Using Data Mining Strategies in Clinical Decision Making: A Literature Review

Automated data extraction of electronic medical records: Validity of data mining to construct research databases for eligibility in gastroenterological clinical trials

BACKGROUND

Electronic medical records (EMRs) are adopted for storing patient-related healthcare information. Using data mining techniques, it is possible to make use of and derive benefit from this massive amount of data effectively. We aimed to evaluate validity of data extracted by the Customized eXtraction Program (CXP).

METHODS

The CXP extracts and structures data in rapid standardised processes. The CXP was programmed to extract TNFα-native active ulcerative colitis (UC) patients from EMRs using defined International Classification of Disease-10 (ICD-10) codes. Extracted data were read in parallel with manual assessment of the EMR to compare with CXP-extracted data.

RESULTS

From the complete EMR set, 2,802 patients with code K51 (UC) were extracted. Then, CXP extracted 332 patients according to inclusion and exclusion criteria. Of these, 97.5% were correctly identified, resulting in a final set of 320 cases eligible for the study. When comparing CXP-extracted data against manually assessed EMRs, the recovery rate was 95.6-101.1% over the years with 96.1% weighted average sensitivity.

CONCLUSION

Utilisation of the CXP software can be considered as an effective way to extract relevant EMR data without significant errors. Hence, by extracting from EMRs, CXP accurately identifies patients and has the capacity to facilitate research studies and clinical trials by finding patients with the requested code as well as funnel down itemised individuals according to specified inclusion and exclusion criteria. Beyond this, medical procedures and laboratory data can rapidly be retrieved from the EMRs to create tailored databases of extracted material for immediate use in clinical trials.

Nursing Diagnoses for Coronavirus Disease, COVID-19: Identification by Taxonomic Triangulation

PURPOSE

To identify the nursing care problems related to the clinical process of disease by COVID-19.

METHOD

The study applied the taxonomic triangulation technique on a clinical management guide to coronavirus disease, COVID-19, from the World Health Organization. The technique is divided into the phases: extraction of knowledge in natural language about assessment, planning and intervention, translation into standard language NOC and NIC, linking to NANDA-I diagnoses, triangulation looking for diagnostic matches in the three sets, and, finally, validation by a panel of experts from a hospital and a university.

FINDINGS

The extraction identified 159 terms in natural language that were translated into 173 variables: 34 NOC for assessment, 19 NOC for planning, and 120 NIC for intervention. The relationships to NANDA-I diagnoses recorded 2,182 links and the triangulation returned 109 diagnoses, 54 of them for a critical situation. The panel of experts unanimously validated the 29 diagnoses with the highest number of links.

CONCLUSION

Coronavirus disease, COVID-19, involves a complex situation with multiple associated care problems that can be identified using the taxonomic triangulation technique.

IMPLICATIONS FOR NURSING PRACTICE

The links between taxonomies and the taxonomic triangulation technique are an important tool for generating knowledge. The results of this study may guide the diagnosis and treatment of coronavirus disease, COVID-19, as well as similar processes that occur with acute respiratory distress syndrome.

New Blood Biomarkers for the Diagnosis of AFP-Negative Hepatocellular Carcinoma

An early diagnosis of hepatocellular carcinoma (HCC) followed by effective treatment is currently critical for improving the prognosis and reducing the associated economic burden. Alpha-fetoprotein (AFP) is the most widely used biomarker for HCC diagnosis. Based on elevated serum AFP levels as well as typical imaging features, AFP-positive HCC (APHC) can be easily diagnosed, but AFP-negative HCC (ANHC) is not easily detected due to lack of ideal biomarkers and thus mainly reliance on imaging. Imaging for the diagnosis of ANHC is probably insufficient in sensitivity and/or specificity because most ANHC tumors are small and early-stage HCC, and it is involved in sophisticated techniques and high costs. Moreover, ANHC accounts for nearly half of HCC and exhibits a better prognosis compared with APHC. Therefore, the diagnosis of ANHC in clinical practice has been a critical issue for the early treatment and prognosis improvement of HCC. In recent years, tremendous efforts have been made to discover new biomarkers complementary to AFP for HCC diagnosis. In this review, we systematically review and discuss the recent advances of blood biomarkers for HCC diagnosis, including DNA biomarkers, RNA biomarkers, protein biomarkers, and conventional laboratory metrics, focusing on their diagnostic evaluation alone and in combination, in particular on their diagnostic performance for ANHC.

Taxonomic Triangulation of Care in Healthcare Protocols: Mapping of Diagnostic Knowledge From Standardized Language

Taxonomic triangulation is a data mining technique for the management of care knowledge. This technique uses standardized languages, such as North American Nursing Diagnosis Association International, Nursing Outcomes Classification, and Nursing Interventions Classification, as well as logic. Its purpose is to find patterns in the data and identify care diagnoses. Triangulation can be applied to databases (clinical records) or to bibliographic sources (eg, protocols). The objective of this study is to identify the care diagnoses implicit in the nursing care protocols of the Community of Madrid. The method followed has three phases: knowledge extraction for mapping of variables, linking to diagnoses, and triangulation with analysis. The study analyzes six protocols, and 344 variables (167 assessment, 29 planning, and 148 intervention) and 6118 links have been extracted. Triangulation identified 165 NANDA diagnoses (68.48%), and only 25 labels were not revealed through this process. As a limitation, the results depend on the knowledge presented in protocols and change with language editions. Some labels included in the sample are recent and are not included in the links with nursing outcomes classification and nursing interventions classification. In conclusion, taxonomic triangulation makes it possible to manage knowledge, discover data patterns, and represent care situations.