Geological reports are a significant accomplishment for geologists involved in geological investigations and scientific research as they contain rich data and textual information.With the rapid development of science ...Geological reports are a significant accomplishment for geologists involved in geological investigations and scientific research as they contain rich data and textual information.With the rapid development of science and technology,a large number of textual reports have accumulated in the field of geology.However,many non-hot topics and non-English speaking regions are neglected in mainstream geoscience databases for geological information mining,making it more challenging for some researchers to extract necessary information from these texts.Natural Language Processing(NLP)has obvious advantages in processing large amounts of textual data.The objective of this paper is to identify geological named entities from Chinese geological texts using NLP techniques.We propose the RoBERTa-Prompt-Tuning-NER method,which leverages the concept of Prompt Learning and requires only a small amount of annotated data to train superior models for recognizing geological named entities in low-resource dataset configurations.The RoBERTa layer captures context-based information and longer-distance dependencies through dynamic word vectors.Finally,we conducted experiments on the constructed Geological Named Entity Recognition(GNER)dataset.Our experimental results show that the proposed model achieves the highest F1 score of 80.64%among the four baseline algorithms,demonstrating the reliability and robustness of using the model for Named Entity Recognition of geological texts.展开更多
Readability is a fundamental problem in textbooks assessment.For low resources languages(LRL),however,little investigation has been done on the readability of textbook.In this paper,we proposed a readability assessmen...Readability is a fundamental problem in textbooks assessment.For low resources languages(LRL),however,little investigation has been done on the readability of textbook.In this paper,we proposed a readability assessment method for Tibetan textbook(a low resource language).We extract features based on the information that are gotten by Tibetan segmentation and named entity recognition.Then,we calculate the correlation of different features using Pearson Correlation Coefficient and select some feature sets to design the readability formula.Fit detection,F test and T test are applied on these selected features to generate a new readability assessment formula.Experiment shows that this new formula is capable of assessing the readability of Tibetan textbooks.展开更多
The use of bilateral internal thoracic arteries to treat coronary artery disease is very less despite of improved long-term survival. In this case report</span><span style="font-family:"">,&...The use of bilateral internal thoracic arteries to treat coronary artery disease is very less despite of improved long-term survival. In this case report</span><span style="font-family:"">,</span><span style="font-family:""> a patient presented with TVD has been managed with anaortic off-pump complete arterial revascularization (OPCABG) by using composite LIMA RIMA Y Sequential 5 grafts to achieve complete arterial revascularization successfully. This type of operative procedure is technically difficult in a resource poor country like Bangladesh. However, this has been done by a group of young cardiac surgeon. During follow up</span><span style="font-family:"">,</span><span style="font-family:""> the patient is in a good condition.展开更多
Neural machine translation(NMT)has advanced with deep learning and large-scale multilingual models,yet translating lowresource languages often lacks sufficient training data and leads to hallucinations.This often resu...Neural machine translation(NMT)has advanced with deep learning and large-scale multilingual models,yet translating lowresource languages often lacks sufficient training data and leads to hallucinations.This often results in translated content that diverges significantly from the source text.This research proposes a refined Contrastive Decoding(CD)algorithm that dynamically adjusts weights of log probabilities from strong expert and weak amateur models to mitigate hallucinations in lowresource NMT and improve translation quality.Advanced large language NMT models,including ChatGLM and LLaMA,are fine-tuned and implemented for their superior contextual understanding and cross-lingual capabilities.The refined CD algorithm evaluates multiple candidate translations using BLEU score,semantic similarity,and Named Entity Recognition accuracy.Extensive experimental results show substantial improvements in translation quality and a significant reduction in hallucination rates.Fine-tuned models achieve higher evaluation metrics compared to baseline models and state-of-the-art models.An ablation study confirms the contributions of each methodological component and highlights the effectiveness of the refined CD algorithm and advanced models in mitigating hallucinations.Notably,the refined methodology increased the BLEU score by approximately 30%compared to baseline models.展开更多
Text clustering is an important task because of its vital role in NLP-related tasks.However,existing research on clustering is mainly based on the English language,with limited work on low-resource languages,such as U...Text clustering is an important task because of its vital role in NLP-related tasks.However,existing research on clustering is mainly based on the English language,with limited work on low-resource languages,such as Urdu.Low-resource language text clustering has many drawbacks in the form of limited annotated collections and strong linguistic diversity.Theprimary aim of this paper is twofold:(1)By introducing a clustering dataset namedUNC-2025 comprises 100k Urdu news documents,and(2)a detailed empirical standard of Large Language Model(LLM)improved clusteringmethods for Urdu text.We explicitly evaluate the behavior of the 11multilingual and Urdu-specific embeddings on 3 different clustering algorithms.We carefully evaluated our performance based on a set of internal and external measurements of validity.We discover the best configuration of the mBERT embedding with the HDBSCAN algorithm that attains a new state-of-the-art performance with a high score of external validity of 0.95.This new LLM method has created a new strong standard of Urdu text clustering.Importantly,the results confirm the strength and high scalability of the LLM-generated embeddings towards the ability to generalise the fine,subtle semantics needed to discover topics in low-resource settings and open the door to novel NLP applications in underrepresented languages.展开更多
Introduction: Standard procedures for surgical fixation of proximal femoral fractures (PFF) require an image intensifier which in developing countries remains a luxury. We hypothesized that, with a well-codified techn...Introduction: Standard procedures for surgical fixation of proximal femoral fractures (PFF) require an image intensifier which in developing countries remains a luxury. We hypothesized that, with a well-codified technique, the Watson Jones approach (WJA) without image intensifier nor traction table, can allow open reduction and internal fixation (ORIF) of PFF using Dynamic hip screw (DHS), with satisfactory outcome. Patients and methods: Forty one consecutive patients (mean age 59.5 ± 21.6 years, 61% males) who were followed in a Teaching Hospital for PFF treated by ORIF using the WJA and DHS from January 2016 to December 2020 were reassessed. The outcome measures were the quality of the reduction, the positioning of the implants, the tip-apex distance (TAD), the rate and delay of consolidation, the functional results using Postel Merle d’Aubigné (PMA) score, the rate of surgical site infection (SSI) and the overall mortality. Logistic regression was used to determine factors associated with mechanical failure. Results: The mean follow-up period was 33.8 ± 15.0 months. Fracture reduction was good in 31 (75.6%) cases and acceptable in 8(19.5%) cases. Implant position was fair to good in 37 (90.2%) patients. The mean TAD was 26.1 ± 3.9 mm. Three patients developed SSI. Consolidation was achieved in 38 (92.6%) patients. The functional results were good to excellent in 80.5% of patients. The overall mortality rate was 7.3%. There were an association between mechanical failure and osteoporosis (p = 0.04), fracture reduction (p = 0.003), and TAD (p = 0.025). In multivariate logistic regression, no independent factors were predictive of mechanical failure. Conclusion: This study shows that ORIF using DHS for PFF via the Watson-Jones approach without an image intensifier can give satisfactory anatomical and functional outcomes in low-resource settings. It provides and validates a reliable and reproducible technique that deserves to be diffused to surgeons in austere areas over the world.展开更多
BACKGROUND Globally,it’s estimated that at least 1 billion people have a near and/or distance vision impairment that could have been prevented or is yet to be addressed.The burden of unaddressed vision impairment and...BACKGROUND Globally,it’s estimated that at least 1 billion people have a near and/or distance vision impairment that could have been prevented or is yet to be addressed.The burden of unaddressed vision impairment and blindness is estimated to be four times higher in low and intermediate-resource settings than in high-income settings.[1]展开更多
Because of its simplicity,reliability,and replicability,the Masquelet induced membrane technique(IMT)has become one of the preferred methods for critical bone defect reconstruction in extremities.Although it is now us...Because of its simplicity,reliability,and replicability,the Masquelet induced membrane technique(IMT)has become one of the preferred methods for critical bone defect reconstruction in extremities.Although it is now used worldwide,few studies have been published about IMT in military practice.Bone reconstruction is particularly challenging in this context of care due to extensive soft-tissue injury,early wound infection,and even delayed management in austere conditions.Based on our clinical expertise,recent research,and a literature analysis,this narrative review provides an overview of the IMT application to combat-related bone defects.It presents technical specificities and future developments aiming to optimize IMT outcomes,including for the management of massive multi-tissue defects or bone reconstruction performed in the field with limited resources.展开更多
Creating a parallel corpus for machine translation is a challenging and time-consuming task,especially in a linguistically diverse country like the Philippines,with 185 languages.Although a wealth of text is available...Creating a parallel corpus for machine translation is a challenging and time-consuming task,especially in a linguistically diverse country like the Philippines,with 185 languages.Although a wealth of text is available,annotated data is scarce,particularly for languages like Bikol.Bikol is one of the major languages in the Philippines;however,its underrepresentation in the digital sphere is attributed to the absence of annotated data.This study outlines the development process of BFParCo,a proposed gold standard dataset for the Bikol and Filipino parallel corpus.The corpus underwent refinement through manual phrase alignment,translation,and evaluation.Subsequently,T5 and mT5 transformer models were fine-tuned with the parallel corpus and were evaluated using the BLEU metric.The results showed a notable improvement in Bilingual Evaluation Understudy(BLEU)score after fine-tuning,with an increase of 60.68 in BIK→FIL and 58.93 in FIL→BIK translations.Additionally,human evaluators comprehensively assessed the fine-tuned models'results using Multidimensional Quality Metrics and Scalar Quality Metrics error taxonomies.The fine-tuned models then were made publicly accessible through Hugging Face.This study represents a significant stride in advancing machine translation tools for Bikol and Filipino languages.展开更多
Low-resource text plagiarism detection faces a significant challenge due to the limited availability of labeled data for training.This task requires the development of sophisticated algorithms capable of identifying s...Low-resource text plagiarism detection faces a significant challenge due to the limited availability of labeled data for training.This task requires the development of sophisticated algorithms capable of identifying similarities and differences in texts,particularly in the realm of semantic rewriting and translation-based plagiarism detection.In this paper,we present an enhanced attentive Siamese Long Short-Term Memory(LSTM)network designed for Tibetan-Chinese plagiarism detection.Our approach begins with the introduction of translation-based data augmentation,aimed at expanding the bilingual training dataset.Subsequently,we propose a pre-detection method leveraging abstract document vectors to enhance detection efficiency.Finally,we introduce an improved attentive Siamese LSTM network tailored for Tibetan-Chinese plagiarism detection.We conduct comprehensive experiments to showcase the effectiveness of our proposed plagiarism detection framework.展开更多
Thanks to the strong representation capability of pre-trained language models,supervised machine translation models have achieved outstanding performance.However,the performances of these models drop sharply when the ...Thanks to the strong representation capability of pre-trained language models,supervised machine translation models have achieved outstanding performance.However,the performances of these models drop sharply when the scale of the parallel training corpus is limited.Considering the pre-trained language model has a strong ability for monolingual representation,it is the key challenge for machine translation to construct the in-depth relationship between the source and target language by injecting the lexical and syntactic information into pre-trained language models.To alleviate the dependence on the parallel corpus,we propose a Linguistics Knowledge-Driven MultiTask(LKMT)approach to inject part-of-speech and syntactic knowledge into pre-trained models,thus enhancing the machine translation performance.On the one hand,we integrate part-of-speech and dependency labels into the embedding layer and exploit large-scale monolingual corpus to update all parameters of pre-trained language models,thus ensuring the updated language model contains potential lexical and syntactic information.On the other hand,we leverage an extra self-attention layer to explicitly inject linguistic knowledge into the pre-trained language model-enhanced machine translation model.Experiments on the benchmark dataset show that our proposed LKMT approach improves the Urdu-English translation accuracy by 1.97 points and the English-Urdu translation accuracy by 2.42 points,highlighting the effectiveness of our LKMT framework.Detailed ablation experiments confirm the positive impact of part-of-speech and dependency parsing on machine translation.展开更多
Background: COVID-19 was initially described to affect the respiratory system, it is now known to affect other systems and present in a myriad of ways. This has often proved to be a diagnostic and therapeutic challeng...Background: COVID-19 was initially described to affect the respiratory system, it is now known to affect other systems and present in a myriad of ways. This has often proved to be a diagnostic and therapeutic challenge to clinicians and stretched healthcare resources, sometimes with poor outcomes. We report on an atypical presentation of SARS-CoV-2 infection as an acute exacerbation of myasthenia gravis in a low resource setting. Case Presentation: A middle aged male with myasthenia gravis presented with a three weeks’ history of worsening generalized weakness and three days’ history of new-onset dysphagia with no history of fever and cough. After a few days of apparent improvement, the patient developed progressive respiratory distress. Further evaluation confirmed bilateral pneumonia and COVID-19. Conclusion: There is a need for clinicians and health service providers to have a low threshold for suspicion and testing for COVID-19 in patients with pre-existing medical conditions and immunomodulatory therapy such as Myasthenia gravis.展开更多
Kidney transplantation(KT)is the optimal form of renal replacement therapy for patients with end-stage renal diseases.However,this health service is not available to all patients,especially in developing countries.The...Kidney transplantation(KT)is the optimal form of renal replacement therapy for patients with end-stage renal diseases.However,this health service is not available to all patients,especially in developing countries.The deceased donor KT programs are mostly absent,and the living donor KT centers are scarce.Single-center studies presenting experiences from developing countries usually report a variety of challenges.This review addresses these challenges and the opposing strategies by reviewing the single-center experiences of developing countries.The financial challenges hamper the infrastructural and material availability,coverage of transplant costs,and qualification of medical personnel.The sociocultural challenges influence organ donation,equity of beneficence,and regular follow-up work.Low interests and motives for transplantation may result from high medicolegal responsibilities in KT practice,intense potential psychosocial burdens,complex qualification protocols,and low productivity or compensation for KT practice.Low medical literacy about KT advantages is prevalent among clinicians,patients,and the public.The inefficient organizational and regulatory oversight is translated into inefficient healthcare systems,absent national KT programs and registries,uncoordinated job descriptions and qualification protocols,uncoordinated on-site investigations with regulatory constraints,and the prevalence of commercial KT practices.These challenges resulted in noticeable differences between KT services in developed and developing countries.The coping strategies can be summarized in two main mechanisms:The first mechanism is maximizing the available resources by increasing the rates of living kidney donation,promoting the expertise of medical personnel,reducing material consumption,and supporting the establishment and maintenance of KT programs.The latter warrants the expansion of the public sector and the elimination of non-ethical KT practices.The second mechanism is recruiting external resources,including financial,experience,and training agreements.展开更多
The performance of a machine translation system heavily depends on the quantity and quality of the bilingual language resource. However,getting a parallel corpus,which has a large scale and is of high quality,is a ver...The performance of a machine translation system heavily depends on the quantity and quality of the bilingual language resource. However,getting a parallel corpus,which has a large scale and is of high quality,is a very difficult task especially for low resource languages such as Chinese-Vietnamese. Fortunately,multilingual user generated contents（ UGC）,such as bilingual movie subtitles,provide us access to automatic construction of the parallel corpus. Although the amount of UGC parallel corpora can be considerable,the original corpus is not suitable for statistical machine translation（ SMT） systems. The corpus may contain translation errors,sentence mismatching,free translations,etc. To improve the quality of the bilingual corpus for SMT systems,three filtering methods are proposed： sentence length difference,the semantic of sentence pairs,and machine learning. Experiments are conducted on the Chinese to Vietnamese translation corpus.Experimental results demonstrate that all the three methods effectively improve the corpus quality,and the machine translation performance（ BLEU score） can be improved by 1. 32.展开更多
Background and Aim:Despite their flaws,the low-cost but powerful economical solutions can ensure everyone has access to health.The main aim of this study is to extract characteristics of frugal innovation(FI)and socia...Background and Aim:Despite their flaws,the low-cost but powerful economical solutions can ensure everyone has access to health.The main aim of this study is to extract characteristics of frugal innovation(FI)and social innovation(SI)for Primary Health Centers(PHCs)in low resource settings(LRS)for sustainable development.We will use the gained insights to design the mobile primary healthcare infrastructure using FI and SI strategies.There is a lack of methodology to design sustainable healthcare infrastructure for LRS.There is a gap in the literature about building sustainable infrastructure to provide basic healthcare facilities essential to the community.This article studies several factors necessary for designing sustainable infrastructure from the lens of FI,SI,and sustainability to develop a mobile healthcare infrastructure for last-mile people.Methods:Started with purposive sampled case studies to find out factors and criteria that most affect the success for an innovation to be frugal,social,and sustainable.The established criteria were used to design,develop,and deploy the mobile Primary Health Center(mPHC).Moving forward,we tested the system designed with stakeholders to gather insights.At this stage we found the feedback loop from the stakeholders and the role of interdisciplinary discussions between experts,medical officers,nurses,patient,and other staff of PHCs during the design,development,deployment,and test stage to be useful in taking design decisions efficiently.Results:The designed healthcare infrastructure of mPHC through the aspects of FI and SI proves to be efficient in providing key healthcare services to LRS.Conclusion:Focusing on essential capabilities and optimizing performance with technology,methodologies,and processes reduces costs in an innovation.Focus on socially inclusive and rebalancing power disparities,overcome societal challenges and improve human capabilities will create a sustainable and novel solution.展开更多
We present a novel standard convolutional symbols generator(SCSG)block for a multi-parameter reconfigurable Viterbi decoder to optimize resource consumption and adaption of multiple parameters.The SCSG block generates...We present a novel standard convolutional symbols generator(SCSG)block for a multi-parameter reconfigurable Viterbi decoder to optimize resource consumption and adaption of multiple parameters.The SCSG block generates all the states and calculates all the possible standard convolutional symbols corresponding to the states using an iterative approach.The architecture of the Viterbi decoder based on the SCSG reduces resource consumption for recalculating the branch metrics and rearranging the correspondence between branch metrics and transition paths.The proposed architecture supports constraint lengths from 3 to 9,code rates of 1/2,1/3,and 1/4,and fully optional polynomials.The proposed Viterbi decoder has been implemented on the Xilinx XC7VX485T device with a high throughput of about 200 Mbps and a low resource consumption of 162k logic gates.展开更多
Since the onset of the HIV epidemic,assessing CD4+T-cells has become a routine procedure for evaluating immune deficiency,with flow cytometry established as the gold standard.Over time,various strategies and platforms...Since the onset of the HIV epidemic,assessing CD4+T-cells has become a routine procedure for evaluating immune deficiency,with flow cytometry established as the gold standard.Over time,various strategies and platforms have been introduced to improve CD4+cell enumeration,aiming to enhance the performance of diagnostic devices and bring the service closer to patients.These advancements are particularly critical for low-resource settings and point-of-care applications,where the excellent performance of flow cytometry is hindered by its unsuitability in such environments.This work presents an innovative electrochemical microfluidic device that,with further development,could be applied for HIV management in low resource settings.The setup integrates an electrochemical sensor within a PDMS microfluidic structure,allowing for on-chip electrode functionalization and cell detection.Using electrochemical impedance spectroscopy,the biosensor demonstrates a linear detection range from 1.25×105 to 2×106 cells/mL,with a detection limit of 1.41×105 cells/mL for CD4+cells isolated from blood samples,aligning with clinical ranges for both healthy and HIV+patients.The biosensor shows specificity towards CD4+cells with negligible response to monocytes,neutrophils,and bovine serum albumin.Its integration with a microfluidic chip for sensor fabrication and cell detection,compact size,minimal manual handling,ease of fabrication,electrochemical detection capability,and potential for multiplexing together with the detection range make the device particularly advantageous for use in low-resource settings,standing out among other devices described in the literature.This study also investigates the integration of a microfluidic Dean Flow Fractionation(DFF)chip for cell separation.展开更多
Objective:We conducted this study to assess the etiopathogenic relation of otitis media with effusion(OME)in a group of children aged 1-5 years among the local population of Rawalpindi.Methods:This was a cross-section...Objective:We conducted this study to assess the etiopathogenic relation of otitis media with effusion(OME)in a group of children aged 1-5 years among the local population of Rawalpindi.Methods:This was a cross-sectional retrospective study.Study was conducted among the children presenting to the immunization center of three tertiary care hospitals of Rawalpindi.Otitis media was assessed by clinical examination and tympanometry from August 2019 to January 2020.Multi-factor regression analysis was then applied to recognize the statistical significance and association of various risk factors to OME.Results:Out of 400 children enrolled in this study,108(27.0%)had OME,out of which 65(60.2%)were males and 30(27.8%)were of age group 2-3 years.Multivariable regression analysis of risk factors associated with OME showed it was strongly associated with snoring(P<0.001),last year symptoms(attack of ear aches with hearing loss[P=0.002]),drugs(URTI antibiotics[P=0.026],All 3 drugs[P=0.013]).Conclusions:We found out that OME is a common disease which if not identified or treated timely can lead to other hard to cure health problems.Control of its etiopathogenic factors can play a major role in its prevention.展开更多
基金supported by the National Natural Science Foundation of China(Nos.42488201,42172137,42050104,and 42050102)the National Key R&D Program of China(No.2023YFF0804000)Sichuan Provincial Youth Science&Technology Innovative Research Group Fund(No.2022JDTD0004)
文摘Geological reports are a significant accomplishment for geologists involved in geological investigations and scientific research as they contain rich data and textual information.With the rapid development of science and technology,a large number of textual reports have accumulated in the field of geology.However,many non-hot topics and non-English speaking regions are neglected in mainstream geoscience databases for geological information mining,making it more challenging for some researchers to extract necessary information from these texts.Natural Language Processing(NLP)has obvious advantages in processing large amounts of textual data.The objective of this paper is to identify geological named entities from Chinese geological texts using NLP techniques.We propose the RoBERTa-Prompt-Tuning-NER method,which leverages the concept of Prompt Learning and requires only a small amount of annotated data to train superior models for recognizing geological named entities in low-resource dataset configurations.The RoBERTa layer captures context-based information and longer-distance dependencies through dynamic word vectors.Finally,we conducted experiments on the constructed Geological Named Entity Recognition(GNER)dataset.Our experimental results show that the proposed model achieves the highest F1 score of 80.64%among the four baseline algorithms,demonstrating the reliability and robustness of using the model for Named Entity Recognition of geological texts.
基金This work was supported by the China National Natural Science Foundation No.(61331013)the Young faculty scientific research ability promotion program of Minzu University of China.
文摘Readability is a fundamental problem in textbooks assessment.For low resources languages(LRL),however,little investigation has been done on the readability of textbook.In this paper,we proposed a readability assessment method for Tibetan textbook(a low resource language).We extract features based on the information that are gotten by Tibetan segmentation and named entity recognition.Then,we calculate the correlation of different features using Pearson Correlation Coefficient and select some feature sets to design the readability formula.Fit detection,F test and T test are applied on these selected features to generate a new readability assessment formula.Experiment shows that this new formula is capable of assessing the readability of Tibetan textbooks.
文摘The use of bilateral internal thoracic arteries to treat coronary artery disease is very less despite of improved long-term survival. In this case report</span><span style="font-family:"">,</span><span style="font-family:""> a patient presented with TVD has been managed with anaortic off-pump complete arterial revascularization (OPCABG) by using composite LIMA RIMA Y Sequential 5 grafts to achieve complete arterial revascularization successfully. This type of operative procedure is technically difficult in a resource poor country like Bangladesh. However, this has been done by a group of young cardiac surgeon. During follow up</span><span style="font-family:"">,</span><span style="font-family:""> the patient is in a good condition.
基金M.Faheem is supported by VTT Technical Research Center of Finland.
文摘Neural machine translation(NMT)has advanced with deep learning and large-scale multilingual models,yet translating lowresource languages often lacks sufficient training data and leads to hallucinations.This often results in translated content that diverges significantly from the source text.This research proposes a refined Contrastive Decoding(CD)algorithm that dynamically adjusts weights of log probabilities from strong expert and weak amateur models to mitigate hallucinations in lowresource NMT and improve translation quality.Advanced large language NMT models,including ChatGLM and LLaMA,are fine-tuned and implemented for their superior contextual understanding and cross-lingual capabilities.The refined CD algorithm evaluates multiple candidate translations using BLEU score,semantic similarity,and Named Entity Recognition accuracy.Extensive experimental results show substantial improvements in translation quality and a significant reduction in hallucination rates.Fine-tuned models achieve higher evaluation metrics compared to baseline models and state-of-the-art models.An ablation study confirms the contributions of each methodological component and highlights the effectiveness of the refined CD algorithm and advanced models in mitigating hallucinations.Notably,the refined methodology increased the BLEU score by approximately 30%compared to baseline models.
基金Chang Gung University and Chang Gung Memorial Hospital under project number NERPD4Q0021.
文摘Text clustering is an important task because of its vital role in NLP-related tasks.However,existing research on clustering is mainly based on the English language,with limited work on low-resource languages,such as Urdu.Low-resource language text clustering has many drawbacks in the form of limited annotated collections and strong linguistic diversity.Theprimary aim of this paper is twofold:(1)By introducing a clustering dataset namedUNC-2025 comprises 100k Urdu news documents,and(2)a detailed empirical standard of Large Language Model(LLM)improved clusteringmethods for Urdu text.We explicitly evaluate the behavior of the 11multilingual and Urdu-specific embeddings on 3 different clustering algorithms.We carefully evaluated our performance based on a set of internal and external measurements of validity.We discover the best configuration of the mBERT embedding with the HDBSCAN algorithm that attains a new state-of-the-art performance with a high score of external validity of 0.95.This new LLM method has created a new strong standard of Urdu text clustering.Importantly,the results confirm the strength and high scalability of the LLM-generated embeddings towards the ability to generalise the fine,subtle semantics needed to discover topics in low-resource settings and open the door to novel NLP applications in underrepresented languages.
文摘Introduction: Standard procedures for surgical fixation of proximal femoral fractures (PFF) require an image intensifier which in developing countries remains a luxury. We hypothesized that, with a well-codified technique, the Watson Jones approach (WJA) without image intensifier nor traction table, can allow open reduction and internal fixation (ORIF) of PFF using Dynamic hip screw (DHS), with satisfactory outcome. Patients and methods: Forty one consecutive patients (mean age 59.5 ± 21.6 years, 61% males) who were followed in a Teaching Hospital for PFF treated by ORIF using the WJA and DHS from January 2016 to December 2020 were reassessed. The outcome measures were the quality of the reduction, the positioning of the implants, the tip-apex distance (TAD), the rate and delay of consolidation, the functional results using Postel Merle d’Aubigné (PMA) score, the rate of surgical site infection (SSI) and the overall mortality. Logistic regression was used to determine factors associated with mechanical failure. Results: The mean follow-up period was 33.8 ± 15.0 months. Fracture reduction was good in 31 (75.6%) cases and acceptable in 8(19.5%) cases. Implant position was fair to good in 37 (90.2%) patients. The mean TAD was 26.1 ± 3.9 mm. Three patients developed SSI. Consolidation was achieved in 38 (92.6%) patients. The functional results were good to excellent in 80.5% of patients. The overall mortality rate was 7.3%. There were an association between mechanical failure and osteoporosis (p = 0.04), fracture reduction (p = 0.003), and TAD (p = 0.025). In multivariate logistic regression, no independent factors were predictive of mechanical failure. Conclusion: This study shows that ORIF using DHS for PFF via the Watson-Jones approach without an image intensifier can give satisfactory anatomical and functional outcomes in low-resource settings. It provides and validates a reliable and reproducible technique that deserves to be diffused to surgeons in austere areas over the world.
文摘BACKGROUND Globally,it’s estimated that at least 1 billion people have a near and/or distance vision impairment that could have been prevented or is yet to be addressed.The burden of unaddressed vision impairment and blindness is estimated to be four times higher in low and intermediate-resource settings than in high-income settings.[1]
文摘Because of its simplicity,reliability,and replicability,the Masquelet induced membrane technique(IMT)has become one of the preferred methods for critical bone defect reconstruction in extremities.Although it is now used worldwide,few studies have been published about IMT in military practice.Bone reconstruction is particularly challenging in this context of care due to extensive soft-tissue injury,early wound infection,and even delayed management in austere conditions.Based on our clinical expertise,recent research,and a literature analysis,this narrative review provides an overview of the IMT application to combat-related bone defects.It presents technical specificities and future developments aiming to optimize IMT outcomes,including for the management of massive multi-tissue defects or bone reconstruction performed in the field with limited resources.
文摘Creating a parallel corpus for machine translation is a challenging and time-consuming task,especially in a linguistically diverse country like the Philippines,with 185 languages.Although a wealth of text is available,annotated data is scarce,particularly for languages like Bikol.Bikol is one of the major languages in the Philippines;however,its underrepresentation in the digital sphere is attributed to the absence of annotated data.This study outlines the development process of BFParCo,a proposed gold standard dataset for the Bikol and Filipino parallel corpus.The corpus underwent refinement through manual phrase alignment,translation,and evaluation.Subsequently,T5 and mT5 transformer models were fine-tuned with the parallel corpus and were evaluated using the BLEU metric.The results showed a notable improvement in Bilingual Evaluation Understudy(BLEU)score after fine-tuning,with an increase of 60.68 in BIK→FIL and 58.93 in FIL→BIK translations.Additionally,human evaluators comprehensively assessed the fine-tuned models'results using Multidimensional Quality Metrics and Scalar Quality Metrics error taxonomies.The fine-tuned models then were made publicly accessible through Hugging Face.This study represents a significant stride in advancing machine translation tools for Bikol and Filipino languages.
基金supported by the National Natural Science Foundation of China(No.62271456)the Open Projects Program of State Key Laboratory of Multimodal Artificial Intelligence Systems.
文摘Low-resource text plagiarism detection faces a significant challenge due to the limited availability of labeled data for training.This task requires the development of sophisticated algorithms capable of identifying similarities and differences in texts,particularly in the realm of semantic rewriting and translation-based plagiarism detection.In this paper,we present an enhanced attentive Siamese Long Short-Term Memory(LSTM)network designed for Tibetan-Chinese plagiarism detection.Our approach begins with the introduction of translation-based data augmentation,aimed at expanding the bilingual training dataset.Subsequently,we propose a pre-detection method leveraging abstract document vectors to enhance detection efficiency.Finally,we introduce an improved attentive Siamese LSTM network tailored for Tibetan-Chinese plagiarism detection.We conduct comprehensive experiments to showcase the effectiveness of our proposed plagiarism detection framework.
基金supported by the National Natural Science Foundation of China under Grant(61732005,61972186)Yunnan Provincial Major Science and Technology Special Plan Projects(Nos.202103AA080015,202203AA080004).
文摘Thanks to the strong representation capability of pre-trained language models,supervised machine translation models have achieved outstanding performance.However,the performances of these models drop sharply when the scale of the parallel training corpus is limited.Considering the pre-trained language model has a strong ability for monolingual representation,it is the key challenge for machine translation to construct the in-depth relationship between the source and target language by injecting the lexical and syntactic information into pre-trained language models.To alleviate the dependence on the parallel corpus,we propose a Linguistics Knowledge-Driven MultiTask(LKMT)approach to inject part-of-speech and syntactic knowledge into pre-trained models,thus enhancing the machine translation performance.On the one hand,we integrate part-of-speech and dependency labels into the embedding layer and exploit large-scale monolingual corpus to update all parameters of pre-trained language models,thus ensuring the updated language model contains potential lexical and syntactic information.On the other hand,we leverage an extra self-attention layer to explicitly inject linguistic knowledge into the pre-trained language model-enhanced machine translation model.Experiments on the benchmark dataset show that our proposed LKMT approach improves the Urdu-English translation accuracy by 1.97 points and the English-Urdu translation accuracy by 2.42 points,highlighting the effectiveness of our LKMT framework.Detailed ablation experiments confirm the positive impact of part-of-speech and dependency parsing on machine translation.
文摘Background: COVID-19 was initially described to affect the respiratory system, it is now known to affect other systems and present in a myriad of ways. This has often proved to be a diagnostic and therapeutic challenge to clinicians and stretched healthcare resources, sometimes with poor outcomes. We report on an atypical presentation of SARS-CoV-2 infection as an acute exacerbation of myasthenia gravis in a low resource setting. Case Presentation: A middle aged male with myasthenia gravis presented with a three weeks’ history of worsening generalized weakness and three days’ history of new-onset dysphagia with no history of fever and cough. After a few days of apparent improvement, the patient developed progressive respiratory distress. Further evaluation confirmed bilateral pneumonia and COVID-19. Conclusion: There is a need for clinicians and health service providers to have a low threshold for suspicion and testing for COVID-19 in patients with pre-existing medical conditions and immunomodulatory therapy such as Myasthenia gravis.
文摘Kidney transplantation(KT)is the optimal form of renal replacement therapy for patients with end-stage renal diseases.However,this health service is not available to all patients,especially in developing countries.The deceased donor KT programs are mostly absent,and the living donor KT centers are scarce.Single-center studies presenting experiences from developing countries usually report a variety of challenges.This review addresses these challenges and the opposing strategies by reviewing the single-center experiences of developing countries.The financial challenges hamper the infrastructural and material availability,coverage of transplant costs,and qualification of medical personnel.The sociocultural challenges influence organ donation,equity of beneficence,and regular follow-up work.Low interests and motives for transplantation may result from high medicolegal responsibilities in KT practice,intense potential psychosocial burdens,complex qualification protocols,and low productivity or compensation for KT practice.Low medical literacy about KT advantages is prevalent among clinicians,patients,and the public.The inefficient organizational and regulatory oversight is translated into inefficient healthcare systems,absent national KT programs and registries,uncoordinated job descriptions and qualification protocols,uncoordinated on-site investigations with regulatory constraints,and the prevalence of commercial KT practices.These challenges resulted in noticeable differences between KT services in developed and developing countries.The coping strategies can be summarized in two main mechanisms:The first mechanism is maximizing the available resources by increasing the rates of living kidney donation,promoting the expertise of medical personnel,reducing material consumption,and supporting the establishment and maintenance of KT programs.The latter warrants the expansion of the public sector and the elimination of non-ethical KT practices.The second mechanism is recruiting external resources,including financial,experience,and training agreements.
基金Supported by the National Basic Research Program of China(973Program)(2013CB329303)the National Natural Science Foundation of China(61502035)
文摘The performance of a machine translation system heavily depends on the quantity and quality of the bilingual language resource. However,getting a parallel corpus,which has a large scale and is of high quality,is a very difficult task especially for low resource languages such as Chinese-Vietnamese. Fortunately,multilingual user generated contents（ UGC）,such as bilingual movie subtitles,provide us access to automatic construction of the parallel corpus. Although the amount of UGC parallel corpora can be considerable,the original corpus is not suitable for statistical machine translation（ SMT） systems. The corpus may contain translation errors,sentence mismatching,free translations,etc. To improve the quality of the bilingual corpus for SMT systems,three filtering methods are proposed： sentence length difference,the semantic of sentence pairs,and machine learning. Experiments are conducted on the Chinese to Vietnamese translation corpus.Experimental results demonstrate that all the three methods effectively improve the corpus quality,and the machine translation performance（ BLEU score） can be improved by 1. 32.
文摘Background and Aim:Despite their flaws,the low-cost but powerful economical solutions can ensure everyone has access to health.The main aim of this study is to extract characteristics of frugal innovation(FI)and social innovation(SI)for Primary Health Centers(PHCs)in low resource settings(LRS)for sustainable development.We will use the gained insights to design the mobile primary healthcare infrastructure using FI and SI strategies.There is a lack of methodology to design sustainable healthcare infrastructure for LRS.There is a gap in the literature about building sustainable infrastructure to provide basic healthcare facilities essential to the community.This article studies several factors necessary for designing sustainable infrastructure from the lens of FI,SI,and sustainability to develop a mobile healthcare infrastructure for last-mile people.Methods:Started with purposive sampled case studies to find out factors and criteria that most affect the success for an innovation to be frugal,social,and sustainable.The established criteria were used to design,develop,and deploy the mobile Primary Health Center(mPHC).Moving forward,we tested the system designed with stakeholders to gather insights.At this stage we found the feedback loop from the stakeholders and the role of interdisciplinary discussions between experts,medical officers,nurses,patient,and other staff of PHCs during the design,development,deployment,and test stage to be useful in taking design decisions efficiently.Results:The designed healthcare infrastructure of mPHC through the aspects of FI and SI proves to be efficient in providing key healthcare services to LRS.Conclusion:Focusing on essential capabilities and optimizing performance with technology,methodologies,and processes reduces costs in an innovation.Focus on socially inclusive and rebalancing power disparities,overcome societal challenges and improve human capabilities will create a sustainable and novel solution.
基金Project supported by the Natural Science Foundation of Jiangsu Province,China(No.BK20130156)the Summit of the Six Top Talents Program of Jiangsu Province,China(No.2013-DZXX-027)+1 种基金the Fundamental Research Funds for the Central Universities,China(No.JUSRP51510)the Graduate Student Innovation Program for Universities of Jiangsu Province,China(Nos.KYLX15_1192,KYLX16_0776,and SJLX16_0500)
文摘We present a novel standard convolutional symbols generator(SCSG)block for a multi-parameter reconfigurable Viterbi decoder to optimize resource consumption and adaption of multiple parameters.The SCSG block generates all the states and calculates all the possible standard convolutional symbols corresponding to the states using an iterative approach.The architecture of the Viterbi decoder based on the SCSG reduces resource consumption for recalculating the branch metrics and rearranging the correspondence between branch metrics and transition paths.The proposed architecture supports constraint lengths from 3 to 9,code rates of 1/2,1/3,and 1/4,and fully optional polynomials.The proposed Viterbi decoder has been implemented on the Xilinx XC7VX485T device with a high throughput of about 200 Mbps and a low resource consumption of 162k logic gates.
基金funding from Santander postgraduate mobility awards and Department of Electronic&Electrical Engineering,University of BathR.S.was funded through UK Engineering and Physical Sciences Research Council grant number EP/V040189/1.
文摘Since the onset of the HIV epidemic,assessing CD4+T-cells has become a routine procedure for evaluating immune deficiency,with flow cytometry established as the gold standard.Over time,various strategies and platforms have been introduced to improve CD4+cell enumeration,aiming to enhance the performance of diagnostic devices and bring the service closer to patients.These advancements are particularly critical for low-resource settings and point-of-care applications,where the excellent performance of flow cytometry is hindered by its unsuitability in such environments.This work presents an innovative electrochemical microfluidic device that,with further development,could be applied for HIV management in low resource settings.The setup integrates an electrochemical sensor within a PDMS microfluidic structure,allowing for on-chip electrode functionalization and cell detection.Using electrochemical impedance spectroscopy,the biosensor demonstrates a linear detection range from 1.25×105 to 2×106 cells/mL,with a detection limit of 1.41×105 cells/mL for CD4+cells isolated from blood samples,aligning with clinical ranges for both healthy and HIV+patients.The biosensor shows specificity towards CD4+cells with negligible response to monocytes,neutrophils,and bovine serum albumin.Its integration with a microfluidic chip for sensor fabrication and cell detection,compact size,minimal manual handling,ease of fabrication,electrochemical detection capability,and potential for multiplexing together with the detection range make the device particularly advantageous for use in low-resource settings,standing out among other devices described in the literature.This study also investigates the integration of a microfluidic Dean Flow Fractionation(DFF)chip for cell separation.
文摘Objective:We conducted this study to assess the etiopathogenic relation of otitis media with effusion(OME)in a group of children aged 1-5 years among the local population of Rawalpindi.Methods:This was a cross-sectional retrospective study.Study was conducted among the children presenting to the immunization center of three tertiary care hospitals of Rawalpindi.Otitis media was assessed by clinical examination and tympanometry from August 2019 to January 2020.Multi-factor regression analysis was then applied to recognize the statistical significance and association of various risk factors to OME.Results:Out of 400 children enrolled in this study,108(27.0%)had OME,out of which 65(60.2%)were males and 30(27.8%)were of age group 2-3 years.Multivariable regression analysis of risk factors associated with OME showed it was strongly associated with snoring(P<0.001),last year symptoms(attack of ear aches with hearing loss[P=0.002]),drugs(URTI antibiotics[P=0.026],All 3 drugs[P=0.013]).Conclusions:We found out that OME is a common disease which if not identified or treated timely can lead to other hard to cure health problems.Control of its etiopathogenic factors can play a major role in its prevention.
