Search

Your search keyword '"Liu, Jinghui"' showing total 617 results
Start Over Author "Liu, Jinghui"
617 results on '"Liu, Jinghui"'

1. Rephrasing Electronic Health Records for Pretraining Clinical Language Models

Author

Liu, Jinghui and Nguyen, Anthony

Subjects
Computer Science - Computation and Language
Abstract

Clinical language models are important for many applications in healthcare, but their development depends on access to extensive clinical text for pretraining. However, obtaining clinical notes from electronic health records (EHRs) at scale is challenging due to patient privacy concerns. In this study, we rephrase existing clinical notes using LLMs to generate synthetic pretraining corpora, drawing inspiration from previous work on rephrasing web data. We examine four popular small-sized LLMs (<10B) to create synthetic clinical text to pretrain both decoder-based and encoder-based language models. The method yields better results in language modeling and downstream tasks than previous synthesis approaches without referencing real clinical text. We find that augmenting original clinical notes with synthetic corpora from different LLMs improves performances even at a small token budget, showing the potential of this method to support pretraining at the institutional level or be scaled to synthesize large-scale clinical corpora.

Published
2024

2. Selective excitation of work-generating cycles in nonreciprocal living solids

Author

Chao, Yu-Chen, Gokhale, Shreyas, Lin, Lisa, Hastewell, Alasdair, Bacanu, Alexandru, Chen, Yuchao, Li, Junang, Liu, Jinghui, Lee, Hyunseok, Dunkel, Jorn, and Fakhri, Nikta

Subjects
Condensed Matter - Soft Condensed Matter, Condensed Matter - Statistical Mechanics, Physics - Biological Physics
Abstract

Emergent nonreciprocity in active matter drives the formation of self-organized states that transcend the behaviors of equilibrium systems. Integrating experiments, theory and simulations, we demonstrate that active solids composed of living starfish embryos spontaneously transition between stable fluctuating and oscillatory steady states. The nonequilibrium steady states arise from two distinct chiral symmetry breaking mechanisms at the microscopic scale: the spinning of individual embryos resulting in a macroscopic odd elastic response, and the precession of their rotation axis, leading to active gyroelasticity. In the oscillatory state, we observe long-wavelength optical vibrational modes that can be excited through mechanical perturbations. Strikingly, these excitable nonreciprocal solids exhibit nonequilibrium work generation without cycling protocols, due to coupled vibrational modes. Our work introduces a novel class of tunable nonequilibrium processes, offering a framework for designing and controlling soft robotic swarms and adaptive active materials, while opening new possibilities for harnessing nonreciprocal interactions in engineered systems.

Published
2024

3. Light-induced cortical excitability reveals programmable shape dynamics in starfish oocytes

Author

Liu, Jinghui, Burkart, Tom, Ziepke, Alexander, Reinhard, John, Chao, Yu-Chen, Tan, Tzer Han, Swartz, S. Zachary, Frey, Erwin, and Fakhri, Nikta

Subjects
Physics - Biological Physics
Abstract

Chemo-mechanical waves on active deformable surfaces are a key component for many vital cellular functions. In particular, these waves play a major role in force generation and long-range signal transmission in cells that dynamically change shape, as encountered during cell division or morphogenesis. Reconstituting and controlling such chemically controlled cell deformations is a crucial but unsolved challenge for the development of synthetic cells. Here, we develop an optogenetic method to elucidate the mechanism responsible for coordinating surface contraction waves that occur in oocytes of the starfish Patiria miniata during meiotic cell division. Using spatiotemporally-patterned light stimuli as a control input, we create chemo-mechanical cortical excitations that are decoupled from meiotic cues and drive diverse shape deformations ranging from local pinching to surface contraction waves and cell lysis. We develop a quantitative model that entails the hierarchy of chemical and mechanical dynamics, which allows to relate the variety of mechanical responses to optogenetic stimuli. Our framework systematically predicts and explains transitions of programmed shape dynamics. Finally, we qualitatively map the observed shape dynamics to elucidate how the versatility of intracellular protein dynamics can give rise to a broad range of mechanical phenomenologies. More broadly, our results pave the way toward real-time control over dynamical deformations in living organisms and can advance the design of synthetic cells and life-like cellular functions., Comment: 36 pages, 16 figures, 11 movies

Published
2024

4. e-Health CSIRO at RRG24: Entropy-Augmented Self-Critical Sequence Training for Radiology Report Generation

Author

Nicolson, Aaron, Liu, Jinghui, Dowling, Jason, Nguyen, Anthony, and Koopman, Bevan

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

The Shared Task on Large-Scale Radiology Report Generation (RRG24) aims to expedite the development of assistive systems for interpreting and reporting on chest X-ray (CXR) images. This task challenges participants to develop models that generate the findings and impression sections of radiology reports from CXRs from a patient's study, using five different datasets. This paper outlines the e-Health CSIRO team's approach, which achieved multiple first-place finishes in RRG24. The core novelty of our approach lies in the addition of entropy regularisation to self-critical sequence training, to maintain a higher entropy in the token distribution. This prevents overfitting to common phrases and ensures a broader exploration of the vocabulary during training, essential for handling the diversity of the radiology reports in the RRG24 datasets. Our model is available on Hugging Face https://huggingface.co/aehrc/cxrmate-rrg24.

Published
2024

5. e-Health CSIRO at 'Discharge Me!' 2024: Generating Discharge Summary Sections with Fine-tuned Language Models

Author

Liu, Jinghui, Nicolson, Aaron, Dowling, Jason, Koopman, Bevan, and Nguyen, Anthony

Subjects
Computer Science - Computation and Language
Abstract

Clinical documentation is an important aspect of clinicians' daily work and often demands a significant amount of time. The BioNLP 2024 Shared Task on Streamlining Discharge Documentation (Discharge Me!) aims to alleviate this documentation burden by automatically generating discharge summary sections, including brief hospital course and discharge instruction, which are often time-consuming to synthesize and write manually. We approach the generation task by fine-tuning multiple open-sourced language models (LMs), including both decoder-only and encoder-decoder LMs, with various configurations on input context. We also examine different setups for decoding algorithms, model ensembling or merging, and model specialization. Our results show that conditioning on the content of discharge summary prior to the target sections is effective for the generation task. Furthermore, we find that smaller encoder-decoder LMs can work as well or even slightly better than larger decoder based LMs fine-tuned through LoRA. The model checkpoints from our team (aehrc) are openly available., Comment: BioNLP @ ACL 2024

Published
2024

7. Design and Implementation of Intelligent Coupled Rearview Mirror Variable Position Structure

Author

Liu, Jinghui, Xu, Tongqiang, Du, Guangfu, Wu, Liying, Chaari, Fakher, Series Editor, Gherardini, Francesco, Series Editor, Ivanov, Vitalii, Series Editor, Haddar, Mohamed, Series Editor, Cavas-Martínez, Francisco, Editorial Board Member, di Mare, Francesca, Editorial Board Member, Kwon, Young W., Editorial Board Member, Tolio, Tullio A. M., Editorial Board Member, Trojanowska, Justyna, Editorial Board Member, Schmitt, Robert, Editorial Board Member, Xu, Jinyang, Editorial Board Member, Pham, Duc Truong, editor, Lei, Yaguo, editor, and Lou, Yanshan, editor

Published
2025
Full Text
View/download PDF

10. Improving Text-based Early Prediction by Distillation from Privileged Time-Series Text

Author

Liu, Jinghui, Capurro, Daniel, Nguyen, Anthony, and Verspoor, Karin

Subjects
Computer Science - Computation and Language
Abstract

Modeling text-based time-series to make prediction about a future event or outcome is an important task with a wide range of applications. The standard approach is to train and test the model using the same input window, but this approach neglects the data collected in longer input windows between the prediction time and the final outcome, which are often available during training. In this study, we propose to treat this neglected text as privileged information available during training to enhance early prediction modeling through knowledge distillation, presented as Learning using Privileged tIme-sEries Text (LuPIET). We evaluate the method on clinical and social media text, with four clinical prediction tasks based on clinical notes and two mental health prediction tasks based on social media posts. Our results show LuPIET is effective in enhancing text-based early predictions, though one may need to consider choosing the appropriate text representation and windows for privileged text to achieve optimal performance. Compared to two other methods using transfer learning and mixed training, LuPIET offers more stable improvements over the baseline, standard training. As far as we are concerned, this is the first study to examine learning using privileged information for time-series in the NLP context., Comment: Accepted by ALTA 2022

Published
2023

21. Impact of Planting Spacing and Soil Amendment with Water Absorbent on Root and Leave Senescence Traits of Drip Irrigated Oat

Author

TIAN Lu, LIU Jinghui, MI Junzhen, ZHAO Baoping, LI Yinghao, ZHANG Sheng, WANG Fengwu, JIAO Weihong, XU Zhenpeng, and ZHENG Chengzhong

Subjects
oat, drip irrigation, row spacing, super absorbent polymer, senescence characteristic, Agriculture (General), S1-972, Irrigation engineering. Reclamation of wasteland. Drainage, TC801-978
Abstract

【Objective】 Roots and leaves are highly influenced by a multitude of biotic and abiotic factors. Taking oat under drip irrigation as an example, this study delves into the response of its root and leave traits to planting density and soil amendment with water absorbent polymer. 【Method】 The field experiment was conducted in 2019—2020 with intra-row planting spacing being 20 cm, and the inter-row planting spacing being 20 cm (A), 15 cm (B) and 10 cm (C), respectively. In each treatment there were two soil amendments by applying 0 and 22.5 kg/hm2 of a water absorbent polymer. Senescence traits of the roots and flag leaves in each treatment were measured. 【Result】 ①Both row spacing and soil amendment affected root activity, SOD and MDA content, and soluble protein content in the roots and leaves significantly with PA>C, while the impact on MDA content in the roots and leaves was in the order of C>A>B. Combined with soil amendment, reducing row spacing from 20 cm to 15 cm increased root activity, SOD activity in the roots and leaves, soluble protein content in the roots and leaves by 5.56%~15.27%, 4.78%~5.89%, 3.98%~10.62%, 5.20%~23.74% and 5.89%~15.04% respectively, while reducing MDA content in the roots and leaves by 11.79%~13.86% and 6.01%~14.08%, respectively. Reducing the row spacing further to 10 cm significantly reduced the above traits, except the soluble protein content, which did not show significant changes, and MDA content in the roots and leaves, which showed a significant increase. Without soil amendment, the impact of row spacing on root and leaf traits varied with growth stage and year. ③When row spacing was 20 cm, compared with no amendment, amending the soil increased SOD activity and soluble protein content in the roots and leaves by 7.52%~15.75%, 3.21%~6.47%, 7.25%~10.47%, 18.01%~22.47% and 4.87%~9.10%, respectively, while reducing MDA content in the roots and leaves by 8.79%~13.96% and 7.35%~11.67%, respectively. In contrast, the effect of soil amendment on the above traits varied when the row spacing was 10 cm. ④The senescence traits of roots and leaves were correlated with biomass and grain yield of the oat, both at significant levels. 【Conclusion】 Reducing the row spacing from 20 cm to 15 cm, combined with soil amendment by applying 22.5 kg/hm2 of the water-absorbent polymer, is effective in reducing senescence of roots and leaves of the drip-irrigated oat and consequently promoting its biological yield and grain yield.

Published
2023
Full Text
View/download PDF

25. Effects of Reduced Nitrogen Fertilization Via Controlled Release on Dry Matter Accumulation, Photosynthetic Traits and Yield of Proso Millet

Author

WANG Ying, LIU Jinghui, WU Junying, MI Junzhen, SUN Mengyuan, ZHANG Lanying, and FENG Xueying

Subjects
dry matter accumulation and distribution, nitrogen balance index, net photosynthetic rate, spad, Agriculture (General), S1-972, Irrigation engineering. Reclamation of wasteland. Drainage, TC801-978
Abstract

【Objective】 Proso millet is a common crop cultivated in the Loess Plateau in northwestern China and sustainable production needs to reduce nitrogen fertilization. This paper presents the findings of an experimental study that investigated the impact of reducing nitrogen fertilization through controlled release on photosynthetic traits and yield of the millet. 【Method】 The field experiment was conducted with application of 120 kg/hm2 of urea taken as the control (TN). There were seven controlled-release treatments, which applied N fertilizer at 120 kg/hm2 (T1), 108 kg/hm2 (T2), 96 kg/hm2 (T3), 84 kg/hm2 (T4), 72 kg/hm2 (T5), and zero (T0). During the experiment, we measured the changes in dry matter accumulation, photosynthetic traits of the flag leaves, SPAD, NBI and yield of the millets in each treatment and the CK. 【Result】 ①Nitrogen fertilization significantly improved the dry matter accumulation and photosynthetic traits of the flag leaves at all growing stages, regardless of treatments. Compared with CK, applying the same amount of N fertilizer but at controlled release increased the dry matter accumulation in the middle and late stages by 4.44%~8.77%, proportion of the dry matter allocated to the ears at maturity stage by 1.58%~2.75%. ②Depending on the application rate, controlled release of N fertilizer led to a 2.19% to 5.05% increase in chlorophyll content and a 2.66% to 8.34% increase in nitrogen balance index of the flag leaves; it also increased the net photosynthetic rate, stomatal conductance and transpiration rate at the filling stage by 3.90%, 6.40% and 4.20%, respectively, while reducing the average intercellular CO2 concentration by 4.45%. ③With the decrease in controlled-release N fertilization, both dry matter accumulation and photosynthetic rate decreased, and the decrease was more significant in T3—T0. SPAD and NBI of the flag leaves were positively correlated with the dry matter accumulation, net photosynthetic rate and the yield; the correlations between yield and NBI at the jointing and filling stages was significantly positive. ④Applying the same amount of N as in TN but with controlled release increased the yield of the millet by 4.42%, equivalent to an increase in net income by 121~280 Yuan/hm2. Reducing 10%~20% of N used in the TN with the controlled release did not reduce the yield significantly, but further reduction beyond 20% led to a decrease in yield by 3.19%~5.23%. 【Conclusion】 Applying N fertilizer in controlled release to proso millet in hilly areas in Loess Plateau can improve its yield. It hence has a potential application for other crops in this region and other regions with similar growing environments.

Published
2023
Full Text
View/download PDF

30. Influence of UHV GIL access on transient recovery voltage of circuit breaker

Author

SUN Yi, LIU Jinghui, ZHENG Xinlei, LI Jiangtao, XU Yiqing, and SHI Dajun

Subjects
ultra high voltage (uhv) gas-insulated transmission line (gil), circuit breaker, transient recovery voltage (trv), rate of rise of trv (rrrv), hybrid transmission line, emtp-atp simulation, Applications of electric power, TK4001-4102
Abstract

The access of ultra high voltage (UHV) gas-insulated transmission line (GIL) affects both the transient recovery voltage (TRV) amplitude and the rate of rise of TRV (RRRV) of circuit breakers,which may endanger the breaking capacity of circuit breakers. Based on the double-circuit UHV accurate current (AC) overhead line on the same tower and the double-circuit GIL hybrid transmission line,EMTP-ATP simulation is used to study the influence of GIL length,access position and GIL lead station measurement equipment on the TRV amplitude and RRRV of the circuit breaker when a three-phase short-circuit fault occurs at the end of the circuit breaker in the substation. The simulation results show that the access of the GIL has a greater impact on the TRV amplitude and RRRV of the circuit breaker. The increase in the length of the GIL has a greater impact on the TRV amplitude,but the impact on RRRV is small. In order to limit the RRRV,the GIL should be installed in the middle of the overhead line,and the metal oxide varistor (MOV) should be installed to limit the TRV amplitude. The measurement equipment of the GIL terminal station has little effect on the TRV amplitude and RRRV of the circuit breaker. The research results can provide a reference for the TRV calculation and parameter design of circuit breakers in UHV lines with GIL.

Published
2022
Full Text
View/download PDF

35. Research on Optimization Design Method of Reliability Validation Test in the Case of Zero-Failure Data

Author

Qi, Ming, Yang, Zekun, Liu, Jinghui, Li, Xuefeng, Wu, Dengyun, Angrisani, Leopoldo, Series Editor, Arteaga, Marco, Series Editor, Panigrahi, Bijaya Ketan, Series Editor, Chakraborty, Samarjit, Series Editor, Chen, Jiming, Series Editor, Chen, Shanben, Series Editor, Chen, Tan Kay, Series Editor, Dillmann, Rüdiger, Series Editor, Duan, Haibin, Series Editor, Ferrari, Gianluigi, Series Editor, Ferre, Manuel, Series Editor, Hirche, Sandra, Series Editor, Jabbari, Faryar, Series Editor, Jia, Limin, Series Editor, Kacprzyk, Janusz, Series Editor, Khamis, Alaa, Series Editor, Kroeger, Torsten, Series Editor, Liang, Qilian, Series Editor, Martín, Ferran, Series Editor, Ming, Tan Cher, Series Editor, Minker, Wolfgang, Series Editor, Misra, Pradeep, Series Editor, Möller, Sebastian, Series Editor, Mukhopadhyay, Subhas, Series Editor, Ning, Cun-Zheng, Series Editor, Nishida, Toyoaki, Series Editor, Pascucci, Federica, Series Editor, Qin, Yong, Series Editor, Seng, Gan Woon, Series Editor, Speidel, Joachim, Series Editor, Veiga, Germano, Series Editor, Wu, Haitao, Series Editor, Zhang, Junjie James, Series Editor, Wang, Yue, editor, Xu, Lexi, editor, Yan, Yufeng, editor, and Zou, Jiaqi, editor

Published
2021
Full Text
View/download PDF

47. The Effects of Controlled-release of Urea on Yield and Water Use Efficiency of Rain-fed Potato

Author

SUN Mengyuan, LIU Jinghui, MI Junzhen, ZAHNG Lanying, and WANG Ying

Subjects
nitrogen application, controlled-releasing urea, potato yield, water use efficiency, nitrogen partial productivity, economic benefit, Agriculture (General), S1-972, Irrigation engineering. Reclamation of wasteland. Drainage, TC801-978
Abstract

【Background】 Drought and low nitrogen use efficiency are double whammy of developing sustainable agriculture in many countries in semi-arid areas that face increased land degradation and climate change. 【Objective】 This paper is to investigate the effects of controlled-release of urea on nitrogen form in soil, as well as yield, economic benefit, nitrogen partial productivity and water use efficiency of rain-fed potato, in attempts to provide guidance for improving fertilization management in loess hilly-gully region. 【Method】 The experiment was conducted in a field, and it compared the impact of common urea application (CF) and the controlled-release of urea (CRU) on yield, chlorophyll (Chl), flavonols (Flav), nitrogen balance index (NBI), nitrogen partial productivity and water use efficiency of the potato. The amount of urea used in the conventional application was 200 kg N/hm2 (CF), and those used in the controlled-release were 10% (CRU1), 20% (CRU2) 30% (CRU3) of that used in CF, respectively. The control was non-fertilization (CK). 【Result】 With nitrogen application increasing, the Chl and NBI content in the leaves increased whereas the Flav decreased. Growth stage and the amount of nitrogen application both affected the Chl, NBI and Flav at significant level (P0.05). Both NBI and Flav were highly correlated with the yield during the tuber expansion stage. Reducing nitrogen application led to a decrease in tuber yield and economic benefit, and the reduction in economic benefit was ranked in the order of CF>CRU1>CRU2>CRU3>CK. The yields of potato in CRU2 and CRU3 were 88.90% and 85.33% of that in CF, respectively. Compared to CRU1, CF increased the tuber yield and economic benefit by 5.18% and 24.26% respectively, but it reduced the marketable potato rate and nitrogen partial productivity by 6.50% and 5.33% respectively. Reducing nitrogen application resulted in a decrease in water consumption and water use efficiency. CF increased water consumption and improved water use efficiency significantly (P0.05). 【Conclusion】 Our results showed that compared to CRU1, CRU2, CRU3 and CF increased water consumption by 2.79%~5.02%, and water use efficiency by 2.33%~11.58% respectively. They are thus optimal urea applications to improve both potato yield and water use efficiency, while in the meantime increased the economic benefit in the loess hilly-gully region we studied.

Published
2021
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results