the perioperative risk factors for spinal cord ischemia (SCI) in patients who underwent aortic repair with the frozen elephant trunk technique (FET) after acute aortic Stanford A dissection. METHODS: From May 2015 to April 2019, 31 patients underwent aortic arch replacement with the FET technique, and spinal ischemia was observed in 4 patients. The risk factors for postoperative SCI were analyzed. RESULTS: The mean age of patients with acute aortic dissection was 57.1 years, and 29.4% were female. Four patients developed SCI. There were no significant differences in characteristics such as age and body mass index. The female gender was associated with most of the SCI cases in the univariate analysis (75%, p = 0.016). Known perioperative and intraoperative risk factors were not related to postoperative SCI in our study. Patients who developed SCI had increased serum postoperative creatinine levels (p = 0.03). Twenty-four patients showed complete false lumen thrombosis up to zones 3-4, five patients up to zones 5-6 and two patients up to zones 7-9, which correlates with the postoperative development of SCI (p = 0.02). The total number of patent intercostal arteries was significantly reduced postoperatively in SCI patients (p = 0.044). CONCLUSIONS: Postoperative acute kidney injury, the reduction in patent intercostal arteries after surgery and the extension of false lumen thrombosis up to and beyond zone 5 may play a significant role in the development of clinically relevant spinal cord injury after FET.

Convolutional neural networks to predict brain tumor grades and Alzheimer's disease with MR spectroscopic imaging data.

PURPOSE: To evaluate the value of convolutional neural network (CNN) in the diagnosis of human brain tumor or Alzheimer's disease by MR spectroscopic imaging (MRSI) and to compare its Matthews correlation coefficient (MCC) score against that of other machine learning methods and previous evaluation of the same data. We address two challenges: 1) limited number of cases in MRSI datasets and 2) interpretability of results in the form of relevant spectral regions. METHODS: A shallow CNN with only one hidden layer and an ad-hoc loss function was constructed involving two branches for processing spectral and image features of a brain voxel respectively. Each branch consists of a single convolutional hidden layer. The output of the two convolutional layers is merged and fed to a classification layer that outputs class predictions for the given brain voxel. RESULTS: Our CNN method separated glioma grades 3 and 4 and identified Alzheimer's disease patients using MRSI and complementary MRI data with high MCC score (Area Under the Curve were 0.87 and 0.91 respectively). The results demonstrated superior effectiveness over other popular methods as Partial Least Squares or Support Vector Machines. Also, our method automatically identified the spectral regions most important in the diagnosis process and we show that these are in good agreement with existing biomarkers from the literature. CONCLUSION: Shallow CNNs models integrating image and spectral features improved quantitative and exploration and diagnosis of brain diseases for research and clinical purposes. Software is available at https://bitbucket.org/TeslaH2O/cnn_mrsi.

The metagenomic basis of anammox metabolism in Candidatus 'Brocadia fulgida'.

Anammox (anaerobic ammonium oxidation) coupled to nitrite reduction is an important step in the nitrogen cycle and has been recognized as an important sink for fixed nitrogen in the ocean. Still little is known about the genomic blueprint of different anammox species. In the present article, we discuss the important genes of anammox metabolism in Candidatus 'Brocadia fulgida' that were retrieved via a metagenomic approach.

Encoder-decoder models for chest X-ray report generation perform no better than unconditioned baselines.

High quality radiology reporting of chest X-ray images is of core importance for high-quality patient diagnosis and care. Automatically generated reports can assist radiologists by reducing their workload and even may prevent errors. Machine Learning (ML) models for this task take an X-ray image as input and output a sequence of words. In this work, we show that ML models for this task based on the popular encoder-decoder approach, like 'Show, Attend and Tell' (SA&T) have similar or worse performance than models that do not use the input image, called unconditioned baseline. An unconditioned model achieved diagnostic accuracy of 0.91 on the IU chest X-ray dataset, and significantly outperformed SA&T (0.877) and other popular ML models (p-value < 0.001). This unconditioned model also outperformed SA&T and similar ML methods on the BLEU-4 and METEOR metrics. Also, an unconditioned version of SA&T obtained by permuting the reports generated from images of the test set, achieved diagnostic accuracy of 0.862, comparable to that of SA&T (p-value ≥ 0.05).