Design of a Recommendation System for Adding Support in the Treatment of Chronic Patients

Rapid growth of chronic disease cases around the world is adding pressure on healthcare providers to ensure a structured patent follow-up during chronic disease management process. In response to the increasing demand for better chronic disease management and improved health care efficiency, nursing roles have been specialized or enhanced in the primary health care setting. Nurses become key players in chronic disease management process. Study describes a system to help nurses manage the care process of patient with chronic disease. It supports focusing nurse's attention on those resources/solutions that are likely to be most relevant to their particular situation/problem in nursing domain. System is based on multi-relational property graph representing a flexible modeling construct. Graph allows modeling a nursing ontology and the indices that partition domain into an efficient, searchable space where the solution to a problem is seen as abstractly defined traversals through its vertices and edges.

A systematic comparison of feature space effects on disease classifier performance for phenotype identification of five diseases

Automated phenotype identification plays a critical role in cohort selection and bioinformatics data mining. Natural Language Processing (NLP)-informed classification techniques can robustly identify phenotypes in unstructured medical notes. In this paper, we systematically assess the effect of naive, lexically normalized, and semantic feature spaces on classifier performance for obesity, atherosclerotic cardiovascular disease (CAD), hyperlipidemia, hypertension, and diabetes. We train support vector machines (SVMs) using individual feature spaces as well as combinations of these feature spaces on two small training corpora (730 and 790 documents) and a combined (1520 documents) training corpus. We assess the importance of feature spaces and training data size on SVM model performance. We show that inclusion of semantically-informed features does not statistically improve performance for these models. The addition of training data has weak effects of mixed statistical significance across disease classes suggesting larger corpora are not necessary to achieve relatively high performance with these models.

Generalized discriminant analysis for congestive heart failure risk assessment based on long-term heart rate variability

The aims of this study are summarized in the following items: first, to investigate the class discrimination power of long-term heart rate variability (HRV) features for risk assessment in patients suffering from congestive heart failure (CHF); second, to introduce the most discriminative features of HRV to discriminate low risk patients (LRPs) and high risk patients (HRPs), and third, to examine the influence of feature dimension reduction in order to achieve desired accuracy of the classification. We analyzed two public Holter databases: 12 data of patients suffering from mild CHF (NYHA class I and II), labeled as LRPs and 32 data of patients suffering from severe CHF (NYHA class III and IV), labeled as HRPs. A K-nearest neighbor classifier was used to evaluate the performance of feature set in the classification. Moreover, to reduce the number of features as well as the overlap of the samples of two classes in feature space, we used generalized discriminant analysis (GDA) as a feature extraction method. By applying GDA to the discriminative nonlinear features, we achieved sensitivity and specificity of 100% having the least number of features. Finally, the results were compared with other similar conducted studies regarding the performance of feature selection procedure and classifier besides the number of features used in training.

Using association rule mining to identify risk factors for early childhood caries

BACKGROUND AND OBJECTIVE

Early childhood caries (ECC) is a potentially severe disease affecting children all over the world. The available findings are mostly based on a logistic regression model, but data mining, in particular association rule mining, could be used to extract more information from the same data set.

METHODS

ECC data was collected in a cross-sectional analytical study of the 10% sample of preschool children in the South Bačka area (Vojvodina, Serbia). Association rules were extracted from the data by association rule mining. Risk factors were extracted from the highly ranked association rules.

RESULTS

Discovered dominant risk factors include male gender, frequent breastfeeding (with other risk factors), high birth order, language, and low body weight at birth. Low health awareness of parents was significantly associated to ECC only in male children.

CONCLUSIONS

The discovered risk factors are mostly confirmed by the literature, which corroborates the value of the methods.

Technology-Enhanced Stepped Collaborative Care Targeting Posttraumatic Stress Disorder and Comorbidity After Injury: A Randomized Controlled Trial

Posttraumatic stress disorder (PTSD) and its comorbidities are endemic among injured trauma survivors. Previous collaborative care trials targeting PTSD after injury have been effective, but they have required intensive clinical resources. The present pragmatic clinical trial randomized acutely injured trauma survivors who screened positive on an automated electronic medical record PTSD assessment to collaborative care intervention (n = 60) and usual care control (n = 61) conditions. The stepped measurement-based intervention included care management, psychopharmacology, and psychotherapy elements. Embedded within the intervention were a series of information technology (IT) components. PTSD symptoms were assessed with the PTSD Checklist at baseline prerandomization and again, 1-, 3-, and 6-months postinjury. IT utilization was also assessed. The technology-assisted intervention required a median of 2.25 hours (interquartile range = 1.57 hours) per patient. The intervention was associated with modest symptom reductions, but beyond the margin of statistical significance in the unadjusted model: F(2, 204) = 2.95, p = .055. The covariate adjusted regression was significant: F(2, 204) = 3.06, p = .049. The PTSD intervention effect was greatest at the 3-month (Cohen's effect size d = 0.35, F(1, 204) = 4.11, p = .044) and 6-month (d = 0.38, F(1, 204) = 4.10, p = .044) time points. IT-enhanced collaborative care was associated with modest PTSD symptom reductions and reduced delivery times; the intervention model could potentially facilitate efficient PTSD treatment after injury.

Promising approaches of computer-supported dietary assessment and management-Current research status and available applications

PURPOSE

The aim of this review was to analyze computer-based tools for dietary management (including web-based and mobile devices) from both scientific and applied perspectives, presenting advantages and disadvantages as well as the state of validation.

METHODS

For this cross-sectional analysis, scientific results from 41 articles retrieved via a medline search as well as 29 applications from online markets were identified and analyzed.

RESULTS

Results show that many approaches computerize well-established existing nutritional concepts for dietary assessment, e.g., food frequency questionnaires (FFQ) or dietary recalls (DR). Both food records and barcode scanning are less prominent in research but are frequently offered by commercial applications. Integration with a personal health record (PHR) or a health care workflow is suggested in the literature but is rarely found in mobile applications.

CONCLUSIONS

It is expected that employing food records for dietary assessment in research settings will be increasingly used when simpler interfaces, e.g., barcode scanning techniques, and comprehensive food databases are applied, which can also support user adherence to dietary interventions and follow-up phases of nutritional studies.

Impact of an inpatient geriatric consultative service on outcomes for cognitively impaired patients

BACKGROUND

Impact of geriatric consultative services (GCS) on hospital readmission and mortality outcomes for cognitively impaired (CI) patients is not known.

OBJECTIVE

Evaluate impact of GCS on hospital readmission and mortality among CI inpatients.

DESIGN

Secondary data analysis of a prospective trial of a computerized decision support system between July 1, 2006 and May 30, 2008.

SETTING

Study conducted at Eskenazi hospital, Indianapolis, Indiana, a 340-bed, public hospital with over 2300 yearly admissions of patients ages 65 years or older.

PATIENTS

There were 415 inpatients aged 65 years and older with CI enrolled from July 2006 to March 2008.

MEASUREMENTS

Thirty-day and 1-year mortality and hospital readmission following the index admission. Cox proportional hazard models were used to determine the association between receiving GCS, readmission, or mortality while adjusting for demographics, discharge destination, delirium, Charlson Comorbidity Index, and prior hospitalizations. The propensity score method was used to adjust for the nonrandom assignment of GCS.

RESULTS

Patients receiving GCS were older (79 years old, 8.1 standard deviation [SD] vs 76 years old, 7.8 SD; P < 0.001) with higher incidence of delirium (49% vs 29%; P < 0.001). No significant differences were found between the groups for hospital readmission (hazard ratio [HR] = 1.19; 95% confidence interval = 0.89-1.59) and mortality at 12 months of index admission (HR = 0.91; 95% confidence interval = 0.59-1.40). However, a significant increase in readmissions was observed for the GCS group (HR = 1.75; 95% confidence interval = 1.06-2.88) at 30 days postdischarge.

CONCLUSION

One-year postdischarge outcomes of CI patients who received GCS were not different from patients who did not receive the service. New models of care are needed to improve postdischarge readmission and mortality among hospitalized patients with CI.

DyKOSMap: A framework for mapping adaptation between biomedical knowledge organization systems

BACKGROUND

Knowledge Organization Systems (KOS) and their associated mappings play a central role in several decision support systems. However, by virtue of knowledge evolution, KOS entities are modified over time, impacting mappings and potentially turning them invalid. This requires semi-automatic methods to maintain such semantic correspondences up-to-date at KOS evolution time.

METHODS

We define a complete and original framework based on formal heuristics that drives the adaptation of KOS mappings. Our approach takes into account the definition of established mappings, the evolution of KOS and the possible changes that can be applied to mappings. This study experimentally evaluates the proposed heuristics and the entire framework on realistic case studies borrowed from the biomedical domain, using official mappings between several biomedical KOSs.

RESULTS

We demonstrate the overall performance of the approach over biomedical datasets of different characteristics and sizes. Our findings reveal the effectiveness in terms of precision, recall and F-measure of the suggested heuristics and methods defining the framework to adapt mappings affected by KOS evolution. The obtained results contribute and improve the quality of mappings over time.

CONCLUSIONS

The proposed framework can adapt mappings largely automatically, facilitating thus the maintenance task. The implemented algorithms and tools support and minimize the work of users in charge of KOS mapping maintenance.

EMPOWER--pathways for supporting the self-management of diabetes patients

Diabetes is a serious world-wide medical challenge and there is a recognised need for improved diabetes care outcomes. This paper describes results of the EMPOWER project, to foster the self-management of diabetes patients by integration of existing and new services offered to patients after having been diagnosed with diabetes. The Self-Management Pathway described in this paper helps patients in the specification of personalized activities based on medical recommendations and personal goals, as well as self-monitoring of the results. The whole process is supported by innovative ICT services that motivate patients to change their lifestyle and adhere to defined medication and activity plans. We describe the approach and present the findings of the validation phase in Germany and Turkey.

Ontology-based Medical Decision Support System to Enhance Chronic Patients' Lifestyle within E-care Telemonitoring Platform

The aim of this paper is to describe an original approach which consists of designing ontology based Medical Decision Support System (MDSS) to enhance the patients' lifestyle. This system is composed of two main parts: data collector which collects relevant lifestyle-related patent data by prompting the only significant questions in connection with the patient's medical background, and advices provider which provides personalized lifestyle advices to the patients regarding their lifestyle changes. The proposed MDSS is integrated within E-care home health monitoring platform in order to: (i) improve the patient's healthy lifestyle; (ii) educate the patients towards their disease; (iii) increase the early detection of risky situation.

Archetype based patient data modeling to support treatment of pituitary adenomas

The treatment of patients with pituitary adenoma requires the assessment of various patient data by the clinician. Because of their heterogeneity, they are stored in different sub-information systems, limiting a fast and easy access. The objective of this paper is to apply and test the tools provided by the openEHR Foundation to model the patient data relevant for diagnosis and treatment of the disease with the future intention to implement a centralised standard-based information platform. This platform should support the clinician in the treatment of the disease and improve the information exchange with other healthcare institutions. Some results of the domain modeling, so far obtained, are presented, and the advantages of openEHR emphasized. The free tools and the large database of existing structured and standard archetypes facilitated the modeling task. The separation of the domain modeling from the application development will support the next step of development of the information platform.

Evaluating Methods for Identifying Cancer in Free-Text Pathology Reports Using Various Machine Learning and Data Preprocessing Approaches

Automated detection methods can address delays and incompleteness in cancer case reporting. Existing automated efforts are largely dependent on complex dictionaries and coded data. Using a gold standard of manually reviewed pathology reports, we evaluated the performance of alternative input formats and decision models on a convenience sample of free-text pathology reports. Results showed that the input format significantly impacted performance, and specific algorithms yielded better results for presicion, recall and accuracy. We conclude that our approach is sufficiently accurate for practical purposes and represents a generalized process.

Trigger Development for the Improvement of Neurological Patient Care

By analyzing medical records, we developed triggers for epilepsy patients' care coordination. Thirteen triggers with potential to affect patient care outcomes and safety were found.

Oncotherapy: A System for Requesting Chemotherapy Protocols

A clinical decision support system is able to provide oncologists with suitable treatment options at the moment of decision making regarding which chemotherapy protocol is the best to apply to a particular oncological case. The National Cancer Institute has created a Guidelines Committee that establishes therapeutical options for each clinical case. The Health Informatics Department has developed Oncotherapy, a knowledge database that incorporates information provided by the Guidelines Committee. Oncotherapy includes a tailored information repository to provide oncologists in the public health system with the chemotherapy protocols available given three types of data: clinical diagnosis, clinical stage and therapy criteria. The protocol selected by the treating oncologist is sent back to Oncotherapy, which may create new knowledge that can be incorporated into the knowledge database. In this way, the system supports making the best decision according to the chemotherapy protocol options available. Furthermore, it can warn of errors that could result from mistakenly chosen therapies.

Conceptual Knowledge Discovery in Databases for Drug Combinations Predictions in Malignant Melanoma

The worldwide incidence of melanoma is rising faster than any other cancer, and prognosis for patients with metastatic disease is poor. Current targeted therapies are limited in their durability and/or effect size in certain patient populations due to acquired mechanisms of resistance. Thus, the development of synergistic combinatorial treatment regimens holds great promise to improve patient outcomes. We have previously shown that a model for in-silico knowledge discovery, Translational Ontology-anchored Knowledge Discovery Engine (TOKEn), is able to generate valid relationships between bimolecular and clinical phenotypes. In this study, we have aggregated observational and canonical knowledge consisting of melanoma-related biomolecular entities and targeted therapeutics in a computationally tractable model. We demonstrate here that the explicit linkage of therapeutic modalities with biomolecular underpinnings of melanoma utilizing the TOKEn pipeline yield a set of informed relationships that have the potential to generate combination therapy strategies.

Classification of Contextual Use of Left Ventricular Ejection Fraction Assessments

Knowledge of the left ventricular ejection fraction is critical for the optimal care of patients with heart failure. When a document contains multiple ejection fraction assessments, accurate classification of their contextual use is necessary to filter out historical findings or recommendations and prioritize the assessments for selection of document level ejection fraction information. We present a natural language processing system that classifies the contextual use of both quantitative and qualitative left ventricular ejection fraction assessments in clinical narrative documents. We created support vector machine classifiers with a variety of features extracted from the target assessment, associated concepts, and document section information. The experimental results showed that our classifiers achieved good performance, reaching 95.6% F1-measure for quantitative assessments and 94.2% F1-measure for qualitative assessments in a five-fold cross-validation evaluation.

ClinicalAccess: a clinical decision support tool

ClinicalAccess is a new clinical decision support tool that uses a question-and-answer format to mirror clinical decision-making strategies. The unique format of ClinicalAccess delivers concise, authoritative answers to more than 120,000 clinical questions. This column presents a review of the product, a sample search, and a comparison with other point-of-care search engines.

Synthesizing analytic evidence to refine care pathways

Care pathways play significant roles in delivering evidence-based and coordinated care to patients with specific conditions. In order to put care pathways into practice, clinical institutions always need to adapt them based on local care settings so that the best local practices can be incorporated and used to develop refined pathways. However, it is knowledge-intensive and error-prone to incorporate various analytic insights from local data sets. In order to assist care pathway developers in working effectively and efficiently, we propose to automatically synthesize the analytical evidences derived from multiple analysis methods, and recommend modelling operations accordingly to derive a refined care pathway for a specific patient cohort. We validated our method by adapting a Congestive Heart Failure (CHF) Ambulatory Care Pathway for patients with additional condition of COPD through synthesizing the results of variation analysis and frequent pattern mining against patient records.

Improvement of Diagnosis Coding by Analysing EHR and Using Rule Engine: Application to the Chronic Kidney Disease

Coding medical diagnosis in case mix databases is a time-consuming task as every information available in patient records has to be taken into account. We developed rules based on EHR data with the Drools rules engine in order to support diagnosis coding of chronic kidney disease (CKD) in our hospital. 520 patients had a GFR &lt; 60 ml/min as estimated by the Cockroft-Gault formula and corresponded to 429 case mix database entries. We compared stays in which the patient was older than 12 and younger than 65 or 80 at the time of the stay. We concluded that our rules engine implementation may improve coding of CKD for 45.6% of patients with a GFR &lt; 60 ml/min and younger than 65. When patients are older than 65 our rule engine may be less useful for suggesting missing codes of CKD because the estimation of GFR by the Cockroft-Gault formula becomes less reliable as patients get older.

Problem Solving for Volatilizing Situation in Nursing: Developing Thinking Process Supporting System using NursingNAVI® Contents

We have identified three foci of the nursing observation and nursing action respectively. Using these frameworks, we have developed the structured knowledge model for a number of diseases and medical interventions. We developed this structure based NursingNAVI® contents collaborated with some quality centered hospitals. Authors analysed the nursing care documentations of post-gastrectomy patients in light of the standardized nursing care plan in the "NursingNAVI®" developed by ourselves and revealed the "failure to observe" and "failure to document", which leaded to the volatility of the patients' data, conditions and some situation. This phenomenon should have been avoided if nurses had employed a standardized nursing care plan. So, we developed thinking process support system for planning, delivering, recording and evaluating in daily nursing using NursingNAVI® contents. A hospital decided to use NursingNAVI® contents in HIS. It was suggested that the system has availability for nursing OJT and time reduction of planning and recording without volatilizing situation.

Advances In Infection Surveillance and Clinical Decision Support With Fuzzy Sets and Fuzzy Logic

By the use of extended intelligent information technology tools for fully automated healthcare-associated infection (HAI) surveillance, clinicians can be informed and alerted about the emergence of infection-related conditions in their patients. Moni--a system for monitoring nosocomial infections in intensive care units for adult and neonatal patients--employs knowledge bases that were written with extensive use of fuzzy sets and fuzzy logic, allowing the inherent un-sharpness of clinical terms and the inherent uncertainty of clinical conclusions to be a part of Moni's output. Thus, linguistic as well as propositional uncertainty became a part of Moni, which can now report retrospectively on HAIs according to traditional crisp HAI surveillance definitions, as well as support clinical bedside work by more complex crisp and fuzzy alerts and reminders. This improved approach can bridge the gap between classical retrospective surveillance of HAIs and ongoing prospective clinical-decision-oriented HAI support.

Identification of Incidental Pulmonary Nodules in Free-text Radiology Reports: An Initial Investigation

Advances in image quality produced by computed tomography (CT) and the growth in the number of image studies currently performed has made the management of incidental pulmonary nodules (IPNs) a challenging task. This research aims to identify IPNs in radiology reports of chest and abdominal CT by Natural Language Processing techiniques to recognize IPN in sentences of radiology reports. Our preliminary analysis indicates vastly different pulmonary incidental findings rates for two different patient groups.

Towards a Clinical Decision Support System for Drug Allergy Management: Are Existing Drug Reference Terminologies Sufficient for Identifying Substitutes and Cross-Reactants?

Drug allergy cross-reactivity checking is an important component of electronic health record systems. Currently, a single, open-source medication dictionary that can provide this function does not exist. In this study, we assessed the feasibility of using RxNorm and NDF-RT (National Drug File--Reference Terminology) for allergy management decision support. We evaluated the performance of using the Pharmacological Class, Mechanism of Action and Chemical Structure NDF-RT classifications in discriminating between safe and cross-reactive alternatives to a sample of common drug allergens. The positive predictive values for the three approaches were 96.3%, 99.3% and 96.2% respectively. The negative predictive values were 94.7%, 56.8% and 92.6%. Our findings suggest that in the absence of an established medication allergy classification system, using the Pharmacologic Class and Chemical Structure classifications in NDF-RT may still be effective for discriminating between safe and cross-reactive alternatives to potential allergens.