Articles in 2020

Predicting overall survival for patients with confirmed non-small-cell lung cancer is an important issue in clinical practice. The authors developed and validated in four independent patient cohorts a shallow convolutional neural network that can predict the outcomes of individuals using pre-treatment CT images. The authors further show that the survival model can be used, via transfer learning, for classifying benign versus malignant nodules.

While computerization and digitization of medicine have advanced substantially, management tools in healthcare have not yet benefited much from these developments due to the extreme complexity and variability of healthcare operations. The ability of machine learning algorithms to build strong models from a large number of weakly predictive features, and to identify key factors in complex feature sets, is tested in operational problems involving hospital datasets on workflow and patient waiting time.

Early and accurate clinical assessment of disease severity in COVID-19 patients is essential for planning the allocation of scarce hospital resources. An explainable machine learning tool trained on blood sample data from 485 patients from Wuhan selected three biomarkers for predicting mortality of individual patients with high accuracy.

Current neural networks attempt to learn spatial and temporal information as a whole, limiting their ability to process complex video data. Pang et al. improve performance by introducing a network structure which learns to implicitly decouple complex spatial and temporal concepts.

The deep convolutional recurrent neural network ‘PredNet’ can be trained to predict future video frames in a self-supervised manner. A surprising result is that it captures a wide array of phenomena observed in natural neuronal systems, ranging from low-level visual cortical neuron response properties to high-level perceptual illusions, hinting at potential similarities between recurrent predictive neural network models and computations in the brain.

4D MRI scans can reconstruct cardiovascular flow, although they typically take many minutes, hindering real-time assessment. Vishnevskiy et al. develop a deep variational network to permit high-fidelity image reconstruction in a matter of seconds, allowing integration of 4D flow MRI into clinical workflows.

A lack of accurate and efficient variant calling methods has held back single-molecule sequencing technologies from clinical applications. The authors present a deep-learning method for fast and accurate germline small variant calling, using single-molecule sequencing data.

Finding the best ratio of ingredients for polymerization reactions can be time consuming and wasteful. An automated microreactor process with integrated machine learning analysis initiates reactions, measures the resulting yield and cleans itself without human intervention. It can test concentrations of reagents systematically to find the combination with the highest production, while producing a low amount of waste.

Integrating knowledge about the circuit-level organization of the brain into neuromorphic artificial systems is a challenging research problem. The authors present a neural algorithm for the learning of odourant signals and their robust identification under noise, based on the architecture of the mammalian olfactory bulb and implemented on the Intel Loihi neuromorphic system.

With the aid of deep learning, the space of chemical molecules, such as candidates for drugs, can be constrained to find new bioactive molecules. A new open source tool can generate libraries of novel molecules with user defined properties.

When predicting the interaction of proteins with potential drugs, the protein can be encoded as its one-dimensional sequence or a three-dimensional structure, which can capture more relevant features of the protein, but also makes the task to predict the interactions harder. A new method predicts these interactions using a two-dimensional distance matrix representation of a protein, which can be processed like a two-dimensional image, striking a balance between the data being simple to process and rich in relevant structures.

Getting safe and fast access to blood vessels is vital to many methods of treatment and diagnosis in medicine. Robot-assisted or even fully autonomous methods can potentially do the task more reliably than humans, especially when veins are hard to detect. In this work, a method is tested that uses deep learning to find blood vessels and track the movement of a patient’s arm.

Persistent homology provides an efficient approach to simplifying the complexity of protein structure. Wang et al. combine this approach with convolutional neural networks and gradient-boosting trees to improve predictions of protein–protein interactions.

Age-related macular degeneration is a serious eye disease which should be detected as early as possible. Using both fundus images and genetic information, a deep neural network is able to detect the severity of the disease and predict its progression seven years into the future.

Counting different types of circulating tumour cells can give valuable information on the severity of the disease and on whether treatments are effective for a specific patient. In this work, the authors show that their method based on autoencoders can identify and count cells more accurately and faster than human experts.

The dynamics of paper shapes in free fall are still not fully understood, despite being discussed for more than 150 years. Collecting large amounts of data has the potential to give us new insights and a robotics system could generate and analyse data in large quantities.

Neural networks are often implemented with reduced precision in order to meet the tight energy and memory budget required by edge computing devices. Chakraborty et al. develop a technique for assessing which layers can be quantized, and by how much, without sacrificing too much on performance.

By assembling conceptual systems from real-word datasets of text, images and audio, Roads and Love propose that objects embedded within a conceptual system have a unique signature that allows for conceptual systems to be aligned in an unsupervised fashion.

Tree-based machine learning models are widely used in domains such as healthcare, finance and public services. The authors present an explanation method for trees that enables the computation of optimal local explanations for individual predictions, and demonstrate their method on three medical datasets.

Magnetic resonance scans use different contrast agents to generate different images, each giving specific clinical information. Lee et al. use a collaborative generative model to synthesize some magnetic resonance contrasts from others, providing guidance for how clinical imaging times can be reduced.