1,827 results on '"Ho Lam"'
Search Results
2. Loss of function of GATA3 regulates FRA1 and c-FOS to activate EMT and promote mammary tumorigenesis and metastasis
- Author
-
Xiong Liu, Feng Bai, Yuchan Wang, Chuying Wang, Ho Lam Chan, Chenglong Zheng, Jian Fang, Wei-Guo Zhu, and Xin-Hai Pei
- Subjects
Cytology ,QH573-671 - Abstract
Abstract Basal-like breast cancers (BLBCs) are among the most aggressive cancers, partly due to their enrichment of cancer stem cells (CSCs). Breast CSCs can be generated from luminal-type cancer cells via epithelial-mesenchymal transition (EMT). GATA3 maintains luminal cell fate, and its expression is lost or reduced in BLBCs. However, deletion of Gata3 in mice or cells results in early lethality or proliferative defects. It is unknown how loss-of-function of GATA3 regulates EMT and CSCs in breast cancer. We report here that haploid loss of Gata3 in mice lacking p18Ink4c, a cell cycle inhibitor, up-regulates Fra1, an AP-1 family protein that promotes mesenchymal traits, and downregulates c-Fos, another AP-1 family protein that maintains epithelial fate, leading to activation of EMT and promotion of mammary tumor initiation and metastasis. Depletion of Gata3 in luminal tumor cells similarly regulates Fra1 and c-Fos in activation of EMT. GATA3 binds to FOSL1 (encoding FRA1) and FOS (encoding c-FOS) loci to repress FOSL1 and activate FOS transcription. Deletion of Fra1 or reconstitution of Gata3, but not reconstitution of c-Fos, in Gata3 deficient tumor cells inhibits EMT, preventing tumorigenesis and/or metastasis. In human breast cancers, GATA3 expression is negatively correlated with FRA1 and positively correlated with c-FOS. Low GATA3 and FOS, but high FOSL1, are characteristics of BLBCs. Together, these data provide the first genetic evidence indicating that loss of function of GATA3 in mammary tumor cells activates FOSL1 to promote mesenchymal traits and CSC function, while concurrently repressing FOS to lose epithelial features. We demonstrate that FRA1 is required for the activation of EMT in GATA3 deficient tumorigenesis and metastasis.
- Published
- 2023
- Full Text
- View/download PDF
3. Current microfluidic platforms for reverse engineering of cornea
- Author
-
Qinyu Li, Ho Lam Wong, Yan Lam Ip, Wang Yee Chu, Man Shek Li, Chinmoy Saha, Kendrick Co Shih, and Yau Kei Chan
- Subjects
Cornea-on-a-chip ,Microfluidic platform ,Tissue engineering ,Corneal microenvironment ,Pharmacokinetics ,Medicine (General) ,R5-920 ,Biology (General) ,QH301-705.5 - Abstract
According to the World Health Organization, corneal blindness constitutes 5.1% of global blindness population. Surgical outcomes have been improved significantly in the treatment of corneal blindness. However, corneal transplantation is limited by global shortage of donor tissue, prompting researchers to explore alternative therapies such as novel ocular pharmaceutics to delay corneal disease progression. Animal models are commonly adopted for investigating pharmacokinetics of ocular drugs. However, this approach is limited by physiological differences in the eye between animals and human, ethical issues and poor bench-to-bedside translatability. Cornea-on-a-chip (CoC) microfluidic platforms have gained great attention as one of the advanced in vitro strategies for constructing physiologically representative corneal models. With significant improvements in tissue engineering technology, CoC integrates corneal cells with microfluidics to recapitulate human corneal microenvironment for the study of corneal pathophysiological changes and evaluation of ocular drugs. Such model, in complement to animal studies, can potentially accelerate translational research, in particular the pre-clinical screening of ophthalmic medication, driving clinical treatment advancement for corneal diseases. This review provides an overview of engineered CoC platforms with respect to their merits, applications, and technical challenges. Emerging directions in CoC technology are also proposed for further investigations, to accentuate preclinical obstacles in corneal research.
- Published
- 2023
- Full Text
- View/download PDF
4. Transcriptomic study of anastasis for reversal of ethanol-induced apoptosis in mouse primary liver cells
- Author
-
Ho Man Tang, C. Conover Talbot, Ming Chiu Fung, and Ho Lam Tang
- Subjects
Science - Abstract
Measurement(s) Microarray Analysis Technology Type(s) Microarray Analysis Sample Characteristic - Organism Mus musculus
- Published
- 2022
- Full Text
- View/download PDF
5. The association between altered intestinal microbiome, impaired systemic and ocular surface immunity, and impaired wound healing response after corneal alkaline-chemical injury in diabetic mice
- Author
-
Yashan Bu, Kendrick Co Shih, Ho Lam Wong, Sum Sum Kwok, Amy Cheuk-Yin Lo, Joseph Yau-Kei Chan, Alex Lap-Ki Ng, Tommy Chung-Yan Chan, Vishal Jhanji, and Louis Tong
- Subjects
intestinal microbiome ,diabetes ,corneal wound healing ,alkaline chemical injury ,T-cell mediated immunity ,ocular surface ,Immunologic diseases. Allergy ,RC581-607 - Abstract
PurposeWe aim to investigate the effect of sustained hyperglycemia on corneal epithelial wound healing, ocular surface and systemic immune response, and microbiome indices in diabetic mice compared to controls after alkaline chemical injury of the eye.MethodsCorneal alkaline injury was induced in the right eye of Ins2Akita (Akita) mice and wild-type mice. The groups were observed at baseline and subsequently days 0, 3, and 7 after injury. Corneal re-epithelialization was observed under slit lamp with fluorescein staining using a cobalt blue light filter. Enucleated cornea specimens were compared at baseline and after injury for changes in cornea thickness under hematoxylin and eosin staining. Tear cytokine and growth factor levels were measured using protein microarray assay and compared between groups and time points. Flow cytometry was conducted on peripheral blood and ocular surface samples to determine CD3+CD4+ cell count. Fecal samples were collected, and gut microbiota composition and diversity pattern were measured using shotgun sequencing.ResultsAkita mice had significantly delayed corneal wound healing compared to controls. This was associated with a reduction in tear levels of vascular endothelial growth factor A, angiopoietin 2, and insulin growth factor 1 on days 0, 3, and 7 after injury. Furthermore, there was a distinct lack of upregulation of peripheral blood and ocular surface CD3+CD4+ cell counts in response to injury in Akita mice compared to controls. This was associated with a reduction in intestinal microbiome diversity indices in Akita mice compared to controls after injury. Specifically, there was a lower abundance of Firmicutes bacterium M10-2 in Akita mice compared to controls after injury.ConclusionIn diabetic mice, impaired cornea wound healing was associated with an inability to mount systemic and local immune response to ocular chemical injury. Baseline and post-injury differences in intestinal microbial diversity and abundance patterns between diabetic mice and controls may potentially play a role in this altered response.
- Published
- 2023
- Full Text
- View/download PDF
6. Survey of the management of patients with bronchiectasis: a pilot investigation in Asian populations
- Author
-
Ho Cheol Kim, Masaru Suzuki, Hui Fang Lim, Le Thi Tuyet Lan, Ho Lam Nguyen, Jeng-Shing Wang, Kang-Yun Lee, Jae Seung Lee, Yeon-Mok Oh, Sang Do Lee, Hayoung Choi, Hyun Lee, and Sei Won Lee
- Subjects
bronchiectasis ,survey ,asia ,Medicine - Abstract
Background/Aims Although international guidelines for bronchiectasis management have been published in Western countries, there is a lack of data about their application in Asian populations including patients with different phenotypes. We aimed to investigate the current status of bronchiectasis management in Asian populations. Methods A nationwide questionnaire survey was performed of Asian respiratory specialists from South Korea, Japan, Taiwan, Singapore, Vietnam, and Sri Lanka. Participants were invited by e-mail to answer a questionnaire comprising 25 questions based on international guidelines for the management of bronchiectasis. Results A total of 221 physicians participated in the survey. About half of them were Korean (50.2%), with the next most common nationalities being Japanese (23.1%), Taiwanese (13.6%), and Singaporean (7.7%). Only 18 (8.1%) responders had local guidelines for bronchiectasis. While 85 (38.5%) responders checked sputum acid-fast bacillus smear/culture about 1 to 3 times per year, only a small proportion of responders routinely performed a serum immunoglobulin test (36/221, 16.3%) or evaluated for allergic bronchopulmonary aspergillosis (41/221, 18.6%). Less than half (43.4%) of responders performed eradication treatment in patients with drug-sensitive Pseudomonas aeruginosa infection, mainly due to the limited availability of inhaled antibiotics (34.8%). In addition, 58.6% of responders considered physiotherapy such as airway clearance and pulmonary rehabilitation. Conclusions Discrepancies might exist between guideline recommendations and practice for bronchiectasis management in Asian populations, partly due to the limited availability of treatment in each country. The development of local guidelines that consider the phenotypes and situation will help to standardize and improve the management of bronchiectasis.
- Published
- 2021
- Full Text
- View/download PDF
7. Nonparametric Inference Framework for Time-dependent Epidemic Models
- Author
-
Luu, Son, Susko, Edward, and Ho, Lam Si Tung
- Subjects
Statistics - Methodology - Abstract
Compartmental models, especially the Susceptible-Infected-Removed (SIR) model, have long been used to understand the behaviour of various diseases. Allowing parameters, such as the transmission rate, to be time-dependent functions makes it possible to adjust for and make inferences about changes in the process due to mitigation strategies or evolutionary changes of the infectious agent. In this article, we attempt to build a nonparametric inference framework for stochastic SIR models with time dependent infection rate. The framework includes three main steps: likelihood approximation, parameter estimation and confidence interval construction. The likelihood function of the stochastic SIR model, which is often intractable, can be approximated using methods such as diffusion approximation or tau leaping. The infection rate is modelled by a B-spline basis whose knot location and number of knots are determined by a fast knot placement method followed by a criterion-based model selection procedure. Finally, a point-wise confidence interval is built using a parametric bootstrap procedure. The performance of the framework is observed through various settings for different epidemic patterns. The model is then applied to the Ontario COVID-19 data across multiple waves.
- Published
- 2024
8. Codec-SUPERB @ SLT 2024: A lightweight benchmark for neural audio codec models
- Author
-
Wu, Haibin, Chen, Xuanjun, Lin, Yi-Cheng, Chang, Kaiwei, Du, Jiawei, Lu, Ke-Han, Liu, Alexander H., Chung, Ho-Lam, Wu, Yuan-Kuei, Yang, Dongchao, Liu, Songxiang, Wu, Yi-Chiao, Tan, Xu, Glass, James, Watanabe, Shinji, and Lee, Hung-yi
- Subjects
Electrical Engineering and Systems Science - Audio and Speech Processing ,Computer Science - Sound - Abstract
Neural audio codec models are becoming increasingly important as they serve as tokenizers for audio, enabling efficient transmission or facilitating speech language modeling. The ideal neural audio codec should maintain content, paralinguistics, speaker characteristics, and audio information even at low bitrates. Recently, numerous advanced neural codec models have been proposed. However, codec models are often tested under varying experimental conditions. As a result, we introduce the Codec-SUPERB challenge at SLT 2024, designed to facilitate fair and lightweight comparisons among existing codec models and inspire advancements in the field. This challenge brings together representative speech applications and objective metrics, and carefully selects license-free datasets, sampling them into small sets to reduce evaluation computation costs. This paper presents the challenge's rules, datasets, five participant systems, results, and findings.
- Published
- 2024
9. Non-Destructive Disassembly of Interference Fit under Wear Conditions for Sustainable Remanufacturing
- Author
-
Ho Lam Au-Yeung, Sabbah Ataya, Hany Hassanin, Mahmoud Ahmed El-Sayed, Mahmoud Ahmadein, Naser A. Alsaleh, Mohamed M. Z. Ahmed, and Khamis Essa
- Subjects
interference fit ,press-fit pin-hub ,automated disassembly ,remanufacturing ,finite element analysis ,Mechanical engineering and machinery ,TJ1-1570 - Abstract
Remanufacturing has been gaining increasing attention in the last few years as a part of green engineering. It is the process of restoring the original specifications of a given product utilizing a combination of new, repaired, and old parts. The present study investigates non-destructive disassembly of an interference fit pin-hub joint to enable the reuse of worn parts with the same loading capacity. The aim is to reduce the disassembly force while preventing plastic deformation and frictional damage on the contact surface to avoid fretting failure and enable further coating. A finite element model of a shaft/hub interference fit was developed, taking into account two cases of damage to the mating parts: deformation and corrosion. The results indicate that thermal disassembly is effective in reducing breaking force by 50% in deformed joints, whereas vibration waves are more suitable for corroded parts with increased friction. In addition, applying a low-frequency oscillation force to the axis of disassembly reduces the pulling out force by 5% and plastic deformation by 99% due to acoustic softening effects. Furthermore, using a heat flux simultaneously with vibration decreases the breaking force by 85%, indicating the higher effectiveness of thermal-aided disassembly and vibration-assisted disassembly in reducing the breaking force of corroded parts with increased friction. This study provides remanufacturing designers with efficient tools to weaken the interference fit and decrease the disconnecting force, ultimately reducing the cost and time required for the disassembly process.
- Published
- 2023
- Full Text
- View/download PDF
10. Epigenetic mechanisms in breast cancer therapy and resistance
- Author
-
Liliana Garcia-Martinez, Yusheng Zhang, Yuichiro Nakata, Ho Lam Chan, and Lluis Morey
- Subjects
Science - Abstract
Endocrine therapy has been the mainstay for hormone responsive breast cancer treatment. Here, Garcia-Martinez and colleagues discuss epigenetic mechanisms regulating ER + breast cancer and endocrine therapy resistance, and highlight approaches to rewire the cancer epigenome to improve targeted therapies for this cancer.
- Published
- 2021
- Full Text
- View/download PDF
11. Photogrammetric reconstruction of 3D carpological collection in high resolution for plants authentication and species discovery
- Author
-
Ho Lam Wang, Tin Hang Wong, Yiu Man Chan, Yat Sum Cheng, and David Tai Wai Lau
- Subjects
Medicine ,Science - Abstract
This study provides an accurate and efficient method to reconstruct detailed and high-resolution digital 3D models of carpological materials by photogrammetric method, in which only about 100 to 150 images are required for each model reconstruction. The 3D models reflect the realistic morphology and genuine color of the carpological materials. The 3D models are scaled to represent the true size of the materials even as small as 3 mm in diameter. The interfaces are interactive, in which the 3D models can be rotated in 360° to observe the structures and be zoomed to inspect the macroscopic details. This new platform is beneficial for developing a virtual herbarium of carpological collection which is thus the most important to botanical authentication and education.
- Published
- 2022
12. Cost-utility analysis of robotic-assisted radical cystectomy for bladder cancer compared to open radical cystectomy in the United Kingdom.
- Author
-
Felix Machleid, Jenessa Ho-Wrigley, Ameera Chowdhury, Anita Paliah, Ho Lam Poon, and Elena Pizzo
- Subjects
Medicine ,Science - Abstract
BackgroundBladder cancer is the tenth most common cancer in the United Kingdom. Currently, open radical cystectomy (ORC) is the gold standard. Due to the risk of complications and a 2.3-8% mortality rate1, there is growing interest in the use of robot-assisted radical cystectomy (RARC). The aim of this study is to perform a cost-utility analysis, comparing RARC to ORC for bladder cancer patients from the perspective of the National Health Service England.MethodsA three-stage decision tree: surgery, post-surgery transfusions and complications, in a 90-day time horizon, was produced to simulate possible pathways of patients. The incremental cost-effectiveness ratio (ICER) was calculated based on data derived from current literature. Multiple univariate sensitivity analysis was carried out to evaluate influences of varying costs of RARC and ORC on the ICER.ResultsThe ICER for RARC compared to ORC resulted in £25,536/QALY. At the lower threshold of £20,000/QALY, RARC resulted in a negative NMB (£-4,843.32) and at the upper threshold of £30,000/QALY, a positive NMB (£624.61) compared to ORC. Threshold analysis showed that the intervention costs of £13,497 and £14,403 are met at the lower and upper threshold respectively. The univariate sensitivity analysis showed that the intervention costs of RARC or ORC, and the probabilities of complications, had the greatest impact on the ICER.ConclusionAs the resultant ICER did not fall below the £20,000/QALY threshold, our study did not provide a definitive recommendation for RARC for bladder cancer. Negative values for the NMB at the lower threshold indicated the intervention was not feasible from a cost perspective. At the upper threshold of £30,000/QALY, this situation was reversed. The intervention became cost-effective. Therefore, further research is needed to justify the intervention.
- Published
- 2022
- Full Text
- View/download PDF
13. Comparison of the Structure and Diversity of Root-Associated and Soil Microbial Communities Between Acacia Plantations and Native Tropical Mountain Forests
- Author
-
Kozue Sawada, Shinichi Watanabe, Ho Lam Nguyen, Soh Sugihara, Mayuko Seki, Hana Kobayashi, Koki Toyota, and Shinya Funakawa
- Subjects
acacia plantations ,FUNGuild ,microbial diversity ,Oxisols ,root microbiome ,tropical mountain forests ,Microbiology ,QR1-502 - Abstract
Deforestation of native tropical forests has occurred extensively over several decades. The plantation of fast-growing trees, such as Acacia spp., is expanding rapidly in tropical regions, which can contribute to conserve the remaining native tropical forests. To better understand belowground biogeochemical cycles and the sustainable productivity of acacia plantations, we assessed the effects of vegetation (acacia plantations vs. native forests) and soil types (Oxisols vs. Ultisols) on soil properties, including the diversity and community structures of bacteria- and fungi-colonizing surface and subsurface roots and soil in the Central Highlands of Vietnam. The results in surface soil showed that pH was significantly higher in acacia than in native for Oxisols but not for Ultisols, while exchangeable Al was significantly lower in acacia than in native for Ultisols but not for Oxisols. Bacterial alpha diversity (especially within phylum Chloroflexi) was higher in acacia than in native only for Oxisols but not for Ultisols, which was the same statistical result as soil pH but not exchangeable Al. These results suggest that soil pH, but not exchangeable Al, can be the critical factor to determine bacterial diversity. Acacia tree roots supported greater proportions of copiotrophic bacteria, which may support lower contents of soil inorganic N, compared with native tree roots for both Oxisols and Ultisols. Acacia tree roots also supported greater proportions of plant pathogenic Mycoleptodiscus sp. but appeared to reduce the abundances and diversity of beneficial ECM fungi compared with native tree roots regardless of soil types. Such changes in fungal community structures may threaten the sustainable productivity of acacia plantations in the future.
- Published
- 2021
- Full Text
- View/download PDF
14. Researchy Questions: A Dataset of Multi-Perspective, Decompositional Questions for LLM Web Agents
- Author
-
Rosset, Corby, Chung, Ho-Lam, Qin, Guanghui, Chau, Ethan C., Feng, Zhuo, Awadallah, Ahmed, Neville, Jennifer, and Rao, Nikhil
- Subjects
Computer Science - Computation and Language ,Computer Science - Artificial Intelligence - Abstract
Existing question answering (QA) datasets are no longer challenging to most powerful Large Language Models (LLMs). Traditional QA benchmarks like TriviaQA, NaturalQuestions, ELI5 and HotpotQA mainly study ``known unknowns'' with clear indications of both what information is missing, and how to find it to answer the question. Hence, good performance on these benchmarks provides a false sense of security. A yet unmet need of the NLP community is a bank of non-factoid, multi-perspective questions involving a great deal of unclear information needs, i.e. ``unknown uknowns''. We claim we can find such questions in search engine logs, which is surprising because most question-intent queries are indeed factoid. We present Researchy Questions, a dataset of search engine queries tediously filtered to be non-factoid, ``decompositional'' and multi-perspective. We show that users spend a lot of ``effort'' on these questions in terms of signals like clicks and session length, and that they are also challenging for GPT-4. We also show that ``slow thinking'' answering techniques, like decomposition into sub-questions shows benefit over answering directly. We release $\sim$ 100k Researchy Questions, along with the Clueweb22 URLs that were clicked.
- Published
- 2024
15. Towards audio language modeling -- an overview
- Author
-
Wu, Haibin, Chen, Xuanjun, Lin, Yi-Cheng, Chang, Kai-wei, Chung, Ho-Lam, Liu, Alexander H., and Lee, Hung-yi
- Subjects
Electrical Engineering and Systems Science - Audio and Speech Processing ,Computer Science - Sound - Abstract
Neural audio codecs are initially introduced to compress audio data into compact codes to reduce transmission latency. Researchers recently discovered the potential of codecs as suitable tokenizers for converting continuous audio into discrete codes, which can be employed to develop audio language models (LMs). Numerous high-performance neural audio codecs and codec-based LMs have been developed. The paper aims to provide a thorough and systematic overview of the neural audio codec models and codec-based LMs.
- Published
- 2024
16. Codec-SUPERB: An In-Depth Analysis of Sound Codec Models
- Author
-
Wu, Haibin, Chung, Ho-Lam, Lin, Yi-Cheng, Wu, Yuan-Kuei, Chen, Xuanjun, Pai, Yu-Chi, Wang, Hsiu-Hsuan, Chang, Kai-Wei, Liu, Alexander H., and Lee, Hung-yi
- Subjects
Electrical Engineering and Systems Science - Audio and Speech Processing ,Computer Science - Sound - Abstract
The sound codec's dual roles in minimizing data transmission latency and serving as tokenizers underscore its critical importance. Recent years have witnessed significant developments in codec models. The ideal sound codec should preserve content, paralinguistics, speakers, and audio information. However, the question of which codec achieves optimal sound information preservation remains unanswered, as in different papers, models are evaluated on their selected experimental settings. This study introduces Codec-SUPERB, an acronym for Codec sound processing Universal PERformance Benchmark. It is an ecosystem designed to assess codec models across representative sound applications and signal-level metrics rooted in sound domain knowledge.Codec-SUPERB simplifies result sharing through an online leaderboard, promoting collaboration within a community-driven benchmark database, thereby stimulating new development cycles for codecs. Furthermore, we undertake an in-depth analysis to offer insights into codec models from both application and signal perspectives, diverging from previous codec papers mainly concentrating on signal-level comparisons. Finally, we will release codes, the leaderboard, and data to accelerate progress within the community., Comment: Github: https://github.com/voidful/Codec-SUPERB
- Published
- 2024
17. Verification of Finger Joint Stiffness Estimation Method With Soft Robotic Actuator
- Author
-
Xiang Qian Shi, Ho Lam Heung, Zhi Qiang Tang, Kai Yu Tong, and Zheng Li
- Subjects
soft-elastic composite actuator (SECA) ,SECA-finger modeling ,passive joint stiffness ,metacarpophalangeal joint ,stroke ,spasticity ,Biotechnology ,TP248.13-248.65 - Abstract
Stroke has been the leading cause of disability due to the induced spasticity in the upper extremity. The constant flexion of spastic fingers following stroke has not been well described. Accurate measurements for joint stiffness help clinicians have a better access to the level of impairment after stroke. Previously, we conducted a method for quantifying the passive finger joint stiffness based on the pressure-angle relationship between the spastic fingers and the soft-elastic composite actuator (SECA). However, it lacks a ground-truth to demonstrate the compatibility between the SECA-facilitated stiffness estimation and standard joint stiffness quantification procedure. In this study, we compare the passive metacarpophalangeal (MCP) joint stiffness measured using the SECA with the results from our designed standalone mechatronics device, which measures the passive metacarpophalangeal joint torque and angle during passive finger rotation. Results obtained from the fitting model that concludes the stiffness characteristic are further compared with the results obtained from SECA-Finger model, as well as the clinical score of Modified Ashworth Scale (MAS) for grading spasticity. These findings suggest the possibility of passive MCP joint stiffness quantification using the soft robotic actuator during the performance of different tasks in hand rehabilitation.
- Published
- 2020
- Full Text
- View/download PDF
18. Protocol, rationale and design of DAbigatran for Stroke PreVention In Atrial Fibrillation in MoDerate or Severe Mitral Stenosis (DAVID-MS): a randomised, open-label study
- Author
-
Andrew Ng, Duo Huang, Ning Tan, Chung Wah Siu, Jo Jo Hai, Chun Ka Wong, Mi Zhou, Yuk Ming LAU, Chor Cheung Frankie Tam, Yiu Tung Anthony Wong, See Yue Arthur Yung, Ki Wan Kelvin Chan, Yingqing Feng, Chi Yui Yung, Kwok Lun Lee, Chun Wai Choi, Ho Lam, Katherine Fan, Man Hong Jim, Kai Hang Yiu, and Bryan P. Yan
- Subjects
Medicine - Published
- 2020
- Full Text
- View/download PDF
19. Soft Rehabilitation Actuator With Integrated Post-stroke Finger Spasticity Evaluation
- Author
-
Ho Lam Heung, Zhi Qiang Tang, Xiang Qian Shi, Kai Yu Tong, and Zheng Li
- Subjects
stroke ,finite element method ,finger spasticity ,soft-elastic composite actuator ,elastomer 3D printing ,Biotechnology ,TP248.13-248.65 - Abstract
Strokes cause severe impairment of hand function because of the spasticity in the affected upper extremities. Proper spasticity evaluation is critical to facilitate neural plasticity for rehabilitation after stroke. However, existing methods for measuring spasticity, e.g. Modified Ashworth Scale (MAS), highly depends on clinicians’ experiences, which are subjective and lacks quantitative details. Here, we introduce the first rehabilitation actuator that objectively reflects the condition of post-stroke finger spasticity. The actuator is 3D printed with soft materials. By considering the finger and the actuator together, the spasticity, i.e. stiffness, in finger is obtained from the pressure–angle relationship. The method is validated by simulations using finite element analysis (FEA) and experiments on mannequin fingers. Furthermore, it is examined on four stroke subjects and four healthy subjects. Results show the finger stiffness increases significantly from healthy subjects to stroke subjects, particularly those with high MAS score. For patients with the same MAS score, stiffness variation can be a few times. With this soft actuator, a hand rehabilitation robot that may tell the therapeutic progress during the rehabilitation training is readily available.
- Published
- 2020
- Full Text
- View/download PDF
20. PDGFRA defines the mesenchymal stem cell Kaposi's sarcoma progenitors by enabling KSHV oncogenesis in an angiogenic environment.
- Author
-
Julian Naipauer, Santas Rosario, Sachin Gupta, Courtney Premer, Omayra Méndez-Solís, Mariana Schlesinger, Virginia Ponzinibbio, Vaibhav Jain, Lauren Gay, Rolf Renne, Ho Lam Chan, Lluis Morey, Daria Salyakina, Martin Abba, Sion Williams, Joshua M Hare, Pascal J Goldschmidt-Clermont, and Enrique A Mesri
- Subjects
Immunologic diseases. Allergy ,RC581-607 ,Biology (General) ,QH301-705.5 - Abstract
Kaposi's sarcoma (KS) is an AIDS-defining cancer caused by the KS-associated herpesvirus (KSHV). Unanswered questions regarding KS are its cellular ontology and the conditions conducive to viral oncogenesis. We identify PDGFRA(+)/SCA-1(+) bone marrow-derived mesenchymal stem cells (Pα(+)S MSCs) as KS spindle-cell progenitors and found that pro-angiogenic environmental conditions typical of KS are critical for KSHV sarcomagenesis. This is because growth in KS-like conditions generates a de-repressed KSHV epigenome allowing oncogenic KSHV gene expression in infected Pα(+)S MSCs. Furthermore, these growth conditions allow KSHV-infected Pα(+)S MSCs to overcome KSHV-driven oncogene-induced senescence and cell cycle arrest via a PDGFRA-signaling mechanism; thus identifying PDGFRA not only as a phenotypic determinant for KS-progenitors but also as a critical enabler for viral oncogenesis.
- Published
- 2019
- Full Text
- View/download PDF
21. Polycomb complexes associate with enhancers and promote oncogenic transcriptional programs in cancer through multiple mechanisms
- Author
-
Ho Lam Chan, Felipe Beckedorff, Yusheng Zhang, Jenaro Garcia-Huidobro, Hua Jiang, Antonio Colaprico, Daniel Bilbao, Maria E. Figueroa, John LaCava, Ramin Shiekhattar, and Lluis Morey
- Subjects
Science - Abstract
The role of Polycomb Repressive Complex 1 (PRC1) is well described in development. Here, the authors investigate canonical PRC1’s regulation of transcriptional programs in breast cancer where, in addition to its repressive function, it is also recruited to oncogenic active enhancers to regulate enhancer activity and chromatin accessibility.
- Published
- 2018
- Full Text
- View/download PDF
22. Detection of evolutionary shifts in variance under an Ornsten-Uhlenbeck model
- Author
-
Zhang, Wensha, Ho, Lam Si Tung, and Kenney, Toby
- Subjects
Quantitative Biology - Populations and Evolution ,Statistics - Methodology - Abstract
1. Abrupt environmental changes can lead to evolutionary shifts in not only mean (optimal value), but also variance of descendants in trait evolution. There are some methods to detect shifts in optimal value but few studies consider shifts in variance. 2. We use a multi-optima and multi-variance OU process model to describe the trait evolution process with shifts in both optimal value and variance and provide analysis of how the covariance between species changes when shifts in variance occur along the path. 3. We propose a new method to detect the shifts in both variance and optimal values based on minimizing the loss function with L1 penalty. We implement our method in a new R package, ShiVa (Detection of evolutionary shifts in variance). 4. We conduct simulations to compare our method with the two methods considering only shifts in optimal values (l1ou; PhylogeneticEM). Our method shows strength in predictive ability and includes far fewer false positive shifts in optimal value compared to other methods when shifts in variance actually exist. When there are only shifts in optimal value, our method performs similarly to other methods. We applied our method to the cordylid data, ShiVa outperformed l1ou and phyloEM, exhibiting the highest log-likelihood and lowest BIC.
- Published
- 2023
23. GSQA: An End-to-End Model for Generative Spoken Question Answering
- Author
-
Shih, Min-Han, Chung, Ho-Lam, Pai, Yu-Chi, Hsu, Ming-Hao, Lin, Guan-Ting, Li, Shang-Wen, and Lee, Hung-yi
- Subjects
Computer Science - Computation and Language ,Computer Science - Artificial Intelligence - Abstract
In recent advancements in spoken question answering (QA), end-to-end models have made significant strides. However, previous research has primarily focused on extractive span selection. While this extractive-based approach is effective when answers are present directly within the input, it falls short in addressing abstractive questions, where answers are not directly extracted but inferred from the given information. To bridge this gap, we introduce the first end-to-end Generative Spoken Question Answering (GSQA) model that empowers the system to engage in abstractive reasoning. The challenge in training our GSQA model lies in the absence of a spoken abstractive QA dataset. We propose using text models for initialization and leveraging the extractive QA dataset to transfer knowledge from the text generative model to the spoken generative model. Experimental results indicate that our model surpasses the previous extractive model by 3% on extractive QA datasets. Furthermore, the GSQA model has only been fine-tuned on the spoken extractive QA dataset. Despite not having seen any spoken abstractive QA data, it can still closely match the performance of the cascade model. In conclusion, our GSQA model shows the potential to generalize to a broad spectrum of questions, thus further expanding the spoken question answering capabilities of abstractive QA. Our code is available at https://voidful.github.io/GSQA, Comment: 5 pages, 2 figures, Interspeech 2024
- Published
- 2023
24. Simple Transferability Estimation for Regression Tasks
- Author
-
Nguyen, Cuong N., Tran, Phong, Ho, Lam Si Tung, Dinh, Vu, Tran, Anh T., Hassner, Tal, and Nguyen, Cuong V.
- Subjects
Computer Science - Machine Learning ,Computer Science - Artificial Intelligence ,Statistics - Machine Learning - Abstract
We consider transferability estimation, the problem of estimating how well deep learning models transfer from a source to a target task. We focus on regression tasks, which received little previous attention, and propose two simple and computationally efficient approaches that estimate transferability based on the negative regularized mean squared error of a linear regression model. We prove novel theoretical results connecting our approaches to the actual transferability of the optimal target models obtained from the transfer learning process. Despite their simplicity, our approaches significantly outperform existing state-of-the-art regression transferability estimators in both accuracy and efficiency. On two large-scale keypoint regression benchmarks, our approaches yield 12% to 36% better results on average while being at least 27% faster than previous state-of-the-art methods., Comment: Paper published at The 39th Conference on Uncertainty in Artificial Intelligence (UAI) 2023
- Published
- 2023
25. Cumulative Damage: Cell Death in Posthemorrhagic Hydrocephalus of Prematurity
- Author
-
Riley Sevensky, Jessie C. Newville, Ho Lam Tang, Shenandoah Robinson, and Lauren L. Jantzie
- Subjects
cell death ,encephalopathy of prematurity ,hydrocephalus ,preterm birth ,neurodevelopment ,inflammation ,Cytology ,QH573-671 - Abstract
Globally, approximately 11% of all infants are born preterm, prior to 37 weeks’ gestation. In these high-risk neonates, encephalopathy of prematurity (EoP) is a major cause of both morbidity and mortality, especially for neonates who are born very preterm (
- Published
- 2021
- Full Text
- View/download PDF
26. IGF2BP3 functions as a potential oncogene and is a crucial target of miR-34a in gastric carcinogenesis
- Author
-
Yuhang Zhou, Tingting Huang, Ho Lam Siu, Chi Chun Wong, Yujuan Dong, Feng Wu, Bin Zhang, William K. K. Wu, Alfred S. L. Cheng, Jun Yu, Ka Fai To, and Wei Kang
- Subjects
Gastric cancer ,IGF2BP3 ,miR-34a ,Neoplasms. Tumors. Oncology. Including cancer and carcinogens ,RC254-282 - Abstract
Abstract Background Gastric cancer (GC) is one of the frequent causes of cancer-related death in eastern Asian population. IGF2BP2 lists in the top rank up-regulated genes in GC, but its functional role is unclear. Method The expression of IGF2BP3 in GC cell lines and primary samples was examined by qRT-PCR and Western blot. The biological role of IGF2BP3 was revealed by a series of functional in vitro studies. Its regulation by microRNAs (miRNAs) was predicted by TargetScan and confirmed by luciferase assays and rescue experiments. Results IGF2BP3 ranked the No.1 of the up-regulated genes by expression microarray analysis in GC cell lines. The expression level of IGF2BP3 was observed in GC tissues comparing with non-tumorous gastric epitheliums. The up-regulated IGF2BP3 expression was associated with poor disease specific survival. IGF2BP3 knockdown significantly inhibited cell proliferation and invasion. Apart from copy number gain, IGF2BP3 has been confirmed to be negatively regulated by tumor-suppressive miRNA, namely miR-34a. The expression of miR-34a showed negative correlation with IGF2BP3 mRNA expression in primary GC samples and more importantly, re-overexpression of IGF2BP3 rescued the inhibitory effect of miR-34a. Conclusion We compressively revealed the oncogenic role of IGF2BP3 in gastric tumorigenesis and confirmed its activation is partly due to the silence of miR-34a. Our findings identified useful prognostic biomarker and provided clinical translational potential.
- Published
- 2017
- Full Text
- View/download PDF
27. Cell recovery by reversal of ferroptosis
- Author
-
Ho Man Tang and Ho Lam Tang
- Subjects
Anastasis ,Ferrostatin-1 ,Glutamate ,Glutathione ,Reversal of apoptosis ,Reversal of ferroptosis ,Science ,Biology (General) ,QH301-705.5 - Abstract
The classical view of cell death has long assumed that, once initiated, the dying process is irreversible. However, recent studies reveal that recovery of dying cells can actually occur, even after initiation of a cell suicide process called apoptosis. This discovery raised fundamental key questions about which forms of the cell death process could be reversible and how reversal is mediated. Here, we uncover an unanticipated reversibility of ferroptotic cell death process. Unlike apoptosis reversal, removal of ferroptosis inducers, such as erastin and glutamate, is insufficient to allow ferroptotic dying cells to escape the cell death process. However, by removing the cell death inducer and providing the reduced form of glutathione or the radical-trapping antioxidant ferrostatin-1, ferroptotic dying cells can be rescued and promoted to recover. Interestingly, although ferroptotic inhibitors such as aminooxyacetic acid, deferoxamine, dopamine and vitamin C can prevent initiation of ferroptosis, added alone they are unable to reverse the initiated ferroptosis, suggesting regulatory distinctions between preventing and reversing ferroptosis. Together, these results reveal the first evidence that ferroptosis is reversible and suggest strategies to enhance its reversibility, thereby providing a useful model for studying the physiological, pathological and therapeutic potentials of this cell recovery process.
- Published
- 2019
- Full Text
- View/download PDF
28. Odour conditioning of positive affective states: Rats can learn to associate an odour with being tickled.
- Author
-
Vincent Bombail, Nathalie Jerôme, Ho Lam, Sacha Muszlak, Simone L Meddle, Alistair B Lawrence, and Birte L Nielsen
- Subjects
Medicine ,Science - Abstract
Most associative learning tests in rodents use negative stimuli, such as electric shocks. We investigated if young rats can learn to associate the presence of an odour with the experience of being tickled (i.e. using an experimenter's hand to mimic rough-and-tumble play), shown to elicit 50 kHz ultrasonic vocalisations (USVs), which are indicative of positive affect. Male, pair-housed Wistar rats (N = 24) were all exposed to two neutral odours (A and B) presented in a perforated container on alternate days in a test arena. Following 60s of exposure, the rats were either tickled on days when odour A (n = 8) or odour B (n = 8) was present, or never tickled (n = 8). When tickled, rats produced significantly more 50 kHz USVs compared to the days when not being tickled, and compared to control rats. The level of anticipatory 50 kHz USVs in the 60s prior to tickling did not differ significantly between the tickled and control rats. As a retrieval test following the odour conditioning, rats were exposed successively in the same arena to three odours: an unknown neutral odour, extract of fox faeces, and either odours A or B. Compared to controls, 50 kHz USVs of tickled rats increased when exposed to the odour they had previously experienced when tickled, indicating that these rats had learned to associate the odour with the positive experience of being tickled. In a test with free access for 5 min to both arms of a T-maze, each containing one of the odours, rats tickled with odour A spent more time in the arm with this odour. This work is the first to test in a fully balanced design whether rats can learn to associate an odour with tickling, and indicates that positive odour conditioning has potential to be used as an alternative to negative conditioning tests.
- Published
- 2019
- Full Text
- View/download PDF
29. Mycobacterium tuberculosis-Associated Necrotizing Pneumonia With Adjunctive Corticosteroid Therapy
- Author
-
Ho Lam Nguyen, Huyen Duong Thanh, Thuong Vu Le, and Ngoc Tran Van
- Subjects
Infectious and parasitic diseases ,RC109-216 - Abstract
Necrotizing pneumonia induced by Mycobacterium tuberculosis is a rare but severe condition. It is difficult to distinguish between M. tuberculosis-associated and bacterial necrotizing pneumonia. The optimal treatment for this condition is controversial. Here, we report a case of M. tuberculosis-associated necrotizing pneumonia treated with the adjunctive corticosteroid and the antituberculosis drugs.
- Published
- 2019
- Full Text
- View/download PDF
30. Real-Time Location-Positioning Technologies for Managing Cart Operations at a Distribution Facility
- Author
-
Chun-Hung Cheng, Yong-Hong Kuo, Ho Lam, and Matthew Petering
- Subjects
RFID ,real-time locating ,traceability ,postal services ,big data ,analytics ,Technology ,Engineering (General). Civil engineering (General) ,TA1-2040 ,Biology (General) ,QH301-705.5 ,Physics ,QC1-999 ,Chemistry ,QD1-999 - Abstract
In this paper, we propose an RFID-based location-positioning platform for managing cart operations at vast and fast-moving distribution facilities. Our work was motivated by a real-world problem in a large airmail center. Our system requires that each cart is affixed with an active RFID tag, and RFID readers are installed at multiple locations of the facility. The locations of the tagged carts are determined by the estimated angles of arrival communication signals between the readers and the tags and the signal strengths, and the information about their positions are stored in the platform continuously. This platform enables the users to locate tagged objects in real time, thereby saving staff members a significant amount of time from searching for the right carts in the highly dynamic environment. Consequently, the mail facility can benefit from the increased efficiency. Furthermore, the system is also able to reduce the chance of having a delay in delivering mail bags to unit load device (ULD) area and misplacements of mail bags to ULDs. We further analyze the the huge amount of historical data collected from the RFID infrastructure of the airmail center for the cart movements within the facility and observed previously unrecognized operational issues. We also discuss some of the challenges and problems faced in this project. We believe that this application can bring huge benefits to businesses and organizations along supply chains by effectively anticipating the increasing demand for logistics services in the age of electronic retailing.
- Published
- 2021
- Full Text
- View/download PDF
31. Findings of the 2023 ML-SUPERB Challenge: Pre-Training and Evaluation over More Languages and Beyond
- Author
-
Shi, Jiatong, Chen, William, Berrebbi, Dan, Wang, Hsiu-Hsuan, Huang, Wei-Ping, Hu, En-Pei, Chuang, Ho-Lam, Chang, Xuankai, Tang, Yuxun, Li, Shang-Wen, Mohamed, Abdelrahman, Lee, Hung-yi, and Watanabe, Shinji
- Subjects
Computer Science - Sound ,Computer Science - Computation and Language ,Electrical Engineering and Systems Science - Audio and Speech Processing - Abstract
The 2023 Multilingual Speech Universal Performance Benchmark (ML-SUPERB) Challenge expands upon the acclaimed SUPERB framework, emphasizing self-supervised models in multilingual speech recognition and language identification. The challenge comprises a research track focused on applying ML-SUPERB to specific multilingual subjects, a Challenge Track for model submissions, and a New Language Track where language resource researchers can contribute and evaluate their low-resource language data in the context of the latest progress in multilingual speech recognition. The challenge garnered 12 model submissions and 54 language corpora, resulting in a comprehensive benchmark encompassing 154 languages. The findings indicate that merely scaling models is not the definitive solution for multilingual speech tasks, and a variety of speech/voice types present significant challenges in multilingual speech processing., Comment: Accepted by ASRU
- Published
- 2023
32. A Generalization Bound of Deep Neural Networks for Dependent Data
- Author
-
Do, Quan Huu, Nguyen, Binh T., and Ho, Lam Si Tung
- Subjects
Statistics - Machine Learning ,Computer Science - Machine Learning - Abstract
Existing generalization bounds for deep neural networks require data to be independent and identically distributed (iid). This assumption may not hold in real-life applications such as evolutionary biology, infectious disease epidemiology, and stock price prediction. This work establishes a generalization bound of feed-forward neural networks for non-stationary $\phi$-mixing data.
- Published
- 2023
33. Towards General-Purpose Text-Instruction-Guided Voice Conversion
- Author
-
Kuan, Chun-Yi, Li, Chen An, Hsu, Tsu-Yuan, Lin, Tse-Yang, Chung, Ho-Lam, Chang, Kai-Wei, Chang, Shuo-yiin, and Lee, Hung-yi
- Subjects
Electrical Engineering and Systems Science - Audio and Speech Processing ,Computer Science - Computation and Language ,Computer Science - Machine Learning ,Computer Science - Sound - Abstract
This paper introduces a novel voice conversion (VC) model, guided by text instructions such as "articulate slowly with a deep tone" or "speak in a cheerful boyish voice". Unlike traditional methods that rely on reference utterances to determine the attributes of the converted speech, our model adds versatility and specificity to voice conversion. The proposed VC model is a neural codec language model which processes a sequence of discrete codes, resulting in the code sequence of converted speech. It utilizes text instructions as style prompts to modify the prosody and emotional information of the given speech. In contrast to previous approaches, which often rely on employing separate encoders like prosody and content encoders to handle different aspects of the source speech, our model handles various information of speech in an end-to-end manner. Experiments have demonstrated the impressive capabilities of our model in comprehending instructions and delivering reasonable results., Comment: Accepted to ASRU 2023
- Published
- 2023
34. Rationale and design of the screening of pulmonary hypertension in systemic lupus erythematosus (SOPHIE) study
- Author
-
Duo Huang, Yang-Yang Cheng, Pak-Hei Chan, Jojo Hai, Kai-Hang Yiu, Hung-Fat Tse, Ka-Lam Wong, Katherine Fan, Ying Wah Li, Woon-Leung Ng, Cheuk-Wan Yim, Cheuk-hon John Wong, Lai-Shan Tam, Priscilla C.H. Wong, Chi-Yuen Wong, Chup-Hei Ho, Alexander M.H. Leung, Chi-Chiu Mok, Ho Lam, Chak-Sing Lau, Tommy Cheung, Carmen Ho, Sharon W.Y. Law, Esther W. Chan, Li-Xue Yin, Wen-Sheng Yue, Toi Meng Mok, Mario Alberto Evora, and Chung-Wah Siu
- Subjects
Medicine - Abstract
Current guideline-recommended screening for pulmonary hypertension in patients with systemic sclerosis has not been evaluated in systemic lupus erythematosus (SLE), which is disproportionately prevalent in Asians. This multicentre, cross-sectional screening study aims to study the prevalence of pulmonary hypertension among SLE patients using these guidelines, and identify independent predictors and develop a prediction model for pulmonary hypertension in SLE patients. SLE patients from participating centres will undergo an echocardiography- and biomarker-based pulmonary hypertension screening procedure as in the DETECT study. Standard right heart catheterisation will be provided to patients with intermediate or high echocardiographic probability of pulmonary hypertension. Those with low echocardiographic probability will rescreen within 1 year. The primary measure will be the diagnosis and types of pulmonary hypertension and prevalence of pulmonary hypertension in SLE patients. The secondary measures will be the predictors and prediction models for pulmonary hypertension in SLE patients. The estimated sample size is approximately 895 participants. The results of the SOPHIE study will be an important contribution to the literature of SLE-related pulmonary hypertension and may be immediately translatable to real clinical practice. Ultimately, this study will provide the necessary evidence for establishing universal guidelines for screening of pulmonary hypertension in SLE patients.
- Published
- 2018
- Full Text
- View/download PDF
35. Correction to: ‘Anastasis: recovery from the brink of cell death’
- Author
-
Ho Man Tang and Ho Lam Tang
- Subjects
Science - Published
- 2018
- Full Text
- View/download PDF
36. Anastasis: recovery from the brink of cell death
- Author
-
Ho Man Tang and Ho Lam Tang
- Subjects
anastasis ,apoptosis ,mutagenesis ,programmed cell death ,reversal of apoptosis ,reversal of cell death process ,Science - Abstract
Anastasis is a natural cell recovery phenomenon that rescues cells from the brink of death. Programmed cell death such as apoptosis has been traditionally assumed to be an intrinsically irreversible cascade that commits cells to a rapid and massive demolition. Interestingly, recent studies have demonstrated recovery of dying cells even at the late stages generally considered immutable. Here, we examine the evidence for anastasis in cultured cells and in animals, review findings illuminating the potential mechanisms of action, discuss the challenges of studying anastasis and explore new strategies to uncover the function and regulation of anastasis, the identification of which has wide-ranging physiological, pathological and therapeutic implications.
- Published
- 2018
- Full Text
- View/download PDF
37. ML-SUPERB: Multilingual Speech Universal PERformance Benchmark
- Author
-
Shi, Jiatong, Berrebbi, Dan, Chen, William, Chung, Ho-Lam, Hu, En-Pei, Huang, Wei Ping, Chang, Xuankai, Li, Shang-Wen, Mohamed, Abdelrahman, Lee, Hung-yi, and Watanabe, Shinji
- Subjects
Computer Science - Sound ,Computer Science - Computation and Language ,Electrical Engineering and Systems Science - Audio and Speech Processing - Abstract
Speech processing Universal PERformance Benchmark (SUPERB) is a leaderboard to benchmark the performance of Self-Supervised Learning (SSL) models on various speech processing tasks. However, SUPERB largely considers English speech in its evaluation. This paper presents multilingual SUPERB (ML-SUPERB), covering 143 languages (ranging from high-resource to endangered), and considering both automatic speech recognition and language identification. Following the concept of SUPERB, ML-SUPERB utilizes frozen SSL features and employs a simple framework for multilingual tasks by learning a shallow downstream model. Similar to the SUPERB benchmark, we find speech SSL models can significantly improve performance compared to FBANK features. Furthermore, we find that multilingual models do not always perform better than their monolingual counterparts. We will release ML-SUPERB as a challenge with organized datasets and reproducible training scripts for future multilingual representation research., Comment: Accepted by Interspeech
- Published
- 2023
38. Determination of ITS1 haplotypes of Fritillariae Cirrhosae Bulbus by amplicon sequencing
- Author
-
Wu, Hoi-Yan, Wong, Ka-Lok, Law, Sean Tsz-Sum, Nong, Wenyang, Chan, Kwun-Tin, Hui, Jerome Ho-Lam, Lin, Ge, Chan, Wing-Han, and Shaw, Pang-Chui
- Published
- 2024
- Full Text
- View/download PDF
39. Evolutionary shift detection with ensemble variable selection
- Author
-
Zhang, Wensha, Kenney, Toby, and Ho, Lam Si Tung
- Published
- 2024
- Full Text
- View/download PDF
40. Cytochrome oxidase I DNA barcodes of crocodilians meat selling in Hong Kong
- Author
-
So, Wai Lok, Chong, Tze Kiu, Lee, Ivy Hoi Ting, So, Miu Tsz Wai, Liu, Avis Mang Yi, Leung, Sam Tsz Chung, Ching, Wai, Yip, Ho Yin, Shaw, Pang Chui, and Hui, Jerome Ho Lam
- Published
- 2024
- Full Text
- View/download PDF
41. Emerging treatments for corneal endothelium decompensation — a systematic review
- Author
-
Cheong, Noel, Chui, Siu Wa, Poon, Stephanie Hiu Ling, Wong, Ho Lam, Shih, Kendrick Co, and Chan, Yau Kei
- Published
- 2024
- Full Text
- View/download PDF
42. Molecular signature of anastasis for reversal of apoptosis [version 2; referees: 2 approved]
- Author
-
Ho Man Tang, C. Conover Talbot Jr, Ming Chiu Fung, and Ho Lam Tang
- Subjects
Cellular Death & Stress Responses ,Genomics ,Medicine ,Science - Abstract
Anastasis (Greek for "rising to life") is a cell recovery phenomenon that rescues dying cells from the brink of cell death. We recently discovered anastasis to occur after the execution-stage of apoptosis in vitro and in vivo. Promoting anastasis could in principle preserve injured cells that are difficult to replace, such as cardiomyocytes and neurons. Conversely, arresting anastasis in dying cancer cells after cancer therapies could improve treatment efficacy. To develop new therapies that promote or inhibit anastasis, it is essential to identify the key regulators and mediators of anastasis – the therapeutic targets. Therefore, we performed time-course microarray analysis to explore the molecular mechanisms of anastasis during reversal of ethanol-induced apoptosis in mouse primary liver cells. We found striking changes in transcription of genes involved in multiple pathways, including early activation of pro-cell survival, anti-oxidation, cell cycle arrest, histone modification, DNA-damage and stress-inducible responses, and at delayed times, angiogenesis and cell migration. Validation with RT-PCR confirmed similar changes in the human liver cancer cell line, HepG2, during anastasis. Here, we present the time-course whole-genome gene expression dataset revealing gene expression profiles during the reversal of apoptosis. This dataset provides important insights into the physiological, pathological, and therapeutic implications of anastasis.
- Published
- 2017
- Full Text
- View/download PDF
43. SPADE4: Sparsity and Delay Embedding based Forecasting of Epidemics
- Author
-
Saha, Esha, Ho, Lam Si Tung, and Tran, Giang
- Subjects
Computer Science - Machine Learning ,Physics - Physics and Society ,Quantitative Biology - Populations and Evolution - Abstract
Predicting the evolution of diseases is challenging, especially when the data availability is scarce and incomplete. The most popular tools for modelling and predicting infectious disease epidemics are compartmental models. They stratify the population into compartments according to health status and model the dynamics of these compartments using dynamical systems. However, these predefined systems may not capture the true dynamics of the epidemic due to the complexity of the disease transmission and human interactions. In order to overcome this drawback, we propose Sparsity and Delay Embedding based Forecasting (SPADE4) for predicting epidemics. SPADE4 predicts the future trajectory of an observable variable without the knowledge of the other variables or the underlying system. We use random features model with sparse regression to handle the data scarcity issue and employ Takens' delay embedding theorem to capture the nature of the underlying system from the observed variable. We show that our approach outperforms compartmental models when applied to both simulated and real data., Comment: 24 pages, 13 figures, 2 tables
- Published
- 2022
- Full Text
- View/download PDF
44. T5lephone: Bridging Speech and Text Self-supervised Models for Spoken Language Understanding via Phoneme level T5
- Author
-
Hsu, Chan-Jan, Chung, Ho-Lam, Lee, Hung-yi, and Tsao, Yu
- Subjects
Computer Science - Computation and Language ,Computer Science - Sound ,Electrical Engineering and Systems Science - Audio and Speech Processing - Abstract
In Spoken language understanding (SLU), a natural solution is concatenating pre-trained speech models (e.g. HuBERT) and pretrained language models (PLM, e.g. T5). Most previous works use pretrained language models with subword-based tokenization. However, the granularity of input units affects the alignment of speech model outputs and language model inputs, and PLM with character-based tokenization is underexplored. In this work, we conduct extensive studies on how PLMs with different tokenization strategies affect spoken language understanding task including spoken question answering (SQA) and speech translation (ST). We further extend the idea to create T5lephone(pronounced as telephone), a variant of T5 that is pretrained using phonemicized text. We initialize T5lephone with existing PLMs to pretrain it using relatively lightweight computational resources. We reached state-of-the-art on NMSQA, and the T5lephone model exceeds T5 with other types of units on end-to-end SQA and ST.
- Published
- 2022
45. Codec-SUPERB: An In-Depth Analysis of Sound Codec Models.
- Author
-
Haibin Wu, Ho-Lam Chung, Yi-Cheng Lin, Yuan-Kuei Wu, Xuanjun Chen, Yu-Chi Pai, Hsiu-Hsuan Wang, Kai-Wei Chang, Alexander H. Liu, and Hung-yi Lee
- Published
- 2024
- Full Text
- View/download PDF
46. Generalization Bounds for Deep Transfer Learning Using Majority Predictor Accuracy
- Author
-
Nguyen, Cuong N., Ho, Lam Si Tung, Dinh, Vu, Hassner, Tal, and Nguyen, Cuong V.
- Subjects
Computer Science - Machine Learning ,Computer Science - Artificial Intelligence - Abstract
We analyze new generalization bounds for deep learning models trained by transfer learning from a source to a target task. Our bounds utilize a quantity called the majority predictor accuracy, which can be computed efficiently from data. We show that our theory is useful in practice since it implies that the majority predictor accuracy can be used as a transferability measure, a fact that is also validated by our experiments., Comment: 5 pages, Paper published at the International Symposium on Information Theory and Its Applications (ISITA 2022)
- Published
- 2022
47. When can we reconstruct the ancestral state? Beyond Brownian motion
- Author
-
Vu, Nhat L., Nguyen, Thanh P., Nguyen, Binh T., Dinh, Vu, and Ho, Lam Si Tung
- Subjects
Quantitative Biology - Populations and Evolution ,Mathematics - Statistics Theory - Abstract
Reconstructing the ancestral state of a group of species helps answer many important questions in evolutionary biology. Therefore, it is crucial to understand when we can estimate the ancestral state accurately. Previous works provide a necessary and sufficient condition, called the big bang condition, for the existence of an accurate reconstruction method under discrete trait evolution models and the Brownian motion model. In this paper, we extend this result to a wide range of continuous trait evolution models. In particular, we consider a general setting where continuous traits evolve along the tree according to stochastic processes that satisfy some regularity conditions. We verify these conditions for popular continuous trait evolution models including Ornstein-Uhlenbeck, reflected Brownian Motion, and Cox-Ingersoll-Ross.
- Published
- 2022
48. Handover QG: Question Generation by Decoder Fusion and Reinforcement Learning.
- Author
-
Ho-Lam Chung, Ying-Hong Chan, and Yao-Chung Fan
- Published
- 2024
- Full Text
- View/download PDF
49. Evolutionary shift detection with ensemble variable selection
- Author
-
Zhang, Wensha, Kenney, Toby, and Ho, Lam Si Tung
- Subjects
Quantitative Biology - Populations and Evolution ,Statistics - Methodology - Abstract
1. Abrupt environmental changes can lead to evolutionary shifts in trait evolution. Identifying these shifts is an important step in understanding the evolutionary history of phenotypes. 2. We propose an ensemble variable selection method (R package ELPASO) for the evolutionary shift detection task and compare it with existing methods (R packages l1ou and PhylogeneticEM) under several scenarios. 3. The performances of methods are highly dependent on the selection criterion. When the signal sizes are small, the methods using the Bayesian information criterion (BIC) have better performances. And when the signal sizes are large enough, the methods using the phylogenetic Bayesian information criterion (pBIC) (Khabbazian et al., 2016) have better performance. Moreover, the performance is heavily impacted by measurement error and tree reconstruction error. 4. Ensemble method + pBIC tends to perform less conservatively than l1ou + pBIC, and Ensemble method + BIC is more conservatively than l1ou + BIC. PhylogeneticEM is even more conservative with small signal sizes and falls between l1ou + pBIC and Ensemble method + BIC with large signal sizes. The results can differ between the methods, but none clearly outperforms the others. By applying multiple methods to a single dataset, we can access the robustness of each detected shift, based on the agreement among methods.
- Published
- 2022
50. DUAL: Discrete Spoken Unit Adaptive Learning for Textless Spoken Question Answering
- Author
-
Lin, Guan-Ting, Chuang, Yung-Sung, Chung, Ho-Lam, Yang, Shu-wen, Chen, Hsuan-Jui, Dong, Shuyan, Li, Shang-Wen, Mohamed, Abdelrahman, Lee, Hung-yi, and Lee, Lin-shan
- Subjects
Computer Science - Computation and Language ,Computer Science - Sound ,Electrical Engineering and Systems Science - Audio and Speech Processing - Abstract
Spoken Question Answering (SQA) is to find the answer from a spoken document given a question, which is crucial for personal assistants when replying to the queries from the users. Existing SQA methods all rely on Automatic Speech Recognition (ASR) transcripts. Not only does ASR need to be trained with massive annotated data that are time and cost-prohibitive to collect for low-resourced languages, but more importantly, very often the answers to the questions include name entities or out-of-vocabulary words that cannot be recognized correctly. Also, ASR aims to minimize recognition errors equally over all words, including many function words irrelevant to the SQA task. Therefore, SQA without ASR transcripts (textless) is always highly desired, although known to be very difficult. This work proposes Discrete Spoken Unit Adaptive Learning (DUAL), leveraging unlabeled data for pre-training and fine-tuned by the SQA downstream task. The time intervals of spoken answers can be directly predicted from spoken documents. We also release a new SQA benchmark corpus, NMSQA, for data with more realistic scenarios. We empirically showed that DUAL yields results comparable to those obtained by cascading ASR and text QA model and robust to real-world data. Our code and model will be open-sourced., Comment: Accepted by Interspeech 2022
- Published
- 2022
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.