MyMedR

Displaying all 3 publications

Abstract:

Sort:

Fulltext Adjusting outliers in univariate circular data

Mahmood, Ehab A., Rana, Sohel, Midi, Habshah, Hussin, Abdul Ghapor

Pertanika Journal of Science & Technology, 2017;25(4):1147-1158.
MyJurnal

Circular data analysis is a particular branch of statistics that sits somewhere between the analysis of linear
data and the analysis of spherical data. Circular data are used in many scientific fields. The efficiency
of the statistical methods that are applied depends on the accuracy of the data in the study. However,
circular data may have outliers that cannot be deleted. If this is the case, we have two ways to avoid the
effect of outliers. First, we can apply robust methods for statistical estimations. Second, we can adjust
the outliers using the other clean data points in the dataset. In this paper, we focus on adjusting outliers in
circular data using the circular distance between the circular data points and the circular mean direction.
The proposed procedure is tested by applying it to a simulation study and to real data sets. The results
show that the proposed procedure can adjust outliers according to the measures used in the paper.
Fulltext Procedure for Detecting Outliers in a Circular Regression Model

Rambli A, Abuzaid AH, Mohamed IB, Hussin AG

PLoS One, 2016;11(4):e0153074.
PMID: 27064566 DOI: 10.1371/journal.pone.0153074

A number of circular regression models have been proposed in the literature. In recent years, there is a strong interest shown on the subject of outlier detection in circular regression. An outlier detection procedure can be developed by defining a new statistic in terms of the circular residuals. In this paper, we propose a new measure which transforms the circular residuals into linear measures using a trigonometric function. We then employ the row deletion approach to identify observations that affect the measure the most, a candidate of outlier. The corresponding cut-off points and the performance of the detection procedure when applied on Down and Mardia's model are studied via simulations. For illustration, we apply the procedure on circadian data.
Cranial Morphology Associated With Syndromic Craniosynostosis: A Potential Detection of Abnormality in Patient's Cranial Growth Using Angular Statistics

Zulkipli NS, Satari SZ, Hariri F, Abdullah NA, Wan Yusoff WNS, Hussin AG

Cleft Palate Craniofac J, 2023 Nov;60(11):1484-1493.
PMID: 35711157 DOI: 10.1177/10556656221107524

INTRODUCTION: Apert, Crouzon, and Pfeiffer syndromes are common genetic syndromes related to syndromic craniosynostosis (SC), whereby it is a congenital defect that occurs when the cranial growth is distorted. Identifying cranial angles associated with these 3 syndromes may assist the surgical team to focus on a specific cranial part during the intervention planning, thus optimizing surgical outcomes and reducing potential morbidity.
OBJECTIVE: The aim of this study is to identify the cranial angles, which are associated with Apert, Crouzon, and Pfeiffer syndromes.
METHODS: The cranial computed tomography scan images of 17 patients with SC and 22 control groups aged 0 to 12 years who were treated in the University Malaya Medical Centre were obtained, while 12 angular measurements were attained using the Mimics software. The angular data were then divided into 2 groups (patients aged 0 to 24 months and >24 months). This work proposes a 95% confidence interval (CI) for angular mean to detect the abnormality in patient's cranial growth for the SC syndromes.
RESULTS: The 95% CI of angular mean for the control group was calculated and used as an indicator to confirm the abnormality in patient's cranial growth that is associated with the 3 syndromes. The results showed that there are different cranial angles associated with these 3 syndromes.
CONCLUSIONS: All cranial angles of the patients with these syndromes lie outside the 95% CI of angular mean of control group, indicating the reliability of the proposed CI in the identification of abnormality in the patient's cranial growth.

Filters

Please provide feedback to Administrator (afdal@afpm.org.my)

External Links