Medical Domain Knowledge Collaborative Graph Learning for Healthcare Event Prediction

BackgroundDeep learning models have attracted significant interest from health care researchers during the last few decades. There have been many studies that apply deep learning to medical applications and achieve promising results. However, there are three limitations to the existing models: (1) most clinicians are unable to interpret the results from the existing models, (2) existing models cannot incorporate complicated medical domain knowledge (eg, a disease causes another disease), and (3) most existing models lack visual exploration and interaction. Both the electronic health record (EHR) data set and the deep model results are complex and abstract, which impedes clinicians from exploring and communicating with the model directly.ObjectiveThe objective of this study is to develop an interpretable and accurate risk prediction model as well as an interactive clinical prediction system to support EHR data exploration, knowledge graph demonstration, and model interpretation.MethodsA domain-knowledge–guided recurrent neural network (DG-RNN) model is proposed to predict clinical risks. The model takes medical event sequences as input and incorporates medical domain knowledge by attending to a subgraph of the whole medical knowledge graph. A global pooling operation and a fully connected layer are used to output the clinical outcomes. The middle results and the parameters of the fully connected layer are helpful in identifying which medical events cause clinical risks. DG-Viz is also designed to support EHR data exploration, knowledge graph demonstration, and model interpretation.ResultsWe conducted both risk prediction experiments and a case study on a real-world data set. A total of 554 patients with heart failure and 1662 control patients without heart failure were selected from the data set. The experimental results show that the proposed DG-RNN outperforms the state-of-the-art approaches by approximately 1.5%. The case study demonstrates how our medical physician collaborator can effectively explore the data and interpret the prediction results using DG-Viz.ConclusionsIn this study, we present DG-Viz, an interactive clinical prediction system, which brings together the power of deep learning (ie, a DG-RNN–based model) and visual analytics to predict clinical risks and visually interpret the EHR prediction results. Experimental results and a case study on heart failure risk prediction tasks demonstrate the effectiveness and usefulness of the DG-Viz system. This study will pave the way for interactive, interpretable, and accurate clinical risk predictions.

PurposeInternet technology allows millions of people to find high quality medical resources online, with the result that personal healthcare and medical services have become one of the fastest growing markets in China. Data relating to healthcare search behavior may provide insights that could lead to better provision of healthcare services. However, discrepancies often arise between terminologies derived from professional medical domain knowledge and the more colloquial terms that users adopt when searching for information about ailments. This can make it difficult to match healthcare queries with doctors’ keywords in online medical searches. The paper aims to discuss these issues.Design/methodology/approachTo help address this problem, the authors propose a transfer learning using latent factor graph (TLLFG), which can learn the descriptions of ailments used in internet searches and match them to the most appropriate formal medical keywords.FindingsExperiments show that the TLLFG outperforms competing algorithms in incorporating both medical domain knowledge and patient-doctor Q&A data from online services into a unified latent layer capable of bridging the gap between lay enquiries and professionally expressed information sources, and make more accurate analysis of online users’ symptom descriptions. The authors conclude with a brief discussion of some of the ways in which the model may support online applications and connect offline medical services.Practical implicationsThe authors used an online medical searching application to verify the proposed model. The model can bridge users’ long-tailed description with doctors’ formal medical keywords. Online experiments show that TLLFG can significantly improve the searching experience of both users and medical service providers compared with traditional machine learning methods. The research provides a helpful example of the use of domain knowledge to optimize searching or recommendation experiences.Originality/valueThe authors use transfer learning to map online users’ long-tail queries onto medical domain knowledge, significantly improving the relevance of queries and keywords in a search system reliant on sponsored links.

Clinical decision support system for hypertension medication based on knowledge graph

The text-guided diffusion model GLIDE (Guided Language to Image Diffusion for Generation and Editing) is the state of the art in text-to-image generative artificial intelligence (AI). GLIDE has rich representations, but medical applications of this model have not been systematically explored. If GLIDE had useful medical knowledge, it could be used for medical image analysis tasks, a domain in which AI systems are still highly engineered towards a single use-case. Here we show that the publicly available GLIDE model has reasonably strong representations of key topics in cancer research and oncology, in particular the general style of histopathology images and multiple facets of diseases, pathological processes and laboratory assays. However, GLIDE seems to lack useful representations of the style and content of radiology data. Our findings demonstrate that domain-agnostic generative AI models can learn relevant medical concepts without explicit training. Thus, GLIDE and similar models might be useful for medical image processing tasks in the future - particularly with additional domain-specific fine-tuning.

In recent years, with the better availability of medical data such as Electronic Health Records (EHR), more and more data mining models have been developed to explore the data-driven insights for better human health. However, there are many challenges for analyzing EHR such as high-dimensionality, temporality, sparsity, etc., which make the data-driven models less reliable. Medical knowledge graph (MKG), which encodes comprehensive knowledge about the medical concepts and relationships extracted from medical literature, holds great promise to regularize the data-driven models as prior knowledge. Nonetheless, the MKGs are typically not complete, which limits its utility in helping with the data mining process. In this paper, we propose a mutual enhancement framework MendMKG for predictive modeling of clinical events with both EHR and MKG. In particular, MendMKG first conducts a self-supervised learning strategy to simultaneously pre-train a graph attention network for embedding nodes and complete the MKG. It iteratively performs (1) an embedding-based knowledge graph completion module to derive missing edges, (2) and a reconstruction module of unlabeled EHR data to select high-quality ones from these edges, which would be further appended to the MKG to update the embedding model. Through the iterations, the two modules mutually benefit each other. Then, MendMKG uses the pre-trained graph attention network and the updated MKG to generate the visit embeddings to represent patient’s historical visits, and predict the diagnosis in future visit, through a fine-tuning approach. Experimental results on real world EHR corpus are provided to demonstrate the superiority of the proposed framework, compared to a series of state-of-the-art baselines. <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> The source code and knowledge graph data have been anonymously uploaded to https://github.com/1317375434/MendMKG.

Acute hospital care as performed in the intensive care unit (ICU) is characterized by its frequent, but short-term interventions for patients who are severely ill. Because clinicians have to attend to more than one patient at a time and make decisions in a limited time in acute hospital care environments, the accurate prediction of the in-hospital mortality risk could assist them to pay more attention to patients with a higher in-hospital mortality risk, thereby improving the quality and efficiency of the care. One of the salient features of ICU is the diversity of patients: clinicians are faced by patients with a wide variety of diseases. However, mortality prediction for ICU patients has typically been conducted by building one common predictive model for all the diseases. In this paper, we incorporate disease-specific contexts into mortality modeling by formulating the mortality prediction problem as a multi-task learning problem in which a task corresponds to a disease. Our method effectively integrates medical domain knowledge relating to the similarity among diseases and the similarity among Electronic Health Records (EHRs) into a data-driven approach by incorporating graph Laplacians into the regularization term to encode these similarities. The experimental results on a real dataset from a hospital corroborate the effectiveness of the proposed method. The AUCs of several baselines were improved, including logistic regression without multi-task learning and several multi-task learning methods that do not incorporate the domain knowledge. In addition, we illustrate some interesting results pertaining to disease-specific predictive features, some of which are not only consistent with existing medical domain knowledge, but also contain suggestive hypotheses that could be validated by further investigations in the medical domain.

Entity alignment (EA) is the task of finding equivalent entities in various knowledge graphs (KGs) that represent the same object in the real world, which aims at building a unified large-scale knowledge graph with rich and high-quality content. Existing EA works have achieved promising performance. These works mainly focus on learning the entity embeddings by capturing the structural semantics of entities on the relational triples in KG. Some works also take utilization of attributes information in entity embeddings. However, they are still hindered by the lack of prior seed alignment known as labeled training data. In this paper, we propose a novel approach for KG entity alignment based on transfer learning named DAEA, which incorporates both attributes triples and relational triples to generate the entity embeddings. DAEA first uses the J-GCN model to learn the entity representation model in source domain KG. Then, it combines domain adversarial network to learn entity representation model of the target domain. Since DAEA can transfer common entity information from the source domain to the target domain, it mitigates the problem of poor alignment performance due to lack of prior seed alignment. Experiments results show that DAEA significantly outperforms the state-of-the-art EA approaches on both monolingual and cross-lingual entity alignment datasets. Ablation studies on different components further demonstrate the effectiveness of our approach.

The purpose of this study was to determine the ability to understand students' mathematical concepts and the responses of students who applied the collaborative teamwork learning model and guided note-taking model, as well as to find out the differences in understanding the mathematical concepts of students who applied the collaborative teamwork learning model and the guided note taking model. This study uses a quantitative approach with experimental methods and a quasi-experimental research design in the form of a posttest-only design. The population of this study were students of class X MAN 1 Cirebon, Cirebon Regency, then selected samples of class X IIS 3 as experimental class II with guided note-taking model and X IIS 4 as experimental class I with collaborative teamwork learning model. Sampling is done by cluster random sampling technique. Through the prerequisite test, namely the normality and homogeneity test, the research data obtained were normally distributed and had homogeneous variance. After conducting the prerequisite test, the researcher tested the hypothesis using the Independent Sample T-Test with the asymp value. Sig (2-tailed) of 0.000. This value is smaller than 0.05, so Ha is accepted. In other words, there are differences in the understanding of mathematical concepts between classes that apply the collaborative teamwork learning model with guided note-taking models on the subject of the sine and cosine rules in class X IIS MAN Cirebon I Cirebon Regency. The average value of the class that applies the collaborative teamwork learning model is 72.38, with 11 students achieving the KKM (Minimum Completeness Criteria), and the guided note-taking model of 78.52, with 29 students completing the KKM (Minimum Completeness Criteria). Both models received a good response from students, with a response of 82% for the collaborative teamwork learning model and 76% for the guided note-taking model.

With the rapid advancement of artificial intelligence technology, knowledge graphs are seeing increasingly widespread application in the medical field, offering new technological means for assisted medical diagnosis. However, current research still faces challenges such as difficulties in natural language understanding and insufficient knowledge reasoning capabilities. This paper designs and implements a knowledge graph-based question answering system for assisted medical diagnosis, aiming to enhance diagnostic efficiency and accuracy through structured medical knowledge. The system first uses natural language processing techniques to analyze user-input medical questions, and then performs semantic reasoning by leveraging entities such as diseases, symptoms, medications, and medical tests, along with their relationships in a medical knowledge graph, to generate reliable diagnostic suggestions. Through knowledge graph embedding and relationship path mining, the system achieves multi-hop reasoning and ensures the scientific validity of responses based on authoritative medical databases. Experimental results demonstrate that this system achieves high accuracy and practicality in diagnostic question answering tasks for common diseases. It can provide efficient decision support assistance for both doctors and patients while reducing the risk of misdiagnosis.

To meet the high-efficiency question answering needs of existing patients and doctors, this system integrates medical professional knowledge, knowledge graphs, and question answering systems that conduct man-machine dialogue through natural language. This system locates the medical field, uses crawler technology to use vertical medical websites as data sources, and uses diseases as the core entity to construct a knowledge graph containing 44,000 knowledge entities of 7 types and 300,000 entities of 11 kinds. It is stored in the Neo4j graph database, using rule-based matching methods and string-matching algorithms to construct a domain lexicon to classify and query questions. This system has specific practical value in the medical field knowledge graph and question answering system.

Background The field of information extraction (IE) is currently exploring more versatile and efficient methods for minimization of reliance on extensive annotated datasets and integration of knowledge across tasks and domains. Objective We aim to evaluate and refine the application of the universal IE (UIE) technology in the field of Chinese medical expertise in terms of processing accuracy and efficiency. Methods Our model integrates ontology modeling, web scraping, UIE, fine-tuning strategies, and graph databases, thereby covering knowledge modeling, extraction, and storage techniques. The Enhanced Representation through Knowledge Integration-UIE (ERNIE-UIE) model is fine-tuned and optimized using a small amount of annotated data. A medical knowledge graph is then constructed, followed by validating the graph and conducting knowledge mining on the data stored within it. Results Incorporating the characteristics of whole-course management, we implemented a comprehensive medical knowledge graph–construction model and methodology. Entities and relationships were jointly extracted using the pretrained language model, resulting in 8,525 entity data points and 9,522 triple data points. The accuracy of the knowledge graph was verified using graph algorithms. Conclusion We optimized the construction process of a Chinese medical knowledge graph with minimal annotated data by utilizing a generative extraction paradigm, validating the graph’s efficacy and achieving commendable results. This approach addresses the challenge of insufficient annotated training corpora in low-resource knowledge graph construction, thereby contributing to cost savings in the development of knowledge graphs.

The mathematical generalization ability of students at one junior high school, especially in Makassar, is relatively low. One alternative to improve students' mathematical generalization abilities is to apply active, collaborative and inductive cooperative learning models. This research aims to determine the difference in the average score of mathematical generalization abilities of students taught through active, collaborative and inductive cooperative learning models, and students taught through cooperative learning models on flat-sided geometric material. This research used a quasi-experimental design with a pretest-posttest control group design in class VIII at a junior high school in Makassar, Indonesia. The sample in this study was 35 students in the control group and 31 students in the experimental group. The instrument used is a mathematical generalization ability test. Data were analyzed using the n gain test. Based on the analysis, the average n-gain score for students' mathematical generalization abilities in flat-sided geometric material after being taught by applying active, collaborative and inductive cooperative learning models was 0,51 or experienced a moderate increase and the percentage of learning completeness was 93,55%. Therefore, it can be concluded that junior high school students' mathematical generalization abilities increased after being taught using active, collaborative and inductive cooperative learning models.

Background Problems: In the learning process at SMA Negeri 2 Singaraj, researchers found problems in the learning process in the basketball game material, especially in the basic techniques of passing chest passes and bounce passes, which had a low average below the minimum completion criteria. Research Objectives: This research aims to improve learning outcomes in basketball, the basic technique of passing chest passes and bounce passes through the Collaborative Teamwork Learning model for class XI A1 students at State Senior High School 2 Singaraja. Methods: Classroom action research (CAR) is used. This classroom action research (CAR) uses teachers as researchers and consists of 2 cycles. The subjects of this research were students in class XI A1 of State Senior High School 2 Singaraja, with 40 people consisting of 20 male students and 20 female students. The research instrument is a description test for the knowledge aspect and performance for the skills aspect. Data analysis uses qualitative and quantitative descriptive analysis. Results: The results of the research showed an increase, namely learning completeness in the knowledge aspect in cycle I, which was (85.5%), in cycle II it increased to (92.5%). Furthermore, learning completeness in the skills aspect in cycle I was (87.5%) in cycle II (92.5%). Conclusion: Based on data analysis and discussion, it is concluded that the collaborative teamwork learning model can improve PJOK learning outcomes in basketball passing material for class XI A1 students at SMA Negeri 2 Singaraja for the 2023/2024 academic year. Suggestions for PJOK teachers to be able to implement the Collaborative Teamwork Learning model in other learning materials that are more interesting and effective. The research implies that the Collaborative Teamwork Learning model is proven to be able to improve student learning outcomes.

Collaborative learning has been utilised throughout the years to deliver educational content for various levels of students in schools. Previously, less emphasis was given to the online support mode as the conventional method exists. However, as schools across the globe closed in response to the Covid-19 pandemic, students from various levels experienced a severe disruption to their learning process. Though most schools quickly began offering some online learning, there have been growing concerns about the effects of this unprecedented shift. In addition, such a paradigm shift has also witnessed parents' significant role in ensuring the effective delivery of educational content through online platforms. In such a situation, parents were expected to support the learning process. However, less support from other stakeholders makes it challenging for parents to support their children's learning online. Although various strategies and best practices exist globally, engaging students, parents, teachers, and relevant stakeholders towards collaborative online learning is crucial. Based on relevant previous studies, engagement towards online learning is highly influenced by the user's cultural background in getting successful collaboration among all stakeholders. Thus, the proposed study aims to investigate and propose a conceptual framework for a collaborative online learning model based on a cultural-based approach. It applies a Systematic Literature Review (SLR) based on relevant previous studies. A preliminary study has conducted a comparative review based on strategies and best practices. Initial findings divulge that cultural background influences the teacher-parent engagement towards collaborative online learning provided by the school and relevant authorities. Therefore, findings from the proposed study could improve the quality of learning in primary schools, in line with SDG4 and 4IR.

Objective To identify the reasonable “early exposure to clinical practice” teaching activities of 8-year medical students under the guidance of modern medical model education.Methods 72 teacher questionnaires and 13 student reports were used to survey the understanding,purpose,effectiveness as well as management improvement comments andsuggestions for “early exposure to clinical practice”.Results 83.3％ of teachers ensured the purpose of “early exposure to clinical practice”.59.7％ of teachers were interested in the activities.70.8％ of teachers agreed to continue the current content.Only 31.9％ of teachers thought that the activities achieved the desired purpose,nevertheless 100％ of the students thought that they benefited from the activities and achieved perceptual knowledge of clinical work.61.5％ of students considered a great harvest from “early exposure to clinical practice”.Conclusions The “early exposure to clinical practice” had an active effect on increasing the perceptual knowledge and stimulating the learning interest of students.Teachers should use the modern medical education mode to explore the most suitable teaching content and implementation methods to coincide for training students the ability of “independent learning” and “collaborative learning”,and to prepare for the reform of the future integration of clinical education courses. Key words: Early exposure to clinical practice; 8-year program; Medical students; Teaching reform