Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

Radiology Radiology

Image registration: Maximum likelihood, minimum entropy and deep learning.

In Medical image analysis

In this work, we propose a theoretical framework based on maximum profile likelihood for pairwise and groupwise registration. By an asymptotic analysis, we demonstrate that maximum profile likelihood registration minimizes an upper bound on the joint entropy of the distribution that generates the joint image data. Further, we derive the congealing method for groupwise registration by optimizing the profile likelihood in closed form, and using coordinate ascent, or iterative model refinement. We also describe a method for feature based registration in the same framework and demonstrate it on groupwise tractographic registration. In the second part of the article, we propose an approach to deep metric registration that implements maximum likelihood registration using deep discriminative classifiers. We show further that this approach can be used for maximum profile likelihood registration to discharge the need for well-registered training data, using iterative model refinement. We demonstrate that the method succeeds on a challenging registration problem where the standard mutual information approach does not perform well.

Sedghi Alireza, O’Donnell Lauren J, Kapur Tina, Learned-Miller Erik, Mousavi Parvin, Wells William M


Deep learning, Image registration, Information theory

General General

Graph convolution network with similarity awareness and adaptive calibration for disease-induced deterioration prediction.

In Medical image analysis

Graph convolution networks (GCN) have been successfully applied in disease prediction tasks as they capture interactions (i.e., edges and edge weights on the graph) between individual elements. The interactions in existing works are constructed by fusing similarity between imaging information and distance between non-imaging information, whereas disregarding the disease status of those individuals in the training set. Besides, the similarity is being evaluated by computing the correlation distance between feature vectors, which limits prediction performance, especially for predicting significant memory concern (SMC) and mild cognitive impairment (MCI). In this paper, we propose three mechanisms to improve GCN, namely similarity-aware adaptive calibrated GCN (SAC-GCN), for predicting SMC and MCI. First, we design a similarity-aware graph using different receptive fields to consider disease status. The labelled subjects on the graph are only connected with those labelled subjects with the same status. Second, we propose an adaptive mechanism to evaluate similarity. Specifically, we construct initial GCN with evaluating similarity by using traditional correlation distance, then pre-train the initial GCN by using training samples and use it to score all subjects. Then, the difference between these scores replaces correlation distance to update similarity. Last, we devise a calibration mechanism to fuse functional magnetic resonance imaging (fMRI) and diffusion tensor imaging (DTI) information into edges. The proposed method is tested on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. Experimental results demonstrate that our proposed method is useful to predict disease-induced deterioration and superior to other related algorithms, with a mean classification accuracy of 86.83% in our prediction tasks.

Song Xuegang, Zhou Feng, Frangi Alejandro F, Cao Jiuwen, Xiao Xiaohua, Lei Yi, Wang Tianfu, Lei Baiying


Adaptive mechanism, Calibration mechanism, Disease prediction, Dual-modal information, Graph convolution network, Similarity awareness

General General

Integrated meta-analysis and machine learning approach identifies acyl-CoA thioesterase with other novel genes responsible for biofilm development in Staphylococcus aureus.

In Infection, genetics and evolution : journal of molecular epidemiology and evolutionary genetics in infectious diseases

Biofilm forming Staphylococcus aureus is a major threat to the health-care industry. It is important to understand the differences between planktonic and biofilm growth forms in the pathogen since conventional treatments targeting the planktonic forms are not effective against biofilms. The current study conducts a meta-analysis of three public transcriptomic profiles to examine the differences in gene expression between the planktonic and biofilm states of S. aureus using random-effects modeling. Mean effect sizes were calculated for 2847 genes among which 726 differentially expressed genes were taken for further analysis. Major genes that are discriminatory between the two conditions were mined using supervised learning techniques and validated by high-accuracy classifiers. Ten different feature selection algorithms were applied and used to rank the most important genes in S. aureus biofilms. Finally, an optimal set of 36 genes are presented as candidate genes in biofilm formation or development while throwing light on the novel roles of an acyl-CoA thioesterase enzyme and 10 hypothetical proteins in biofilms. The relevance of the identified gene set was further validated by building five different classification models using SVM, RF, kNN, NB and DT algorithms that were compared with models built from other relevant gene sets and by reviewing the functional role of 25 previously known genes in biofilm development. The study combines meta-analysis of differential expression with supervised machine learning strategies and feature selection for the first time to identify and validate a discriminatory set of genes important in biofilms of S. aureus. The functional roles of the identified genes predicted to be important in biofilms are further scrutinized and can be considered as a signature target list to develop anti-biofilm therapeutics in S. aureus.

Subramanian Devika, Natarajan Jeyakumar


Biofilms, Feature selection, Meta-analysis, Staphylococcus aureus, Supervised machine-learning

Ophthalmology Ophthalmology

TET dependent GDF7 hypomethylation impairs aqueous humor outflow and serves as a potential therapeutic target in glaucoma.

In Molecular therapy : the journal of the American Society of Gene Therapy

Glaucoma is the leading cause of irreversible vision loss, affecting more than 70 million individuals worldwide. Circulatory disturbances of aqueous humor (AH) have long been central pathological contributors to glaucomatous lesions. Thus, targeting the AH outflow is a promising approach to treat glaucoma. However, the epigenetic mechanisms initiating AH outflow disorders and the targeted treatments remain to be developed. Studying glaucoma patients, we identified GDF7 (Growth Differentiation Factor 7) hypomethylation as a crucial event in the onset of AH outflow disorders. Regarding the underlying mechanism, the hypomethylated GDF7 promoter was responsible for the increased GDF7 production and secretion in POAG. Excessive GDF7 protein promoted trabecular meshwork (TM) fibrosis through BMPR2/Smad signaling and up-regulated pro-fibrotic genes, α-smooth muscle actin (α-SMA) and fibronectin (FN). GDF7 protein expression formed a positive feedback loop in GTM. This positive feedback loop was dependent on activated TET (ten-eleven translocation) enzyme, which kept GDF7 promoter region hypomethylated. The phenotypic transition in TM fortified the AH outflow resistance, thus elevating the intraocular pressure (IOP) and attenuating the nerve fiber layer. This methylation dependent mechanism is also confirmed by a machine-learning model in silico with a specificity of 84.38% and a sensitivity of 89.38%. In rhesus monkeys, we developed GDF7 neutralization therapy to inhibit TM fibrosis and consequent AH outflow resistance that contributes to glaucoma. The neutralization therapy achieved high-efficiency control of the IOP (from 21.3±0.3 to 17.6±0.2 mmHg), a three-fold improvement in the outflow facility (from 0.1 to 0.3 μl/min·mmHg), and protection of nerve fibers. This study provides new insights into the epigenetic mechanism of glaucoma and proposes an innovative GDF7 neutralization therapy as a promising intervention.

Wan Peixing, Long Erping, Li Zhidong, Zhu Yingting, Su Wenru, Zhuo Yehong


Computational modeling, DNA methylation, Fibrosis, Glaucoma, Neutralizing antibody, Trabecular meshwork

General General

Neurocognitive Mechanisms Supporting the Generalization of Concepts Across Languages.

In Neuropsychologia

The neurocognitive mechanisms that support the generalization of semantic representations across different languages remain to be determined. Current psycholinguistic models propose that semantic representations are likely to overlap across languages, although there is evidence also to the contrary. Neuroimaging studies observed that brain activity patterns associated with the meaning of words may be similar across languages. However, the factors that mediate cross-language generalization of semantic representations are not known. We here identify a key factor: the depth of processing. Human participants were asked to process visual words as they underwent functional MRI. We found that, during shallow processing, multivariate pattern classifiers could decode the word semantic category within each language in putative substrates of the semantic network, but there was no evidence of cross-language generalization in the shallow processing context. By contrast, when the depth of processing was higher, significant cross-language generalization was observed in several regions, including inferior parietal, ventromedial, lateral temporal, and inferior frontal cortex. These results are in keeping with distributed-only views of semantic processing and favour models based on multiple semantic hubs. The results also have ramifications for existing psycholinguistic models of word processing such as the BIA+, which by default assumes non-selective access to both native and second languages.

Sheikh Usman Ayub, Carreiras Manuel, Soto David


Semantic representation, bilingualism, language, machine learning

General General

Immune Computation and COVID-19 Mortality: A Rationale for IVIg.

In Critical reviews in immunology

COVID-19 infection tends to be more lethal in older persons than in the young; death results from an overactive inflammatory response, leading to cytokine storm and organ failure. Here we describe immune regulation of the inflammatory response phenotype as emerging from a process that is analogous to machine-learning algorithms used in computers. We briefly describe some strategic similarities between immune learning and computer machine learning. We reason that a balanced response to COVID-19 infection might be induced by treating the elderly patient with a wellness repertoire of antibodies obtained from healthy young people. We propose that a beneficial training set of such antibodies might be administered in the form of intravenous immunoglobulin (IVIg).

Cohen Irun R, Efroni Sol, Atlan Henri