RESUMO
Accurately assigning standardized diagnosis and procedure codes from clinical text is crucial for healthcare applications. However, this remains challenging due to the complexity of medical language. This paper proposes a novel model that incorporates extreme multi-label classification tasks to enhance International Classification of Diseases (ICD) coding. The model utilizes deformable convolutional neural networks to fuse representations from hidden layer outputs of pre-trained language models and external medical knowledge embeddings fused using a multimodal approach to provide rich semantic encodings for each code. A probabilistic label tree is constructed based on the hierarchical structure existing in ICD labels to incorporate ontological relationships between ICD codes and enable structured output prediction. Experiments on medical code prediction on the MIMIC-III database demonstrate competitive performance, highlighting the benefits of this technique for robust clinical code assignment.
Assuntos
Classificação Internacional de Doenças , Redes Neurais de Computação , Semântica , Humanos , Processamento de Linguagem Natural , Algoritmos , Bases de Dados FactuaisRESUMO
MOTIVATION: Integrating information from data sources representing different study designs has the potential to strengthen evidence in population health research. However, this concept of evidence "triangulation" presents a number of challenges for systematically identifying and integrating relevant information. These include the harmonization of heterogenous evidence with common semantic concepts and properties, as well as the priortization of the retrieved evidence for triangulation with the question of interest. RESULTS: We present ASQ (Annotated Semantic Queries), a natural language query interface to the integrated biomedical entities and epidemiological evidence in EpiGraphDB, which enables users to extract "claims" from a piece of unstructured text, and then investigate the evidence that could either support, contradict the claims, or offer additional information to the query.This approach has the potential to support the rapid review of preprints, grant applications, conference abstracts and articles submitted for peer review. ASQ implements strategies to harmonize biomedical entities in different taxonomies and evidence from different sources, to facilitate evidence triangulation and interpretation. AVAILABILITY AND IMPLEMENTATION: ASQ is openly available at https://asq.epigraphdb.org and its source code is available at https://github.com/mrcieu/epigraphdb-asq under GPL-3.0 license. SUPPLEMENTARY INFORMATION: Further information can be found in the Supplementary Materials as well as on the ASQ platform via https://asq.epigraphdb.org/docs.
RESUMO
Decision-making in healthcare often relies on narrative guidelines; however, these instruments are poorly accessible for supporting clinical decision-making. This study explores the application of rule-based decision logic in algorithmic modeling, emphasizing its great potential in clinical decision support and research. Integrating rule-based algorithms with existing information systems and real-world data poses a serious challenge. Integrating decision algorithms with information standards increases their effectiveness across various applications. This study outlines a method for constructing clinical decision trees (CDTs), highlighting their transparency and interpretability, using information standards as a design principle. We use the digitization of the Dutch breast cancer guideline through CDTs as a case study to exemplify their versatility and practical significance. The process step 'primary treatment' has been successfully translated from the narrative guidelines format to the anticipated ted computational format.
Assuntos
Algoritmos , Neoplasias da Mama , Sistemas de Apoio a Decisões Clínicas , Oncologia , Humanos , Neoplasias da Mama/terapia , Árvores de Decisões , Guias de Prática Clínica como Assunto , Feminino , Países BaixosRESUMO
OBJECTIVE: Traditional knowledge-based and machine learning diagnostic decision support systems have benefited from integrating the medical domain knowledge encoded in the Unified Medical Language System (UMLS). The emergence of Large Language Models (LLMs) to supplant traditional systems poses questions of the quality and extent of the medical knowledge in the models' internal knowledge representations and the need for external knowledge sources. The objective of this study is three-fold: to probe the diagnosis-related medical knowledge of popular LLMs, to examine the benefit of providing the UMLS knowledge to LLMs (grounding the diagnosis predictions), and to evaluate the correlations between human judgments and the UMLS-based metrics for generations by LLMs. METHODS: We evaluated diagnoses generated by LLMs from consumer health questions and daily care notes in the electronic health records using the ConsumerQA and Problem Summarization datasets. Probing LLMs for the UMLS knowledge was performed by prompting the LLM to complete the diagnosis-related UMLS knowledge paths. Grounding the predictions was examined in an approach that integrated the UMLS graph paths and clinical notes in prompting the LLMs. The results were compared to prompting without the UMLS paths. The final experiments examined the alignment of different evaluation metrics, UMLS-based and non-UMLS, with human expert evaluation. RESULTS: In probing the UMLS knowledge, GPT-3.5 significantly outperformed Llama2 and a simple baseline yielding an F1 score of 10.9% in completing one-hop UMLS paths for a given concept. Grounding diagnosis predictions with the UMLS paths improved the results for both models on both tasks, with the highest improvement (4%) in SapBERT score. There was a weak correlation between the widely used evaluation metrics (ROUGE and SapBERT) and human judgments. CONCLUSION: We found that while popular LLMs contain some medical knowledge in their internal representations, augmentation with the UMLS knowledge provides performance gains around diagnosis generation. The UMLS needs to be tailored for the task to improve the LLMs predictions. Finding evaluation metrics that are aligned with human judgments better than the traditional ROUGE and BERT-based scores remains an open research question.
RESUMO
BACKGROUND: The increasing aging population presents a significant challenge, accompanied by a shortage of professional caregivers, adding to the therapeutic burden. Clinical decision support systems, utilizing computerized clinical guidelines, can improve healthcare quality, reduce expenses, save time, and boost caregiver efficiency. OBJECTIVES: 1) Develop and evaluate an automated quality assessment (QA) system for retrospective longitudinal care quality analysis, focusing on clinical staff adherence to evidence-based guidelines (GLs). 2) Assess the system's technical feasibility and functional capability for senior nurse use in geriatric pressure-ulcer management. METHODS: A computational QA system using our Quality Assessment Temporal Patterns (QATP) methodology was designed and implemented. Our methodology transforms the GL's procedural-knowledge into declarative-knowledge temporal-abstraction patterns representing the expected execution trace in the patient's data for correct therapy application. Fuzzy temporal logic allows for partial compliance, reflecting individual and grouped action performance considering their values and temporal aspects. The system was tested using a pressure ulcer treatment GL and data from 100 geriatric patients' Electronic Medical Records (EMR). After technical evaluation for accuracy and feasibility, an extensive functional evaluation was conducted by an experienced nurse, comparing QA scores with and without system support, and versus automated system scores. Time efficiency was also measured. RESULTS: QA scores from the geriatric nurse, with and without system's support, did not significantly differ from those provided by the automated system (p < 0.05), demonstrating the effectiveness and reliability of both manual and automated methods. The system-supported manual QA process reduced scoring time by approximately two-thirds, from an average of 17.3 min per patient manually to about 5.9 min with the system's assistance, highlighting the system's efficiency potential in clinical practice. CONCLUSION: The QA system based on QATP, produces scores consistent with an experienced nurse's assessment for complex care over extended periods. It enables quick and accurate quality care evaluation for multiple patients after brief training. Such automated QA systems may empower nursing staff, enabling them to manage more patients, accurately and consistently, while reducing costs due to saved time and effort, and enhanced compliance with evidence-based guidelines.
Assuntos
Sistemas de Apoio a Decisões Clínicas , Úlcera por Pressão , Humanos , Idoso , Úlcera por Pressão/terapia , Registros Eletrônicos de Saúde , Garantia da Qualidade dos Cuidados de Saúde/métodos , Idoso de 80 Anos ou mais , Estudos Retrospectivos , Feminino , Masculino , GeriatriaRESUMO
The gait rehabilitation knee exoskeleton is an advanced rehabilitative assistive device designed to help patients with knee joint dysfunction regain normal gait through training and activity support. This paper introduces a design framework based on the process knowledge representation method to optimize the design and control efficiency of the knee exoskeleton. This framework integrates knowledge of design objects and processes, specifically including requirements, functions, principle work areas, and the representation and multi-dimensional dynamic mapping of the Behavior-Structure (RFPBS) matrix, achieving multi-dimensional dynamic mapping of the knee exoskeleton. This method incorporates biomechanical and physiological knowledge from the rehabilitation process to more effectively simulate and support gait movements during rehabilitation. Research results indicate that the knee rehabilitation exoskeleton design, based on the RFPBS process knowledge representation model, accomplishes multi-dimensional dynamic mapping, providing a scientific basis and effective support for the rehabilitation of patients with knee joint dysfunction.
RESUMO
Microscopy image analysis plays a pivotal role in parasitology research. Deep learning (DL), a subset of artificial intelligence (AI), has garnered significant attention. However, traditional DL-based methods for general purposes are data-driven, often lacking explainability due to their black-box nature and sparse instructional resources. To address these challenges, this article presents a comprehensive review of recent advancements in knowledge-integrated DL models tailored for microscopy image analysis in parasitology. The massive amounts of human expert knowledge from parasitologists can enhance the accuracy and explainability of AI-driven decisions. It is expected that the adoption of knowledge-integrated DL models will open up a wide range of applications in the field of parasitology.
Assuntos
Inteligência Artificial , Processamento de Imagem Assistida por Computador , Microscopia , Parasitologia , Parasitologia/métodos , Parasitologia/instrumentação , Parasitologia/tendências , Microscopia/instrumentação , Microscopia/métodos , Microscopia/normas , Humanos , Processamento de Imagem Assistida por Computador/instrumentação , Processamento de Imagem Assistida por Computador/métodos , Aprendizado ProfundoRESUMO
The left and right anterior temporal lobes (ATLs) encode semantic representations. They show graded hemispheric specialization in function, with the left ATL contributing preferentially to verbal semantic processing. We investigated the cognitive correlates of this organization, using resting-state functional connectivity as a measure of functional segregation between ATLs. We analyzed two independent resting-state fMRI datasets (n = 86 and n = 642) in which participants' verbal semantic expertise was measured using vocabulary tests. In both datasets, people with more advanced verbal semantic knowledge showed weaker functional connectivity between left and right ventral ATLs. This effect was highly specific. It was not observed for within-hemisphere connections between semantic regions (ventral ATL and inferior frontal gyrus (IFG), though it was found for left-right IFG connectivity in one dataset). Effects were not found for tasks probing semantic control, nonsemantic cognition, or face recognition. Our results suggest that hemispheric specialization in the ATLs is not an innate property but rather emerges as people develop highly detailed verbal semantic representations. We speculate that this effect is a consequence of the left ATL's greater connectivity with left-lateralized written word recognition regions, which causes it to preferentially represent meaning for advanced vocabulary acquired primarily through reading.
Assuntos
Mapeamento Encefálico , Lateralidade Funcional , Imageamento por Ressonância Magnética , Semântica , Lobo Temporal , Humanos , Lobo Temporal/fisiologia , Lobo Temporal/diagnóstico por imagem , Masculino , Feminino , Adulto , Lateralidade Funcional/fisiologia , Adulto Jovem , Mapeamento Encefálico/métodos , Vias Neurais/fisiologia , Vias Neurais/diagnóstico por imagemRESUMO
Introduction: In recent years, the perceptual capabilities of robots have been significantly enhanced. However, the task execution of the robots still lacks adaptive capabilities in unstructured and dynamic environments. Methods: In this paper, we propose an ontology based autonomous robot task processing framework (ARTProF), to improve the robot's adaptability within unstructured and dynamic environments. ARTProF unifies ontological knowledge representation, reasoning, and autonomous task planning and execution into a single framework. The interface between the knowledge base and neural network-based object detection is first introduced in ARTProF to improve the robot's perception capabilities. A knowledge-driven manipulation operator based on Robot Operating System (ROS) is then designed to facilitate the interaction between the knowledge base and the robot's primitive actions. Additionally, an operation similarity model is proposed to endow the robot with the ability to generalize to novel objects. Finally, a dynamic task planning algorithm, leveraging ontological knowledge, equips the robot with adaptability to execute tasks in unstructured and dynamic environments. Results: Experimental results on real-world scenarios and simulations demonstrate the effectiveness and efficiency of the proposed ARTProF framework. Discussion: In future work, we will focus on refining the ARTProF framework by integrating neurosymbolic inference.
RESUMO
Health and social care systems around the globe currently undergo a transformation towards personalized, preventive, predictive, participative precision medicine (5PM), considering the individual health status, conditions, genetic and genomic dispositions, etc., in personal, social, occupational, environmental and behavioral context. This transformation is strongly supported by technologies such as micro- and nanotechnologies, advanced computing, artificial intelligence, edge computing, etc. For enabling communication and cooperation between actors from different domains using different methodologies, languages and ontologies based on different education, experiences, etc., we have to understand the transformed health ecosystems and all its components in structure, function and relationships in the necessary detail ranging from elementary particles up to the universe. That way, we advance design and management of the complex and highly dynamic ecosystem from data to knowledge level. The challenge is the consistent, correct and formalized representation of the transformed health ecosystem from the perspectives of all domains involved, representing and managing them based on related ontologies. The resulting business view of the real-world ecosystem must be interrelated using the ISO/IEC 21838 Top Level Ontologies standard. Thereafter, the outcome can be transformed into implementable solutions using the ISO/IEC 10746 Open Distributed Processing Reference Model. Model and framework for this system-oriented, architecture-centric, ontology-based, policy-driven approach have been developed by the first author and meanwhile standardized as ISO 23903 Interoperability and Integration Reference Architecture.
Assuntos
Medicina de Precisão , Humanos , Inteligência ArtificialRESUMO
Epidemiological modeling is a key lever for infectious disease control and prevention on farms. It makes it possible to understand the spread of pathogens, but also to compare intervention scenarios even in counterfactual situations. However, the actual capability of decision makers to use mechanistic models to support timely interventions is limited. This study demonstrates how artificial intelligence (AI) techniques can make mechanistic epidemiological models more accessible to farmers and veterinarians, and how to transform such models into user-friendly decision-support tools (DST). By leveraging knowledge representation methods, such as the textual formalization of model components through a domain-specific language (DSL), the co-design of mechanistic models and DST becomes more efficient and collaborative. This facilitates the integration of explicit expert knowledge and practical insights into the modeling process. Furthermore, the utilization of AI and software engineering enables the automation of web application generation based on existing mechanistic models. This automation simplifies the development of DST, as tool designers can focus on identifying users' needs and specifying expected features and meaningful presentations of outcomes, instead of wasting time in writing code to wrap models into web apps. To illustrate the practical application of this approach, we consider the example of Bovine Respiratory Disease (BRD), a tough challenge in fattening farms where young beef bulls often develop BRD shortly after being allocated into pens. BRD is a multi-factorial, multi-pathogen disease that is difficult to anticipate and control, often resulting in the massive use of antimicrobials to mitigate its impact on animal health, welfare, and economic losses. The DST developed from an existing mechanistic BRD model empowers users, including farmers and veterinarians, to customize scenarios based on their specific farm conditions. It enables them to anticipate the effects of various pathogens, compare the epidemiological and economic outcomes associated with different farming practices, and decide how to balance the reduction of disease impact and the reduction of antimicrobial usage (AMU). The generic method presented in this article illustrates the potential of artificial intelligence (AI) and software engineering methods to enhance the co-creation of DST based on mechanistic models in veterinary epidemiology. The corresponding pipeline is distributed as an open-source software. By leveraging these advancements, this research aims to bridge the gap between theoretical models and the practical usage of their outcomes on the field.
Assuntos
Inteligência Artificial , Animais , Bovinos , Software , Técnicas de Apoio para a Decisão , Doenças dos Bovinos/prevenção & controle , Doenças dos Bovinos/epidemiologiaRESUMO
Disease ontologies facilitate the semantic organization and representation of domain-specific knowledge. In the case of prostate cancer (PCa), large volumes of research results and clinical data have been accumulated and needed to be standardized for sharing and translational researches. A formal representation of PCa-associated knowledge will be essential to the diverse data standardization, data sharing and the future knowledge graph extraction, deep phenotyping and explainable artificial intelligence developing. In this study, we constructed an updated PCa ontology (PCAO2) based on the ontology development life cycle. An online information retrieval system was designed to ensure the usability of the ontology. The PCAO2 with a subclass-based taxonomic hierarchy covers the major biomedical concepts for PCa-associated genotypic, phenotypic and lifestyle data. The current version of the PCAO2 contains 633 concepts organized under three biomedical viewpoints, namely, epidemiology, diagnosis and treatment. These concepts are enriched by the addition of definition, synonym, relationship and reference. For the precision diagnosis and treatment, the PCa-associated genes and lifestyles are integrated in the viewpoint of epidemiological aspects of PCa. PCAO2 provides a standardized and systematized semantic framework for studying large amounts of heterogeneous PCa data and knowledge, which can be further, edited and enriched by the scientific community. The PCAO2 is freely available at https://bioportal.bioontology.org/ontologies/PCAO, http://pcaontology.net/ and http://pcaontology.net/mobile/.
Assuntos
Ontologias Biológicas , Neoplasias da Próstata , Humanos , Masculino , Inteligência Artificial , Semântica , Neoplasias da Próstata/genéticaRESUMO
PURPOSE: To examine (1) how much participation is represented in the benchmark Unified Medical Language System (UMLS) resource, and (2) to what extent that representation reflects the definition of child and youth participation and/or its related constructs per the family of Participation-Related Constructs framework. MATERIALS AND METHODS: We searched and analysed UMLS concepts related to the term "participation." Identified UMLS concepts were rated according to their representation of participation (i.e., attendance, involvement, both) as well as participation-related constructs using deductive content analysis. RESULTS: 363 UMLS concepts were identified. Of those, 68 had at least one English definition, resulting in 81 definitions that were further analysed. Results revealed 2 definitions (2/81; 3%; 2/68 UMLS concepts) representing participation "attendance" and 18 definitions (18/81; 22%; 14/68 UMLS concepts) representing participation "involvement." No UMLS concept definition represented both attendance and involvement (i.e., participation). Most of the definitions (11/20; 55%; 9/16 UMLS concepts) representing attendance or involvement also represent a participation-related construct. CONCLUSION(S): The representation of participation within the UMLS is limited and poorly aligned with the contemporary definition of child and youth participation. Expanding ontological resources to represent child and youth participation is needed to enable better data analytics that reflect contemporary paediatric rehabilitation practice.
The representation of participation within the Unified Medical Language System (UMLS) is limited and poorly aligned with the contemporary definition of child and youth participation.From a contemporary paediatric rehabilitation perspective, using the current UMLS concepts for data analytics might result in misrepresentation of child and youth participation.There is need to expand ontological resources within the UMLS to fully and exclusively represent participation dimensions (attendance and involvement) in daily life activities to enable better data analytics that reflect contemporary paediatric rehabilitation practice.
RESUMO
We present a novel approach to representing perceptual and cognitive knowledge, spectral knowledge representation, that is focused on the oscillatory behaviour of the brain. The model is presented in the context of a larger hypothetical cognitive architecture. The model uses literal representations of waves to describe the dynamics of neural assemblies as they process perceived input. We show how the model can be applied to representations of sound, and usefully model music perception, specifically harmonic distance. We demonstrate that the model naturally captures both pitch and chord/key distance as empirically measured by Krumhansl and Kessler, thereby providing an underlying mechanism from which their toroidal model might arise. We evaluate our model with respect to those of Milne and others.
RESUMO
Ontologies serve as comprehensive frameworks for organizing domain-specific knowledge, offering significant benefits for managing clinical data. This study presents the development of the Fall Risk Management Ontology (FRMO), designed to enhance clinical text mining, facilitate integration and interoperability between disparate data sources, and streamline clinical data analysis. By representing major entities within the fall risk management domain, the FRMO supports the unification of clinical language and decision-making processes, ultimately contributing to the prevention of falls among older adults. We used Ontology Web Language (OWL) to build the FRMO in Protégé. Of the seven steps of the Stanford approach, six steps were utilized in the development of the FRMO: (1) defining the domain and scope of the ontology, (2) reusing existing ontologies when possible, (3) enumerating ontology terms, (4) specifying the classes and their hierarchy, (5) defining the properties of the classes, and (6) defining the facets of the properties. We evaluated the FRMO using four main criteria: consistency, completeness, accuracy, and clarity. The developed ontology comprises 890 classes arranged in a hierarchical structure, including six top-level classes with a total of 43 object properties and 28 data properties. FRMO is the first comprehensively described semantic ontology for fall risk management. Healthcare providers can use the ontology as the basis of clinical decision technology for managing falls among older adults.
Assuntos
Acidentes por Quedas , Mineração de Dados , Gestão de Riscos , Acidentes por Quedas/prevenção & controle , Humanos , Mineração de Dados/métodos , Ontologias Biológicas , Registros Eletrônicos de Saúde/organização & administração , SemânticaRESUMO
One of the big challenges in robotics is the generalization necessary for performing unknown tasks in unknown environments on unknown objects. For us humans, this challenge is simplified by the commonsense knowledge we can access. For cognitive robotics, representing and acquiring commonsense knowledge is a relevant problem, so we perform a systematic literature review to investigate the current state of commonsense knowledge exploitation in cognitive robotics. For this review, we combine a keyword search on six search engines with a snowballing search on six related reviews, resulting in 2,048 distinct publications. After applying pre-defined inclusion and exclusion criteria, we analyse the remaining 52 publications. Our focus lies on the use cases and domains for which commonsense knowledge is employed, the commonsense aspects that are considered, the datasets/resources used as sources for commonsense knowledge and the methods for evaluating these approaches. Additionally, we discovered a divide in terminology between research from the knowledge representation and reasoning and the cognitive robotics community. This divide is investigated by looking at the extensive review performed by Zech et al. (The International Journal of Robotics Research, 2019, 38, 518-562), with whom we have no overlapping publications despite the similar goals.
RESUMO
Named entity recognition (NER) is an important task for the natural language processing of biomedical text. Currently, most NER studies standardized biomedical text, but NER for unstandardized biomedical text draws less attention from researchers. Named entities in online biomedical text exist with errors and polymorphisms, which negatively impact NER models' performance and impede support from knowledge representation methods. In this paper, we propose a neural network method that can effectively recognize entities in unstandardized online medical/health text. We introduce a new pre-training scheme that uses large-scale online question-answering pairs to enhance transformers' model capacity on online biomedical text. Moreover, we supply models with knowledge representations from a knowledge base called multi-channel knowledge labels, and this method overcomes the restriction from languages, like Chinese, that require word segmentation tools to represent knowledge. Our model outperforms other baseline methods significantly in experiments on a dataset for Chinese online medical entity recognition and achieves state-of-the-art results.
Assuntos
Processamento de Linguagem Natural , Redes Neurais de ComputaçãoRESUMO
BACKGROUND: Social determinants of health (SDoH) have been described by the World Health Organization as the conditions in which individuals are born, live, work, and age. These conditions can be grouped into 3 interrelated levels known as macrolevel (societal), mesolevel (community), and microlevel (individual) determinants. The scope of SDoH expands beyond the biomedical level, and there remains a need to connect other areas such as economics, public policy, and social factors. OBJECTIVE: Providing a computable artifact that can link health data to concepts involving the different levels of determinants may improve our understanding of the impact SDoH have on human populations. Modeling SDoH may help to reduce existing gaps in the literature through explicit links between the determinants and biological factors. This in turn can allow researchers and clinicians to make better sense of data and discover new knowledge through the use of semantic links. METHODS: An experimental ontology was developed to represent knowledge of the social and economic characteristics of SDoH. Information from 27 literature sources was analyzed to gather concepts and encoded using Web Ontology Language, version 2 (OWL2) and Protégé. Four evaluators independently reviewed the ontology axioms using natural language translation. The analyses from the evaluations and selected terminologies from the Basic Formal Ontology were used to create a revised ontology with a broad spectrum of knowledge concepts ranging from the macrolevel to the microlevel determinants. RESULTS: The literature search identified several topics of discussion for each determinant level. Publications for the macrolevel determinants centered around health policy, income inequality, welfare, and the environment. Articles relating to the mesolevel determinants discussed work, work conditions, psychosocial factors, socioeconomic position, outcomes, food, poverty, housing, and crime. Finally, sources found for the microlevel determinants examined gender, ethnicity, race, and behavior. Concepts were gathered from the literature and used to produce an ontology consisting of 383 classes, 109 object properties, and 748 logical axioms. A reasoning test revealed no inconsistent axioms. CONCLUSIONS: This ontology models heterogeneous social and economic concepts to represent aspects of SDoH. The scope of SDoH is expansive, and although the ontology is broad, it is still in its early stages. To our current understanding, this ontology represents the first attempt to concentrate on knowledge concepts that are currently not covered by existing ontologies. Future direction will include further expanding the ontology to link with other biomedical ontologies, including alignment for granular semantics.
RESUMO
Electronic health records (EHRs) store an extensive array of patient information, encompassing medical histories, diagnoses, treatments, and test outcomes. These records are crucial for enabling healthcare providers to make well-informed decisions regarding patient care. Summarizing clinical notes further assists healthcare professionals in pinpointing potential health risks and making better-informed decisions. This process contributes to reducing errors and enhancing patient outcomes by ensuring providers have access to the most pertinent and current patient data. Recent research has shown that incorporating instruction prompts with large language models (LLMs) substantially boosts the efficacy of summarization tasks. However, we show that this approach also leads to increased performance variance, resulting in significantly distinct summaries even when instruction prompts share similar meanings. To tackle this challenge, we introduce a model-agnostic Soft Prompt-BasedCalibration (SPeC) pipeline that employs soft prompts to lower variance while preserving the advantages of prompt-based summarization. Experimental findings on multiple clinical note tasks and LLMs indicate that our method not only bolsters performance but also effectively regulates variance across different LLMs, providing a more consistent and reliable approach to summarizing critical medical information.
Assuntos
Registros Eletrônicos de Saúde , Processamento de Linguagem Natural , Humanos , Calibragem , Idioma , Pessoal de SaúdeRESUMO
This study proposes an ontology model to represent the surgical pathways for appendicectomy, incorporating domain knowledge extracted from literature and electronic records in an Australian health district. The ontology comprised 108 concepts and 81 object properties. The model is useful for continuous quality improvement initiatives, i.e., process mining to understand what happens in the hospital versus what is required by the policy.