scholarly journals Data Mining to Determine Behavioral Patterns in Respiratory Disease in Pediatric Patients

Author(s):  
Michael Cabanillas-Carbonell ◽  
Randy Verdecia-Peña ◽  
José Luis Herrera Salazar ◽  
Esteban Medina-Rafaile ◽  
Oswaldo Casazola-Cruz
PEDIATRICS ◽  
1951 ◽  
Vol 7 (4) ◽  
pp. 537-549
Author(s):  
ELLEN P. MACKENZIE

Thirteen cases of pneumatosis intestinalis are reported, 12 of them in infants between 12 days and 12 months of age, one in a boy of 6 years. Review of these cases and of 32 reported cases falling within the pediatric age range discloses that the disease occurs most frequently in patients whose general condition is poor, that it is very often associated with congenital or acquired disease of the intestine, and that respiratory disease, usually infectious, frequently co-exists. The presence of pneumatosis in pediatric patients has so far been discovered only at autopsy, but clinical diagnosis, with the aid of the typical roentgenologic findings, is feasible and may be accomplished when the disease is more widely known. The clinical picture and roentgenographic findings in adults are reviewed. The most acceptable theories concerning the pathogenesis are discussed, with their possible relation to infantile diarrhea.


Author(s):  
Mark Last

Data mining is a growing collection of computational techniques for automatic analysis of structured, semi-structured, and unstructured data with the purpose of identifying important trends and previously unknown behavioral patterns. Data mining is widely recognized as the most important and central technology for homeland security in general and for cyber warfare in particular


2019 ◽  
Vol 12 (3) ◽  
pp. 154-168 ◽  
Author(s):  
Luis Naito Mendes Bezerra ◽  
Márcia Terra da Silva

In distance learning, the professor cannot see that the students are having trouble with a subject, and can fail to perceive the problem in time to intervene. However, in learning management systems (LMS's) a large volume of data regarding online access, participation and progress can be registered and collected allowing analysis based on students' behavioral patterns. As traditional methods have a limited capacity to extract knowledge from big volumes of data, educational data mining (EDM) arises as a tool to help teachers interpreting the behavior of students. The objective of the present article is to describe the application of educational data mining technics aiming to obtain relevant knowledge of students' behavioral patterns in an LMS for an online course, with 1,113 students enrolled. This paper applies two algorithms on educational context, decision tree and clustering, unveiling unknown relevant aspects to professors and managers, such as the most important examinations that contribute to students' approval as well as the most significant attributes to their success.


Author(s):  
Ingo Mrosewski ◽  
Tobias Dähn ◽  
Jörg Hehde ◽  
Elena Kalinowski ◽  
Ilona Lindner ◽  
...  

Abstract Objectives Establishing direct reference intervals (RIs) for pediatric patients is a very challenging endeavor. Indirectly determined RIs can address this problem by utilization of existing clinical laboratory databases. In order to provide better laboratory services to the local pediatric population, we established population-specific hematology RIs via data mining. Methods Our laboratory information system (LIS) was searched for pediatric blood counts of patients aged from 0 days to 18 years, performed from 1st of January 2018 until 31st of March 2021. In total, 27,554 blood counts on our SYSMEX XN-9000 were initially identified. After application of pre-defined exclusion criteria, 18,531 sample sets remained. Age- and sex-specific RIs were established in accordance with International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) and Clinical & Laboratory Standards Institute (CLSI) recommendations. Results When compared to pediatric RIs supplied by other authors, the RIs determined specifically for pediatric patients from Berlin and Brandenburg showed several relevant differences, especially with regard to white blood cell counts (WBCs), red blood cell counts (RBCs), red cell distribution widths (RDW) and platelet counts (PLTs) within the distinct age groups. Additionally, alterations to several published age-specific partitions had to be made, while new sex-specific partitions were introduced for WBCs and PLTs. Conclusions Generic RIs from textbooks, manufacturer information and medical publications – even from nationwide or multicenter studies – commonly used in many laboratories might not reflect the specifics of local patient populations properly. RIs should be tailored to the serviced patient population whenever possible. Careful data mining appears to be suitable for this task.


2015 ◽  
pp. 320-340
Author(s):  
Alberto Ochoa-Zezzatti ◽  
Saúl González ◽  
Fernando Montes ◽  
Seyed Amin ◽  
Lourdes Margain ◽  
...  

This chapter proposes a decision support system applied to public schooling, especially for reducing dropout rates for minorities. That is relevant enough to enable understanding of ethical data mining in a strategic planning context. This understanding explains the importance of adequate different aspects related with Strategic Planning. The authors focus their analysis on a specific problem related with reducing dropout rates based on decision support systems under uncertainty. To this end, surveys are performed to gather information about this problem using Data Mining techniques to profile a number of behavioral patterns and choices that describe social behaviors. Ethical Data Mining is used for reasons of culture to improve the socio economic development. In addition, the chapter describes innovative models that capture salient variables of modernization, and how these variables give raise to intervening aspects that end up shaping behavioral patterns in ethical and social aspects. Finally, the chapter remarks and extends discussions of the authors' approach and will provide general guidelines for future work in diverse application domains, including further analysis on how those public politics organize and operate.


2008 ◽  
Vol 135 (2) ◽  
pp. 230-236 ◽  
Author(s):  
Sunchai Payungporn ◽  
Thaweesak Chieochansin ◽  
Chittima Thongmee ◽  
Rujipat Samransamruajkit ◽  
Apiradee Theamboolers ◽  
...  

2020 ◽  
Vol 7 (Supplement_1) ◽  
pp. S602-S603
Author(s):  
Emily Parsons ◽  
Zach Colburn ◽  
Nicholas Friedman ◽  
Jason Caboot ◽  
Anjali Kunz ◽  
...  

Abstract Background Isolation of Candida from the respiratory tract of patients with cystic fibrosis (CF) is common, but its clinical significance remains unclear. We evaluated whether pediatric Candida colonization is associated with specific risk factors, co-pathogens, and degree of respiratory disease. Methods Using the Military Healthcare System database, we identified 273 pediatric patients with CF who were followed for 938 person-years between 2012 and 2017. To determine whether prevalence was associated with different categorical variables, Fisher’s exact tests were performed on 1000 random samples with the constraint that exactly one interval was selected from each individual to generate each sample. When appropriate, follow-up binomial tests were performed to identify species differences. Individuals with a specific Candida species isolated in ≥50% of their respiratory cultures were considered colonized. Those with C. albicans were analyzed separately from all other Candida species. FEV1 values < 80% predicted were used as a surrogate for degree of respiratory disease. Results Candida colonization was not associated with degree of respiratory disease, exocrine pancreatic insufficiency, co-existing diabetes, or the presence of a homozygous F508del CFTR mutation. C. albicans colonization differed by age, and was least prevalent amongst 0-2 year olds (p=0.031) (Fig 1). Compared to those either not colonized with Candida, or colonized with a species other than C. albicans, patients colonized with C. albicans had lower rates of co-infection with Aspergillus (p = 0.041) (Fig 2). Significant differences in Candida colonization between groups was also notable for those colonized with Stenotrophomonas (p=0.014) and Nontuberculous Mycobacterium (p < 0.01), but not for Staphylococcus aureus or Pseudomonas (all p > 0.1). Figure 1. C. albicans prevalence differed by age group (p<0.01). Specifically, prevalence was lower in the 0-2 year old age group (p=0.031). Figure 2. Individuals were grouped into those without a Candida infection (None), those with non-C. albicans colonization (Other), and those with C. albicans colonization. No differences were found with respect to co-infection with MRSA, MSSA, or Pseudomonas. Significant differences were found with respect to Stenotrophomonas (p=0.014), Aspergillus (p < 0.01), and NTM (p < 0.01). The prevalence of Aspergillus in those individuals with C. albicans was lower compared to those with a different Candida infection or no Candida infection (p=0.041). The prevalence of co-infection with Stenotrophomonas was somewhat elevated among those with a non-C. albicans infection (p=0.052). Conclusion C. albicans likely plays a role in influencing the airway microbiome of patients with CF. The significance of colonization with other Candida species warrants further exploration. Our data suggests that further studies are needed to evaluate whether Candida may be seen as protective against certain pathogens and therefore this may influence recommendations to treat patients who have CF with antifungals. Disclosures All Authors: No reported disclosures


Sign in / Sign up

Export Citation Format

Share Document