Prescription Prediction towards Computer-Assisted Diagnosis for Kampo Medicine

Mi, Xiaoyu; Ikeda, Hiroshi; Nakazawa, Fumio; Matsuoka, Hidetoshi; Kataoka, Erika; Hamaya, Satoshi; Odaguchi, Hiroshi; Ishige, Tatsuya; Itô, Yûichi; Wakasugi, Akino; Kawanabe, Tadaaki; Sekine, Mariko; Hanawa, Toshihiko; Yamaguchi, Shinichi

doi:10.1109/ccats.2015.38

Cited by 4 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, a great deal of preventive healthcare research utilized predictive models with electronic health record data [8,9,10,12,13,14], and studies focusing on predictive models for NCDs have also been reported [23,24].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning Based Prediction of Non-communicable Diseases to Improving Intervention Program in Bangladesh

Hu¹,

Nohara²,

Wakata³

et al. 2018

ejbi

View full text Add to dashboard Cite

The prevention of non-communicable diseases (NCDs) has interested researchers in recent decades. According to the World Health Organization, NCD causes more deaths annually than all other causes combine, making it the leading cause of death globally [1]. A study [2] implementing an NCD prevention program using information communication technology was conducted by Kyushu University and Grameen Communications. A Portable Health Clinic (PHC) package with medical sensors (e.g., blood pressure [BP] monitors, blood glucose meters, weighing scales) was developed to provide health checkups, health risk assessments, and teleconsultations with a remote physician over Skype. Data were collected from 16,741 subjects at 10 locations (including urban and rural regions) in Bangladesh between July 2012 and February 2014 [2].

show abstract

Section: Discussionmentioning

confidence: 99%

“…Li et al [11] used a lasso logistic regression model and predicted the risk for colorectal cancer with personal characteristics and fecal immunological test. As popular and effective approaches to predictive analytics, data science and machine learning are highly regarded due to their success in diagnosis, prediction, and choice of treatment [12,13,14].…”

mentioning

confidence: 99%

Machine Learning Based Prediction of Non-communicable Diseases to Improving Intervention Program in Bangladesh

Hu¹,

Nohara²,

Wakata³

et al. 2018

ejbi

View full text Add to dashboard Cite

show abstract

“…(4) Feature fusion of symptoms is carried out by using the mapped concept set and the symptom embedding vector. (5) In prescription recommendation, the patient symptom words are used to form the patient symptom vector through SSTM, and then, the CNN framework is used for training. Finally, the predicted probability of each herb to be selected is output, so as to obtain the predicted prescription.…”

Section: Tcmpr Frameworkmentioning

confidence: 99%

“…Zhou et al [4] extracted the key compatibility of herbs and other knowledge from a large amount of TCM clinical data, indicating that herbs are not independent but closely related. Mi et al [5] used logistic regression, decision tree, and other classical machine learning algorithms and established a prediction model for prescription recommendation. Zhou et al [6] proposed an intelligent prescription recommendation system (FordNet), fusing phenotype and molecule information.…”

Section: Introductionmentioning

confidence: 99%

TCMPR: TCM Prescription recommendation based on subnetwork term mapping and deep learning

Dong

Zheng

Shu

et al. 2021

2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

View full text Add to dashboard Cite

Traditional Chinese medicine (TCM) has played an indispensable role in clinical diagnosis and treatment. Based on a patient's symptom phenotypes, computation-based prescription recommendation methods can recommend personalized TCM prescription using machine learning and artificial intelligence technologies. However, owing to the complexity and individuation of a patient's clinical phenotypes, current prescription recommendation methods cannot obtain good performance. Meanwhile, it is very difficult to conduct effective representation for unrecorded symptom terms in an existing knowledge base. In this study, we proposed a subnetwork-based symptom term mapping method (SSTM) and constructed a SSTM-based TCM prescription recommendation method (termed TCMPR). Our SSTM can extract the subnetwork structure between symptoms from a knowledge network to effectively represent the embedding features of clinical symptom terms (especially the unrecorded terms). The experimental results showed that our method performs better than state-of-the-art methods. In addition, the comprehensive experiments of TCMPR with different hyperparameters (i.e., feature embedding, feature dimension, subnetwork filter threshold, and feature fusion) demonstrate that our method has high performance on TCM prescription recommendation and potentially promote clinical diagnosis and treatment of TCM precision medicine.

show abstract

“…Zhou et al [ 4 ] extracted the key compatibility of herbs and other knowledge from a large amount of TCM clinical data, indicating that herbs are not independent but closely related. Mi et al [ 5 ] used logistic regression, decision tree, and other classical machine learning algorithms and established a prediction model for prescription recommendation. Zhou et al [ 6 ] proposed an intelligent prescription recommendation system (FordNet), fusing phenotype and molecule information.…”

Section: Introductionmentioning

confidence: 99%

TCMPR: TCM Prescription Recommendation Based on Subnetwork Term Mapping and Deep Learning

Dong

Zheng

Shu

et al. 2022

BioMed Research International

View full text Add to dashboard Cite

Traditional Chinese medicine (TCM) has played an indispensable role in clinical diagnosis and treatment. Based on a patient’s symptom phenotypes, computation-based prescription recommendation methods can recommend personalized TCM prescription using machine learning and artificial intelligence technologies. However, owing to the complexity and individuation of a patient’s clinical phenotypes, current prescription recommendation methods cannot obtain good performance. Meanwhile, it is very difficult to conduct effective representation for unrecorded symptom terms in an existing knowledge base. In this study, we proposed a subnetwork-based symptom term mapping method (SSTM) and constructed a SSTM-based TCM prescription recommendation method (termed TCMPR). Our SSTM can extract the subnetwork structure between symptoms from a knowledge network to effectively represent the embedding features of clinical symptom terms (especially the unrecorded terms). The experimental results showed that our method performs better than state-of-the-art methods. In addition, the comprehensive experiments of TCMPR with different hyperparameters (i.e., feature embedding, feature dimension, subnetwork filter threshold, and feature fusion) demonstrate that our method has high performance on TCM prescription recommendation and potentially promote clinical diagnosis and treatment of TCM precision medicine.

show abstract

Prescription Prediction towards Computer-Assisted Diagnosis for Kampo Medicine

Cited by 4 publications

References 0 publications

Machine Learning Based Prediction of Non-communicable Diseases to Improving Intervention Program in Bangladesh

Machine Learning Based Prediction of Non-communicable Diseases to Improving Intervention Program in Bangladesh

TCMPR: TCM Prescription recommendation based on subnetwork term mapping and deep learning

TCMPR: TCM Prescription Recommendation Based on Subnetwork Term Mapping and Deep Learning

Contact Info

Product

Resources

About