Causal knowledge graph construction and evaluation for clinical decision support of diabetic nephropathy

Lyu, Kewei; Tian, Yu; Shang, Yong; Zhou, Tianshu; Yang, Ziyue; Liu, Qianghua; Yao, Xi; Zhang, Ping; Chen, Jianghua; Li, Jingsong

doi:10.1016/j.jbi.2023.104298

Cited by 20 publications

(5 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Knowledge graphs have been widely used in various fields, such as medicine, network security, journalism, finance, and education [ 10 ]. Knowledge graphs in the biomedical domain have applications in studies related to disease associations [ 11 ], genomics [ 12 ], drug interactions [ 13 ], and support for physicians in formulating individualized treatment regimens [ 14 ]. At present, there are well-established knowledge graphs, including DisGeNET [ 15 ], which integrate information on the associations between genes and diseases; DrugBank [ 16 ], a comprehensive bioinformatics and cheminformatics knowledge base; and ClinVar [ 17 ], a compilation of genetic variation information from diverse laboratories worldwide.…”

Section: Introductionmentioning

confidence: 99%

Knowledge Graph for Breast Cancer Prevention and Treatment: Literature-Based Data Analysis Study

Jin,

Liang,

Zhang

et al. 2024

JMIR Med Inform

View full text Add to dashboard Cite

Background The incidence of breast cancer has remained high and continues to rise since the 21st century. Consequently, there has been a significant increase in research efforts focused on breast cancer prevention and treatment. Despite the extensive body of literature available on this subject, systematic integration is lacking. To address this issue, knowledge graphs have emerged as a valuable tool. By harnessing their powerful knowledge integration capabilities, knowledge graphs offer a comprehensive and structured approach to understanding breast cancer prevention and treatment. Objective We aim to integrate literature data on breast cancer treatment and prevention, build a knowledge graph, and provide support for clinical decision-making. Methods We used Medical Subject Headings terms to search for clinical trial literature on breast cancer prevention and treatment published on PubMed between 2018 and 2022. We downloaded triplet data from the Semantic MEDLINE Database (SemMedDB) and matched them with the retrieved literature to obtain triplet data for the target articles. We visualized the triplet information using NetworkX for knowledge discovery. Results Within the scope of literature research in the past 5 years, malignant neoplasms appeared most frequently (587/1387, 42.3%). Pharmacotherapy (267/1387, 19.3%) was the primary treatment method, with trastuzumab (209/1805, 11.6%) being the most commonly used therapeutic drug. Through the analysis of the knowledge graph, we have discovered a complex network of relationships between treatment methods, therapeutic drugs, and preventive measures for different types of breast cancer. Conclusions This study constructed a knowledge graph for breast cancer prevention and treatment, which enabled the integration and knowledge discovery of relevant literature in the past 5 years. Researchers can gain insights into treatment methods, drugs, preventive knowledge regarding adverse reactions to treatment, and the associations between different knowledge domains from the graph.

show abstract

Section: Introductionmentioning

confidence: 99%

Knowledge Graph for Breast Cancer Prevention and Treatment: Literature-Based Data Analysis Study

Jin,

Liang,

Zhang

et al. 2024

JMIR Med Inform

View full text Add to dashboard Cite

show abstract

“…Real-world data (RWD) refers to data related to patient health status or healthcare delivery that is routinely collected from various sources, including electronic health records (EHRs) [1] . Due to the limitations of existing clinical trials, such as high costs and small patient populations, research using RWD accumulated in hospitals is actively being conducted to bridge the gap between clinical research and actual data [2] , [3] , [4] , [5] , [6] . Although RWD also has limitations like potential issues with information quality, confounding variables, and bias, real-world evidence (RWE) may be more reliable and usable than randomized controlled trials (RCT) in detecting side effects.…”

Section: Introductionmentioning

confidence: 99%

Are polypharmacy side effects predicted by public data still valid in real-world data?

Kee,

Kang,

Ahn

et al. 2024

Heliyon

View full text Add to dashboard Cite

“…Zhou et al [ 17 ] used 124 medical records to construct a knowledge graph for recommending hypertension medication. Lyu et al [ 18 ] created a knowledge graph for diabetic nephropathy diagnosis using patient data. Lin et al [ 19 ] extracted knowledge from medical texts and historical prescription data to construct a medical knowledge graph and accurately detect clinical prescription risks.…”

Section: Introductionmentioning

confidence: 99%

Explainable AI Method for Tinnitus Diagnosis via Neighbor-Augmented Knowledge Graph and Traditional Chinese Medicine: Development and Validation Study

Yin,

Kuang,

Zhang

et al. 2024

JMIR Med Inform

View full text Add to dashboard Cite

Background Tinnitus diagnosis poses a challenge in otolaryngology owing to an extremely complex pathogenesis, lack of effective objectification methods, and factor-affected diagnosis. There is currently a lack of explainable auxiliary diagnostic tools for tinnitus in clinical practice. Objective This study aims to develop a diagnostic model using an explainable artificial intelligence (AI) method to address the issue of low accuracy in tinnitus diagnosis. Methods In this study, a knowledge graph–based tinnitus diagnostic method was developed by combining clinical medical knowledge with electronic medical records. Electronic medical record data from 1267 patients were integrated with traditional Chinese clinical medical knowledge to construct a tinnitus knowledge graph. Subsequently, weights were introduced, which measured patient similarity in the knowledge graph based on mutual information values. Finally, a collaborative neighbor algorithm was proposed, which scored patient similarity to obtain the recommended diagnosis. We conducted 2 group experiments and 1 case derivation to explore the effectiveness of our models and compared the models with state-of-the-art graph algorithms and other explainable machine learning models. Results The experimental results indicate that the method achieved 99.4% accuracy, 98.5% sensitivity, 99.6% specificity, 98.7% precision, 98.6% F1-score, and 99% area under the receiver operating characteristic curve for the inference of 5 tinnitus subtypes among 253 test patients. Additionally, it demonstrated good interpretability. The topological structure of knowledge graphs provides transparency that can explain the reasons for the similarity between patients. Conclusions This method provides doctors with a reliable and explainable diagnostic tool that is expected to improve tinnitus diagnosis accuracy.

show abstract

Causal knowledge graph construction and evaluation for clinical decision support of diabetic nephropathy

Cited by 20 publications

References 39 publications

Knowledge Graph for Breast Cancer Prevention and Treatment: Literature-Based Data Analysis Study

Knowledge Graph for Breast Cancer Prevention and Treatment: Literature-Based Data Analysis Study

Are polypharmacy side effects predicted by public data still valid in real-world data?

Explainable AI Method for Tinnitus Diagnosis via Neighbor-Augmented Knowledge Graph and Traditional Chinese Medicine: Development and Validation Study

Contact Info

Product

Resources

About