Search

There has been a growing effort to replace manual extraction of data from research papers with automated data extraction based on natural language processing, language models, and recently, large language models (LLMs). Although these methods enable efficient extraction of data from large sets of research papers, they require a significant amount of up-front effort, expertise, and coding. In this work, we propose the ChatExtract method that can fully automate very accurate data extraction with minimal initial effort and background, using an advanced conversational LLM. ChatExtract consists of a set of engineered prompts applied to a conversational LLM that both identify sentences with data, extract that data, and assure the data's correctness through a series of follow-up questions. These follow-up questions largely overcome known issues with LLMs providing factually inaccurate responses. ChatExtract can be applied with any conversational LLMs and yields very high quality data extraction. In tests on materials data, we find precision and recall both close to 90% from the best conversational LLMs, like GPT-4. We demonstrate that the exceptional performance is enabled by the information retention in a conversational model combined with purposeful redundancy and introducing uncertainty through follow-up prompts. These results suggest that approaches similar to ChatExtract, due to their simplicity, transferability, and accuracy are likely to become powerful tools for data extraction in the near future. Finally, databases for critical cooling rates of metallic glasses and yield strengths of high entropy alloys are developed using ChatExtract. Efficient data extraction from research papers accelerates science and engineering. Here, the authors develop an automated approach which uses conversational large language models to achieve high precision and recall in extracting materials data. [ABSTRACT FROM AUTHOR]

On the 12th of April 2010, Nature Communications published its first editorial and primary research articles. The topics of these first 11 papers represented the multidisciplinary nature of the journal: from DNA damage to optics alongside material science to energy and including polymer chemistry. We have spoken with the corresponding authors of some of these very first papers and asked them about their experience of publishing in this then new journal and how they see Nature Communications now. [ABSTRACT FROM AUTHOR]

The abuse of antibiotics urgently requires rapid identification of drug-resistant bacteria at the point of care (POC). Here we report a visual paper sensor that allows rapid (0.25-3 h) discrimination of the subtypes of β-lactamase (the major cause of bacterial resistance) for precision antibiotic therapy. The sensor exhibits high performance in identifying antibiotic-resistant bacteria with 100 real samples from patients with diverse bacterial infections, demonstrating 100% clinical sensitivity and specificity. Further, this sensor can enhance the accuracy of antibiotic use from 48% empirically to 83%, and further from 50.6% to 97.6% after eliminating fungal infection cases. Our work provides a POC testing platform for guiding effective management of bacterial infections in both hospital and community settings. The rapid identification of drug-resistant bacteria is vital for effective treatment and to avoid antibiotic misuse. Here authors report a paper-based sensor which utilises chromogenic carbapenem and cephalosporin substrates for the identification and discrimination of β-lactamase subtypes. [ABSTRACT FROM AUTHOR]

The lack of data democratization and information leakage from trained models hinder the development and acceptance of robust deep learning-based healthcare solutions. This paper argues that irreversible data encoding can provide an effective solution to achieve data democratization without violating the privacy constraints imposed on healthcare data and clinical models. An ideal encoding framework transforms the data into a new space where it is imperceptible to a manual or computational inspection. However, encoded data should preserve the semantics of the original data such that deep learning models can be trained effectively. This paper hypothesizes the characteristics of the desired encoding framework and then exploits random projections and random quantum encoding to realize this framework for dense and longitudinal or time-series data. Experimental evaluation highlights that models trained on encoded time-series data effectively uphold the information bottleneck principle and hence, exhibit lesser information leakage from trained models. Healthcare data democratization is often hampered by privacy constraints governing the sensitive healthcare data. Here, the authors show that encoding healthcare data could be a potential solution for achieving healthcare democratization within the context of deep learning. [ABSTRACT FROM AUTHOR]

Accurate state-of-health (SOH) estimation is critical for reliable and safe operation of lithium-ion batteries. However, reliable and stable battery SOH estimation remains challenging due to diverse battery types and operating conditions. In this paper, we propose a physics-informed neural network (PINN) for accurate and stable estimation of battery SOH. Specifically, we model the attributes that affect the battery degradation from the perspective of empirical degradation and state space equations, and utilize neural networks to capture battery degradation dynamics. A general feature extraction method is designed to extract statistical features from a short period of data before the battery is fully charged, enabling our method applicable to different battery types and charge/discharge protocols. Additionally, we generate a comprehensive dataset consisting of 55 lithium-nickel-cobalt-manganese-oxide (NCM) batteries. Combined with three other datasets from different manufacturers, we use a total of 387 batteries with 310,705 samples to validate our method. The mean absolute percentage error (MAPE) is 0.87%. Our proposed PINN has demonstrated remarkable performance in regular experiments, small sample experiments, and transfer experiments when compared to alternative neural networks. This study highlights the promise of physics-informed machine learning for battery degradation modeling and SOH estimation. Reliable lithium-ion battery health assessment is vital for safety. Here, authors present a physics-informed neural network for accurate and stable state-of-health estimation, overcoming challenges of varied battery types and usage conditions. [ABSTRACT FROM AUTHOR]

Point-scanning microscopy approaches are transforming super-resolution imaging. Despite achieving parallel high-speed imaging using multifocal techniques, efficient multicolor imaging methods with high-quality illumination are currently lacking. In this paper, we present for the first time Mechanical-scan-free multiColor Super-resolution Microscopy (MCoSM) with spot array illumination, which enables mechanical-scan-free super-resolution imaging with adjustable resolution and a good effective field-of-view based on spatial light modulators. Through 100–2,500 s super-resolution spot illumination with different effective fields of view for imaging, we demonstrate the adjustable capacity of MCoSM. MCoSM extends existing spectral imaging capabilities through a time-sharing process involving different color illumination with phase-shift scanning while retaining the spatial flexibility of super-resolution imaging with diffractive spot array illumination. To demonstrate the prospects of MCoSM, we perform four-color imaging of fluorescent beads at high resolution. MCoSM provides a versatile platform for studying molecular interactions in complex samples at the nanoscale level. Here, the authors use spot array illumination to enable mechanical-scan-free super-resolution microscopy with adjustable resolution and good effective field of view, demonstrating a platform for studying molecular interactions at the nanoscale. [ABSTRACT FROM AUTHOR]

The current thyroid ultrasound relies heavily on the experience and skills of the sonographer and the expertise of the radiologist, and the process is physically and cognitively exhausting. In this paper, we report a fully autonomous robotic ultrasound system, which is able to scan thyroid regions without human assistance and identify malignant nod- ules. In this system, human skeleton point recognition, reinforcement learning, and force feedback are used to deal with the difficulties in locating thyroid targets. The orientation of the ultrasound probe is adjusted dynamically via Bayesian optimization. Experimental results on human participants demonstrated that this system can perform high-quality ultrasound scans, close to manual scans obtained by clinicians. Additionally, it has the potential to detect thyroid nodules and provide data on nodule characteristics for American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) calculation. Current thyroid ultrasounds rely heavily on the experience and skills of the sonographer and of the radiologist, and the process is physically and cognitively exhausting. Here, the authors show a fully autonomous robotic ultrasound system, which is able to scan thyroid regions without human assistance and identify malignant nodules. [ABSTRACT FROM AUTHOR]

In clinical oncology, many diagnostic tasks rely on the identification of cells in histopathology images. While supervised machine learning techniques necessitate the need for labels, providing manual cell annotations is time-consuming. In this paper, we propose a self-supervised framework (enVironment-aware cOntrastive cell represenTation learning: VOLTA) for cell representation learning in histopathology images using a technique that accounts for the cell's mutual relationship with its environment. We subject our model to extensive experiments on data collected from multiple institutions comprising over 800,000 cells and six cancer types. To showcase the potential of our proposed framework, we apply VOLTA to ovarian and endometrial cancers and demonstrate that our cell representations can be utilized to identify the known histotypes of ovarian cancer and provide insights that link histopathology and molecular subtypes of endometrial cancer. Unlike supervised models, we provide a framework that can empower discoveries without any annotation data, even in situations where sample sizes are limited. While machine learning platforms can improve the assessment of Hematoxylin & Eosin (H&E) stained-tumour tissue images, current models typically require manual cell-type annotations in training. Here, the authors develop VOLTA, a self-supervised machine learning framework to improve cell representation learning in H&E images based on the cells environment [ABSTRACT FROM AUTHOR]

One of the most effective ways to advance the performance of quantum computers and quantum sensors is to increase the number of qubits or quantum resources in the system. A major technical challenge that must be solved to realize this goal for trapped-ion systems is scaling the delivery of optical signals to many individual ions. In this paper we demonstrate an approach employing waveguides and multi-mode interferometer splitters to optically address multiple 171Yb+ ions in a surface trap by delivering all wavelengths required for full qubit control. Measurements of hyperfine spectra and Rabi flopping were performed on the E2 clock transition, using integrated waveguides for delivering the light needed for Doppler cooling, state preparation, coherent operations, and detection. We describe the use of splitters to address multiple ions using a single optical input per wavelength and use them to demonstrate simultaneous Rabi flopping on two different transitions occurring at distinct trap sites. This work represents an important step towards the realization of scalable integrated photonics for atomic clocks and trapped-ion quantum information systems. A promising strategy for scaling trapped-ion-based quantum technologies is to use fully integrated optical waveguides to deliver light to numerous ions at multiple sites. Here, the authors. optically address three ions using on-chip waveguides to deliver three distinct wavelengths per ion, and perform Rabi flopping on each ion simultaneously. [ABSTRACT FROM AUTHOR]

An important difference between brains and deep neural networks is the way they learn. Nervous systems learn online where a stream of noisy data points are presented in a non-independent, identically distributed way. Further, synaptic plasticity in the brain depends only on information local to synapses. Deep networks, on the other hand, typically use non-local learning algorithms and are trained in an offline, non-noisy, independent, identically distributed setting. Understanding how neural networks learn under the same constraints as the brain is an open problem for neuroscience and neuromorphic computing. A standard approach to this problem has yet to be established. In this paper, we propose that discrete graphical models that learn via an online maximum a posteriori learning algorithm could provide such an approach. We implement this kind of model in a neural network called the Sparse Quantized Hopfield Network. We show our model outperforms state-of-the-art neural networks on associative memory tasks, outperforms these networks in online, continual settings, learns efficiently with noisy inputs, and is better than baselines on an episodic memory task. Brains and neuromorphic systems learn with local learning rules in online-continual learning scenarios. Designing neural networks that learn effectively under these conditions is challenging. The authors introduce a neural network that implements an effective, principled approach to local, online-continual learning on associative memory tasks. [ABSTRACT FROM AUTHOR]