New knowledge extraction technique using probability for case‐based reasoning: application to medical diagnosis

Park, Yoon-Joo; Kim, Byung-Chun; Shan, Chun

doi:10.1111/j.1468-0394.2006.00321.x

Cited by 49 publications

(41 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, one study [19] used data from a research center and another study [37] utilized telemonitoring data, also known as wearable-based remote patient monitoring data. From the application area perspective, chronic diseases were the most prevalent context.…”

Section: Resultsmentioning

confidence: 99%

Patient Similarity in Prediction Models Based on Health Data: A Scoping Review

Sharafoddini¹,

Dubin²,

Lee³

2017

JMIR Med Inform

View full text Add to dashboard Cite

BackgroundPhysicians and health policy makers are required to make predictions during their decision making in various medical problems. Many advances have been made in predictive modeling toward outcome prediction, but these innovations target an average patient and are insufficiently adjustable for individual patients. One developing idea in this field is individualized predictive analytics based on patient similarity. The goal of this approach is to identify patients who are similar to an index patient and derive insights from the records of similar patients to provide personalized predictions..ObjectiveThe aim is to summarize and review published studies describing computer-based approaches for predicting patients’ future health status based on health data and patient similarity, identify gaps, and provide a starting point for related future research.MethodsThe method involved (1) conducting the review by performing automated searches in Scopus, PubMed, and ISI Web of Science, selecting relevant studies by first screening titles and abstracts then analyzing full-texts, and (2) documenting by extracting publication details and information on context, predictors, missing data, modeling algorithm, outcome, and evaluation methods into a matrix table, synthesizing data, and reporting results.ResultsAfter duplicate removal, 1339 articles were screened in abstracts and titles and 67 were selected for full-text review. In total, 22 articles met the inclusion criteria. Within included articles, hospitals were the main source of data (n=10). Cardiovascular disease (n=7) and diabetes (n=4) were the dominant patient diseases. Most studies (n=18) used neighborhood-based approaches in devising prediction models. Two studies showed that patient similarity-based modeling outperformed population-based predictive methods.ConclusionsInterest in patient similarity-based predictive modeling for diagnosis and prognosis has been growing. In addition to raw/coded health data, wavelet transform and term frequency-inverse document frequency methods were employed to extract predictors. Selecting predictors with potential to highlight special cases and defining new patient similarity metrics were among the gaps identified in the existing literature that provide starting points for future work. Patient status prediction models based on patient similarity and health data offer exciting potential for personalizing and ultimately improving health care, leading to better patient outcomes.

show abstract

Section: Resultsmentioning

confidence: 99%

Patient Similarity in Prediction Models Based on Health Data: A Scoping Review

Sharafoddini¹,

Dubin²,

Lee³

2017

JMIR Med Inform

View full text Add to dashboard Cite

show abstract

“…Statistics were used as part of CBR in [48]. For example, Park et al [46] suggest a case extraction method called Statistical CBR (SCBR) that retrieves the optimal number of neighbours based on their probabilistic similarity. Tsatsoulis et al [60] use Decision Theory but with a different aim from ours.…”

Section: Research Trends In Cbr Retrievalmentioning

confidence: 99%

Loss and gain functions for CBR retrieval

Castro

Navarro

Sánchez

et al. 2009

Information Sciences

View full text Add to dashboard Cite

“…This probabilistic approach has been widely applied in the medical domain in areas such as diagnosis classification, drug testing and advice about medicine [19][20][21]. As far as medical prescription is concerned, Warren et al [22] developed an anticipative data entry interface (Mediface) to intelligently generate 'hot lists' (by learning through probabilistic models) for general practitioners to reduce the time required for selecting the relevant medicines for the patient.…”

Section: Statistical Perspectives Modeling With the Bayesian Theoremmentioning

confidence: 99%

CASESIAN: A knowledge-based system using statistical and experiential perspectives for improving the knowledge sharing in the medical prescription process

Ting

Kwok

Tsang

et al. 2010

Expert Systems with Applications

View full text Add to dashboard Cite

SummaryObjectives: Knowledge sharing is crucial for better patient care in the healthcare industry, but it is challenging for physicians to exchange their clinical insights and practice experiences, particularly with regard to the issuing of prescriptions for medicine. The aim of our study is to facilitate knowledge sharing and information exchange in this area by means of a knowledge-based system.Methods: We propose a knowledge-based system, CASESIAN, to automatically model each physician's prescription experience. This is done by collecting as many as possible instances of when the physician has issued a prescription. These occasions will be analyzed from a statistical perspective to form a reciprocal interactive knowledge sharing process for the issuing of medical prescriptions which we will call the prescription process. With the help of the prescription data in medical organizations, the knowledge-based system employs the Bayesian Theorem to correlate the experience of peers in order to evaluate individual prescription knowledge as retrieved through the Case-based Reasoning technique. In addition, a system prototype was implemented in a Hong Kong medical organization to evaluate the feasibility of such an approach.Results: Our evaluation indicates that there is a significant improvement in knowledge sharing after the adoption of the system. CASESIAN obtains a higher rating in both recall and precision measurement when compared to traditional knowledge-based system. In particular, its information retrieval is much stronger than the baseline in around 40%. Furthermore, regarding the result of the interviews, physicians agree that the system can improve the storing and sharing of medical prescription knowledge.Conclusion: Compared with conventional knowledge-based systems, CASESIAN provides more peer-based evidence that can enhance the learning and sharing process, transforming it from a single loop to a double loop. The quality of shared knowledge is, in addition, more objective and less biased.

show abstract

New knowledge extraction technique using probability for case‐based reasoning: application to medical diagnosis

Cited by 49 publications

References 29 publications

Patient Similarity in Prediction Models Based on Health Data: A Scoping Review

Patient Similarity in Prediction Models Based on Health Data: A Scoping Review

Loss and gain functions for CBR retrieval

CASESIAN: A knowledge-based system using statistical and experiential perspectives for improving the knowledge sharing in the medical prescription process

Contact Info

Product

Resources

About