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

Pathology Pathology

Noninvasive inferring expressed genes and in vivo monitoring of the physiology and pathology of pregnancy using cell-free DNA.

In American journal of obstetrics and gynecology

BACKGROUND : Noninvasive monitoring of fetal development and the early detection of pregnancy-associated complications is challenging, largely due to the lack of information about the molecular spectrum during pregnancy. Recently, cell-free DNA in plasma was found to reflect the global nucleosome footprint and status of gene expression, and showed potential for noninvasive health monitoring during pregnancy.

OBJECTIVE(S) : We aimed to test the relationships between plasma cell-free DNA profiles and pregnancy biology, and evaluate the use of cell-free DNA profile as a non-invasive method for physiological and pathological status monitoring during pregnancy.

STUDY DESIGN : We used genome cell-free DNA sequencing data generated from non-invasive prenatal testing in a total of 2937 pregnant women. For each physiological and pathological conditions, features of cell-free DNA profile were identified using the discovery cohort, and support vector machines classifiers were built and evaluated using independent training and validation cohorts.

RESULTS : We established nucleosome occupancy profiles at transcription start sites in different gestational trimesters, demonstrated the relationships between gene expression and cell-free DNA coverage at transcription start sites, and showed that the cell-free DNA profiles at transcription start sites represented the biological processes of pregnancy. In addition, using cell-free DNA data, nucleosome profiles of transcription factor binding sites were identified to reflect transcription factor footprint, which may help to reveal the molecular mechanisms underlying pregnancy. Finally, by using machine learning models on low coverage non-invasive prenatal testing data, we evaluated the use of cell-free DNA nucleosome profiles for distinguishing gestational trimesters, fetal gender, and fetal trisomy 21, and highlighted its potential utility for predicting physiological and pathological fetal conditions by using low coverage non-invasive prenatal testing data.

CONCLUSION(S) : Our analyses profiled nucleosome footprints and regulatory networks during pregnancy and established a noninvasive, proof-of-principle methodology for health monitoring during pregnancy.

Han Bo-Wei, Yang Fang, Guo Zhi-Wei, Ouyang Guo-Jun, Liang Zhi-Kun, Weng Rong-Tao, Yang Xu, Huang Li-Ping, Wang Ke, Li Fen-Xia, Huang Jie, Yang Xue-Xi, Wu Ying-Song

2020-Aug-29

Cell-free DNA, non-invasive prenatal testing (NIPT), nucleosome footprint, pregnancy, whole genome sequencing

Pathology Pathology

A Machine Learning Approach for the Automated Interpretation of Plasma Amino Acid Profiles.

In Clinical chemistry ; h5-index 61.0

BACKGROUND : Plasma amino acid (PAA) profiles are used in routine clinical practice for the diagnosis and monitoring of inherited disorders of amino acid metabolism, organic acidemias, and urea cycle defects. Interpretation of PAA profiles is complex and requires substantial training and expertise to perform. Given previous demonstrations of the ability of machine learning (ML) algorithms to interpret complex clinical biochemistry data, we sought to determine if ML-derived classifiers could interpret PAA profiles with high predictive performance.

METHODS : We collected PAA profiling data routinely performed within a clinical biochemistry laboratory (2084 profiles) and developed decision support classifiers with several ML algorithms. We tested the generalization performance of each classifier using a nested cross-validation (CV) procedure and examined the effect of various subsampling, feature selection, and ensemble learning strategies.

RESULTS : The classifiers demonstrated excellent predictive performance, with the 3 ML algorithms tested producing comparable results. The best-performing ensemble binary classifier achieved a mean precision-recall (PR) AUC of 0.957 (95% CI 0.952, 0.962) and the best-performing ensemble multiclass classifier achieved a mean F4 score of 0.788 (0.773, 0.803).

CONCLUSIONS : This work builds upon previous demonstrations of the utility of ML-derived decision support tools in clinical biochemistry laboratories. Our findings suggest that, pending additional validation studies, such tools could potentially be used in routine clinical practice to streamline and aid the interpretation of PAA profiles. This would be particularly useful in laboratories with limited resources and large workloads. We provide the necessary code for other laboratories to develop their own decision support tools.

Wilkes Edmund H, Emmett Erin, Beltran Luisa, Woodward Gary M, Carling Rachel S

2020-Sep-01

Machine learning, inherited metabolic disease, plasma amino acids

General General

Modeling cannabinoids from a large-scale sample of Cannabis sativa chemotypes.

In PloS one ; h5-index 176.0

The widespread legalization of Cannabis has opened the industry to using contemporary analytical techniques for chemotype analysis. Chemotypic data has been collected on a large variety of oil profiles inherent to the cultivars that are commercially available. The unknown gene regulation and pharmacokinetics of dozens of cannabinoids offer opportunities of high interest in pharmacology research. Retailers in many medical and recreational jurisdictions are typically required to report chemical concentrations of at least some cannabinoids. Commercial cannabis laboratories have collected large chemotype datasets of diverse Cannabis cultivars. In this work a data set of 17,600 cultivars tested by Steep Hill Inc., is examined using machine learning techniques to interpolate missing chemotype observations and cluster cultivars into groups based on chemotype similarity. The results indicate cultivars cluster based on their chemotypes, and that some imputation methods work better than others at grouping these cultivars based on chemotypic identity. Due to the missing data and to the low signal to noise ratio for some less common cannabinoids, their behavior could not be accurately predicted. These findings have implications for characterizing complex interactions in cannabinoid biosynthesis and improving phenotypical classification of Cannabis cultivars.

Vergara Daniela, Gaudino Reggie, Blank Thomas, Keegan Brian

2020

General General

Attention-Based Road Registration for GPS-Denied UAS Navigation.

In IEEE transactions on neural networks and learning systems

Matching and registration between aerial images and prestored road landmarks are critical techniques to enhance unmanned aerial system (UAS) navigation in the global positioning system (GPS)-denied urban environments. Current registration processes typically consist of two separate stages of road extraction and road registration. These two-stage registration approaches are time-consuming and less robust to noise. To that end, in this article, we, for the first time, investigate the problem of end-to-end Aerial-Road registration. Using deep learning, we develop a novel attention-based neural network architecture for Aerial-Road registration. In this model, we construct two-branch neural networks with shared weights to map two input images into a common embedding space. Besides, considering that road features are sparsely distributed in images, we incorporate a novel multibranch attention module to filter out false descriptor matches from the indiscriminative background in order to improve registration accuracy. Finally, the results from extensive experiments show that compared with state-of-the-art approaches, the mean absolute errors of our approach in rotation angle and the translations in the x- and y-directions are reduced down by a factor of 1.24, 1.38, and 1.44, respectively. Furthermore, as a byproduct, our experimental results prove the feasibility of a neural network multitask learning approach to simultaneously achieve accurate Aerial-Road matching and registration, thus providing an efficient and accurate UAS geolocalization.

Wang Teng, Zhao Ye, Wang Jiawei, Somani Arun K, Sun Changyin

2020-Sep-01

General General

3D Virtual Pancreatography.

In IEEE transactions on visualization and computer graphics

We present 3D virtual pancreatography (VP), a novel visualization procedure and application for non-invasive diagnosis and classification of pancreatic lesions, the precursors of pancreatic cancer. Currently, non-invasive screening of patients is performed through visual inspection of 2D axis-aligned CT images, though the relevant features are often not clearly visible nor automatically detected. VP is an end-to-end visual diagnosis system that includes: a machine learning based automatic segmentation of the pancreatic gland and the lesions, a semi-automatic approach to extract the primary pancreatic duct, a machine learning based automatic classification of lesions into four prominent types, and specialized 3D and 2D exploratory visualizations of the pancreas, lesions and surrounding anatomy. We combine volume rendering with pancreas- and lesion-centric visualizations and measurements for effective diagnosis. We designed VP through close collaboration and feedback from expert radiologists, and evaluated it on multiple real-world CT datasets with various pancreatic lesions and case studies examined by the expert radiologists.

Jadhav Shreeraj, Dmitriev Konstantin, Marino Joseph, Barish Matthew, Kaufman Arie E

2020-Sep-01

General General

Reconstructing 3D Shapes from Multiple Sketches using Direct Shape Optimization.

In IEEE transactions on image processing : a publication of the IEEE Signal Processing Society

3D shape reconstruction from multiple hand-drawn sketches is an intriguing way to 3D shape modeling. Currently, state-of-the-art methods employ neural networks to learn a mapping from multiple sketches from arbitrary view angles to a 3D voxel grid. Because of the cubic complexity of 3D voxel grids, however, neural networks are hard to train and limited to low resolution reconstructions, which leads to a lack of geometric detail and low accuracy. To resolve this issue, we propose to reconstruct 3D shapes from multiple sketches using direct shape optimization (DSO), which does not involve deep learning models for direct voxel-based 3D shape generation. Specifically, we first leverage a conditional generative adversarial network (CGAN) to translate each sketch into an attenuance image that captures the predicted geometry from a given viewpoint. Then, DSO minimizes a project-and-compare loss to reconstruct the 3D shape such that it matches the predicted attenuance images from the view angles of all input sketches. Based on this, we further propose a progressive update approach to handle inconsistencies among a few hand-drawn sketches for the same 3D shape. Our experimental results show that our method significantly outperforms the state-of-the-art methods under widely used benchmarks and produces intuitive results in an interactive application.

Han Zhizhong, Ma Baorui, Liu Yu-Shen, Zwicker Matthias

2020-Sep-01