Dataset Information

The power of data mining in diagnosis of childhood pneumonia.

ABSTRACT: Childhood pneumonia is the leading cause of death of children under the age of 5 years globally. Diagnostic information on the presence of infection, severity and aetiology (bacterial versus viral) is crucial for appropriate treatment. However, the derivation of such information requires advanced equipment (such as X-rays) and clinical expertise to correctly assess observational clinical signs (such as chest indrawing); both of these are often unavailable in resource-constrained settings. In this study, these challenges were addressed through the development of a suite of data mining tools, facilitating automated diagnosis through quantifiable features. Findings were validated on a large dataset comprising 780 children diagnosed with pneumonia and 801 age-matched healthy controls. Pneumonia was identified via four quantifiable vital signs (98.2% sensitivity and 97.6% specificity). Moreover, it was shown that severity can be determined through a combination of three vital signs and two lung sounds (72.4% sensitivity and 82.2% specificity); addition of a conventional biomarker (C-reactive protein) further improved severity predictions (89.1% sensitivity and 81.3% specificity). Finally, we demonstrated that aetiology can be determined using three vital signs and a newly proposed biomarker (lipocalin-2) (81.8% sensitivity and 90.6% specificity). These results suggest that a suite of carefully designed machine learning tools can be used to support multi-faceted diagnosis of childhood pneumonia in resource-constrained settings, compensating for the shortage of expensive equipment and highly trained clinicians.

SUBMITTER: Naydenova E

PROVIDER: S-EPMC4971218 | biostudies-literature | 2016 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

The power of data mining in diagnosis of childhood pneumonia.

Naydenova Elina E Naydenova Elina E Tsanas Athanasios A Howie Stephen S Casals-Pascual Climent C De Vos Maarten M

Journal of the Royal Society, Interface 20160701 120

Childhood pneumonia is the leading cause of death of children under the age of 5 years globally. Diagnostic information on the presence of infection, severity and aetiology (bacterial versus viral) is crucial for appropriate treatment. However, the derivation of such information requires advanced equipment (such as X-rays) and clinical expertise to correctly assess observational clinical signs (such as chest indrawing); both of these are often unavailable in resource-constrained settings. In thi ...[more]

PMID: 27466436

Similar Datasets

Project description:BackgroundHyperbilirubinemia is emerging as an increasingly common problem in newborns due to a decreasing hospital length of stay after birth. Jaundice is the most common disease of the newborn and although being benign in most cases it can lead to severe neurological consequences if poorly evaluated. In different areas of medicine, data mining has contributed to improve the results obtained with other methodologies.Hence, the aim of this study was to improve the diagnosis of neonatal jaundice with the application of data mining techniques.MethodsThis study followed the different phases of the Cross Industry Standard Process for Data Mining model as its methodology.This observational study was performed at the Obstetrics Department of a central hospital (Centro Hospitalar Tâmega e Sousa--EPE), from February to March of 2011. A total of 227 healthy newborn infants with 35 or more weeks of gestation were enrolled in the study. Over 70 variables were collected and analyzed. Also, transcutaneous bilirubin levels were measured from birth to hospital discharge with maximum time intervals of 8 hours between measurements, using a noninvasive bilirubinometer.Different attribute subsets were used to train and test classification models using algorithms included in Weka data mining software, such as decision trees (J48) and neural networks (multilayer perceptron). The accuracy results were compared with the traditional methods for prediction of hyperbilirubinemia.ResultsThe application of different classification algorithms to the collected data allowed predicting subsequent hyperbilirubinemia with high accuracy. In particular, at 24 hours of life of newborns, the accuracy for the prediction of hyperbilirubinemia was 89%. The best results were obtained using the following algorithms: naive Bayes, multilayer perceptron and simple logistic.ConclusionsThe findings of our study sustain that, new approaches, such as data mining, may support medical decision, contributing to improve diagnosis in neonatal jaundice.

Project description:BACKGROUND:Pattern mining utilizes multiple algorithms to explore objective and sometimes unexpected patterns in real-world data. This technique could be applied to electronic medical record data mining; however, it first requires a careful clinical assessment and validation. OBJECTIVE:The aim of this study was to examine the use of pattern mining techniques on a large clinical dataset to detect treatment and medication use patterns for childhood pneumonia. METHODS:We applied 3 pattern mining algorithms to 680,138 medication administration records from 30,512 childhood inpatients with diagnosis of pneumonia during a 6-year period at a children's hospital in China. Patients' ages ranged from 0 to 17 years, where 37.53% (11,453/30,512) were 0 to 3 months old, 86.55% (26,408/30,512) were under 5 years, 60.37% (18,419/30,512) were male, and 60.10% (18,338/30,512) had a hospital stay of 9 to 15 days. We used the FP-Growth, PrefixSpan, and USpan pattern mining algorithms. The first 2 are more traditional methods of pattern mining and mine a complete set of frequent medication use patterns. PrefixSpan also incorporates an administration sequence. The newer USpan method considers medication utility, defined by the dose, frequency, and timing of use of the 652 individual medications in the dataset. Together, these 3 methods identified the top 10 patterns from 6 age groups, forming a total of 180 distinct medication combinations. These medications encompassed the top 40 (73.66%, 500,982/680,138) most frequently used medications. These patterns were then evaluated by subject matter experts to summarize 5 medication use and 2 treatment patterns. RESULTS:We identified 5 medication use patterns: (1) antiasthmatics and expectorants and corticosteroids, (2) antibiotics and (antiasthmatics or expectorants or corticosteroids), (3) third-generation cephalosporin antibiotics with (or followed by) traditional antibiotics, (4) antibiotics and (medications for enteritis or skin diseases), and (5) (antiasthmatics or expectorants or corticosteroids) and (medications for enteritis or skin diseases). We also identified 2 frequent treatment patterns: (1) 42.89% (291,701/680,138) of specific medication administration records were of intravenous therapy with antibiotics, diluents, and nutritional supplements and (2) 11.53% (78,390/680,138) were of various combinations of inhalation of antiasthmatics, expectorants, or corticosteroids. Fleiss kappa for the subject experts' evaluation was 0.693, indicating moderate agreement. CONCLUSIONS:Utilizing a pattern mining approach, we summarized 5 medication use patterns and 2 treatment patterns. These warrant further investigation.

Project description:ObjectiveThis study compares two methods for clinical diagnosis of childhood pneumonia that aim to estimate rates of underdiagnosis and overdiagnosis of childhood pneumonia by examining the sensitivity of Integrated Management of Childhood Diseases implementation in routine care against lung ultrasound (LUS) diagnosis.SettingWe conducted observations in 83 public health facilities (dispensaries, health centres and district hospitals) in Pwani, Dodoma and Tabora, Tanzania between October and December 2017.MethodsWe used a novel method to estimate rates of underdiagnosis and overdiagnosis of childhood pneumonia by comparing directly observed public provider diagnoses to the results of diagnoses made by trained clinicians using Mindray DP-10 ultrasound machines. We perform multivariate analysis to identify confounding effects and robustness checks to bound the result. We also explore a number of observable characteristics correlated with higher rates of agreement between provider diagnoses and ultrasound diagnoses.ResultsWe observed 93 providers conducting exams on patients aged 2 months-5 years who presented respiratory symptoms or were given a respiratory diagnosis by the provider. Of these 957 patients, 110 were excluded from analysis resulting in a final sample of 847.17.6% of cases identified as pneumonia via LUS examinations in our sample were diagnosed as pneumonia by providers, suggesting that a significant number of pneumonia cases for which care is sought in the public sector go undiagnosed. Provider knowledge of breath counting and years of experience are positively correlated with higher agreement. While clinical examination rates are not statistically correlated with agreement, it is notable that providers conducted a clinical examination on only about one-third of patients in the sample.ConclusionOur results suggest that provider training and knowledge of clinical examination protocols for pneumonia diagnosis are predictive of correct diagnosis of pneumonia and should be further explored in future research as a tool for improving quality of care.

Dataset Information

The power of data mining in diagnosis of childhood pneumonia.

Publications

The power of data mining in diagnosis of childhood pneumonia.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets