The Intelligent Diagnostic System for Common Diseases using the Optimized Medical Knowledge Graph

2020 ◽  
Author(s):  
Jing Wang ◽  
Ning Liu ◽  
Qian Hu ◽  
Zhichao Wu ◽  
Wan Dong ◽  
...  
Keyword(s):  
Medical Knowledge ◽  
Diagnostic System ◽  
Knowledge Graph ◽  
Common Diseases

scholarly journals End-to-End Knowledge-Routed Relational Dialogue System for Automatic Diagnosis

2019 ◽  
Vol 33 ◽  
pp. 7346-7353
Author(s):  
Lin Xu ◽  
Qixian Zhou ◽  
Ke Gong ◽  
Xiaodan Liang ◽  
Jianheng Tang ◽  
...  
Keyword(s):  
Natural Language ◽  
Medical Diagnosis ◽  
Expert Knowledge ◽  
Medical Knowledge ◽  
Knowledge Graph ◽  
Dialogue Systems ◽  
Dialogue System ◽  
Self Reports ◽  
Medical Dialogue ◽  
End To End

Beyond current conversational chatbots or task-oriented dialogue systems that have attracted increasing attention, we move forward to develop a dialogue system for automatic medical diagnosis that converses with patients to collect additional symptoms beyond their self-reports and automatically makes a diagnosis. Besides the challenges for conversational dialogue systems (e.g. topic transition coherency and question understanding), automatic medical diagnosis further poses more critical requirements for the dialogue rationality in the context of medical knowledge and symptom-disease relations. Existing dialogue systems (Madotto, Wu, and Fung 2018; Wei et al. 2018; Li et al. 2017) mostly rely on datadriven learning and cannot be able to encode extra expert knowledge graph. In this work, we propose an End-to-End Knowledge-routed Relational Dialogue System (KR-DS) that seamlessly incorporates rich medical knowledge graph into the topic transition in dialogue management, and makes it cooperative with natural language understanding and natural language generation. A novel Knowledge-routed Deep Q-network (KR-DQN) is introduced to manage topic transitions, which integrates a relational refinement branch for encoding relations among different symptoms and symptomdisease pairs, and a knowledge-routed graph branch for topic decision-making. Extensive experiments on a public medical dialogue dataset show our KR-DS significantly beats stateof-the-art methods (by more than 8% in diagnosis accuracy). We further show the superiority of our KR-DS on a newly collected medical dialogue system dataset, which is more challenging retaining original self-reports and conversational data between patients and doctors.

scholarly journals SMR: Medical Knowledge Graph Embedding for Safe Medicine Recommendation

2021 ◽  
Vol 23 ◽  
pp. 100174
Author(s):  
Fan Gong ◽  
Meng Wang ◽  
Haofen Wang ◽  
Sen Wang ◽  
Mengyue Liu
Keyword(s):  
Medical Knowledge ◽  
Graph Embedding ◽  
Knowledge Graph

scholarly journals KGPA: Construction of Knowledge Graph for Pituitary Adenoma (Preprint)

2021 ◽  
Author(s):  
An Fang ◽  
Pei Lou ◽  
Jiahui Hu ◽  
Wanqing Zhao ◽  
Ming Feng ◽  
...  
Keyword(s):  
Pituitary Adenoma ◽  
Data Fusion ◽  
Medical Knowledge ◽  
Medical Data ◽  
Knowledge Graph ◽  
Fusion Method ◽  
Fusion Model ◽  
Construction Of Knowledge ◽  
Icd 10 ◽  
Disease Entities

BACKGROUND Pituitary adenoma is one of the most common central nervous system tumors. The diagnosis and treatment of pituitary adenoma are still very difficult. Misdiagnosis and recurrence occur from time to time, and experienced neurosurgeons are in serious shortage. Knowledge graphs can help interns quickly understand the medical knowledge related to pituitary tumor. OBJECTIVE The aim of this paper is to integrate the data of pituitary adenomas from reliable sources and construct a knowledge graph, and use the knowledge graph for knowledge discovery. METHODS A method of constructing a knowledge graph of diseases was introduced and used to build a knowledge graph for pituitary adenoma (KGPA). The schema of the KGPA was manually constructed. Information of pituitary adenoma were automatically extracted from EMR and the medical websites through the CRF model and web wrappers we designed. An entity fusion method was proposed, based on the head and tail entity fusion models, to fuse the data from heterogeneous sources. The disease entities were standardized to ICD-10. RESULTS Data was extracted from 300 EMRs of pituitary adenoma and 4 medical portals. Entity fusion was carried out by using the data fusion model we proposed. The accuracy of the head and tail entity fusion were more than 97%. Part of the triples were selected for evaluation, and the accuracy was 95.4%. CONCLUSIONS This paper introduced an approach to construct KGPA and proposed a data fusion method suitable for medical data. The evaluation results show that the data in KGPA is of high quality. The constructed KGPA can help physicians in their clinical practice.

scholarly journals Ontology-Oriented Diagnostic System for Traditional Chinese Medicine Based on Relation Refinement

2013 ◽  
Vol 2013 ◽  
pp. 1-11 ◽  
Author(s):  
Peiqin Gu ◽  
Huajun Chen ◽  
Tong Yu
Keyword(s):  
Chinese Medicine ◽  
Traditional Chinese Medicine ◽  
Medical Knowledge ◽  
Diagnostic System ◽  
Syndrome Differentiation ◽  
Diagnosis System ◽  
Knowledge Based ◽  
Interpretable Model ◽  
Online Web ◽  
Clinical Cases

Although Chinese medicine treatments have become popular recently, the complicated Chinese medical knowledge has made it difficult to be applied in computer-aided diagnostics. The ability to model and use the knowledge becomes an important issue. In this paper, we define the diagnosis in Traditional Chinese Medicine (TCM) as discovering the fuzzy relations between symptoms and syndromes. An Ontology-oriented Diagnosis System (ODS) is created to address the knowledge-based diagnosis based on a well-defined ontology of syndromes. The ontology transforms the implicit relationships among syndromes into a machine-interpretable model. The clinical data used for feature selection is collected from a national TCM research institute in China, which serves as a training source for syndrome differentiation. The ODS analyzes the clinical cases to obtain a statistical mapping relation between each syndrome and associated symptom set, before rechecking the completeness of related symptoms via ontology refinement. Our diagnostic system provides an online web interface to interact with users, so that users can perform self-diagnosis. We tested 12 common clinical cases on the diagnosis system, and it turned out that, given the agree metric, the system achieved better diagnostic accuracy compared to nonontology method—92% of the results fit perfectly with the experts’ expectations.

Drug Repurposing for Parkinson’s Disease by Integrating Knowledge Graph Completion Model and Knowledge Fusion of Medical Literature

2021 ◽  
Vol 13 (1) ◽  
pp. 14
Author(s):  
Xiaolin Zhang ◽  
Chao Che
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Knowledge Base ◽  
Medical Literature ◽  
Medical Knowledge ◽  
Drug Repurposing ◽  
New Drugs ◽  
Knowledge Graph ◽  
Knowledge Fusion ◽  
Effective Drugs

The prevalence of Parkinson’s disease increases a tremendous medical and economic burden to society. Therefore, the effective drugs are urgently required. However, the traditional development of effective drugs is costly and risky. Drug repurposing, which identifies new applications for existing drugs, is a feasible strategy for discovering new drugs for Parkinson’s disease. Drug repurposing is based on sufficient medical knowledge. The local medical knowledge base with manually labeled data contains a large number of accurate, but not novel, medical knowledge, while the medical literature containing the latest knowledge is difficult to utilize, because of unstructured data. This paper proposes a framework, named Drug Repurposing for Parkinson’s disease by integrating Knowledge Graph Completion method and Knowledge Fusion of medical literature data (DRKF) in order to make full use of a local medical knowledge base containing accurate knowledge and medical literature with novel knowledge. DRKF first extracts the relations that are related to Parkinson’s disease from medical literature and builds a medical literature knowledge graph. After that, the literature knowledge graph is fused with a local medical knowledge base that integrates several specific medical knowledge sources in order to construct a fused medical knowledge graph. Subsequently, knowledge graph completion methods are leveraged to predict the drug candidates for Parkinson’s disease by using the fused knowledge graph. Finally, we employ classic machine learning methods to repurpose the drug for Parkinson’s disease and compare the results with the method only using the literature-based knowledge graph in order to confirm the effectiveness of knowledge fusion. The experiment results demonstrate that our framework can achieve competitive performance, which confirms the effectiveness of our proposed DRKF for drug repurposing against Parkinson’s disease. It could be a supplement to traditional drug discovery methods.

scholarly journals Path-based knowledge reasoning with textual semantic information for medical knowledge graph completion

2021 ◽  
Vol 21 (S9) ◽  
Author(s):  
Yinyu Lan ◽  
Shizhu He ◽  
Kang Liu ◽  
Xiangrong Zeng ◽  
Shengping Liu ◽  
...  
Keyword(s):  
Semantic Information ◽  
State Of The Art ◽  
Semantic Representation ◽  
Medical Knowledge ◽  
The State ◽  
Language Models ◽  
Knowledge Graph ◽  
Knowledge Reasoning ◽  
Numerical Computing ◽  
Knowledge Graphs

Abstract Background Knowledge graphs (KGs), especially medical knowledge graphs, are often significantly incomplete, so it necessitating a demand for medical knowledge graph completion (MedKGC). MedKGC can find new facts based on the existed knowledge in the KGs. The path-based knowledge reasoning algorithm is one of the most important approaches to this task. This type of method has received great attention in recent years because of its high performance and interpretability. In fact, traditional methods such as path ranking algorithm take the paths between an entity pair as atomic features. However, the medical KGs are very sparse, which makes it difficult to model effective semantic representation for extremely sparse path features. The sparsity in the medical KGs is mainly reflected in the long-tailed distribution of entities and paths. Previous methods merely consider the context structure in the paths of knowledge graph and ignore the textual semantics of the symbols in the path. Therefore, their performance cannot be further improved due to the two aspects of entity sparseness and path sparseness. Methods To address the above issues, this paper proposes two novel path-based reasoning methods to solve the sparsity issues of entity and path respectively, which adopts the textual semantic information of entities and paths for MedKGC. By using the pre-trained model BERT, combining the textual semantic representations of the entities and the relationships, we model the task of symbolic reasoning in the medical KG as a numerical computing issue in textual semantic representation. Results Experiments results on the publicly authoritative Chinese symptom knowledge graph demonstrated that the proposed method is significantly better than the state-of-the-art path-based knowledge graph reasoning methods, and the average performance is improved by 5.83% for all relations. Conclusions In this paper, we propose two new knowledge graph reasoning algorithms, which adopt textual semantic information of entities and paths and can effectively alleviate the sparsity problem of entities and paths in the MedKGC. As far as we know, it is the first method to use pre-trained language models and text path representations for medical knowledge reasoning. Our method can complete the impaired symptom knowledge graph in an interpretable way, and it outperforms the state-of-the-art path-based reasoning methods.

scholarly journals A Method to Learn Embedding of a Probabilistic Medical Knowledge Graph: Algorithm Development (Preprint)

2019 ◽  
Author(s):  
Linfeng Li ◽  
Peng Wang ◽  
Yao Wang ◽  
Shenghui Wang ◽  
Jun Yan ◽  
...  
Keyword(s):  
Medical Records ◽  
Large Scale ◽  
Semantic Representation ◽  
Medical Knowledge ◽  
Mapping Function ◽  
Graph Algorithm ◽  
Knowledge Graph ◽  
Knowledge Graphs ◽  
Representation Method ◽  
Better Than

BACKGROUND Knowledge graph embedding is an effective semantic representation method for entities and relations in knowledge graphs. Several translation-based algorithms, including TransE, TransH, TransR, TransD, and TranSparse, have been proposed to learn effective embedding vectors from typical knowledge graphs in which the relations between head and tail entities are deterministic. However, in medical knowledge graphs, the relations between head and tail entities are inherently probabilistic. This difference introduces a challenge in embedding medical knowledge graphs. OBJECTIVE We aimed to address the challenge of how to learn the probability values of triplets into representation vectors by making enhancements to existing TransX (where X is E, H, R, D, or Sparse) algorithms, including the following: (1) constructing a mapping function between the score value and the probability, and (2) introducing probability-based loss of triplets into the original margin-based loss function. METHODS We performed the proposed PrTransX algorithm on a medical knowledge graph that we built from large-scale real-world electronic medical records data. We evaluated the embeddings using link prediction task. RESULTS Compared with the corresponding TransX algorithms, the proposed PrTransX performed better than the TransX model in all evaluation indicators, achieving a higher proportion of corrected entities ranked in the top 10 and normalized discounted cumulative gain of the top 10 predicted tail entities, and lower mean rank. CONCLUSIONS The proposed PrTransX successfully incorporated the uncertainty of the knowledge triplets into the embedding vectors.

Sign in / Sign up

Export Citation Format

Share Document