Classification of breast cancer recurrence based on imputed data: a simulation study
Abstract Several studies have been conducted to classify various real life events but few are in medical fields; particularly about breast recurrence under statistical techniques. To our knowledge, there is no reported comparison of statistical classification accuracy and classifiers’ discriminative...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2022-12-01
|
| Series: | BioData Mining |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13040-022-00316-8 |
| _version_ | 1828091231509413888 |
|---|---|
| author | Rahibu A. Abassi Amina S. Msengwa |
| author_facet | Rahibu A. Abassi Amina S. Msengwa |
| author_sort | Rahibu A. Abassi |
| collection | DOAJ |
| description | Abstract Several studies have been conducted to classify various real life events but few are in medical fields; particularly about breast recurrence under statistical techniques. To our knowledge, there is no reported comparison of statistical classification accuracy and classifiers’ discriminative ability on breast cancer recurrence in presence of imputed missing data. Therefore, this article aims to fill this analysis gap by comparing the performance of binary classifiers (logistic regression, linear and quadratic discriminant analysis) using several datasets resulted from imputation process using various simulation conditions. Our study aids the knowledge about how classifiers’ accuracy and discriminative ability in classifying a binary outcome variable are affected by the presence of imputed numerical missing data. We simulated incomplete datasets with 15, 30, 45 and 60% of missingness under Missing At Random (MAR) and Missing Completely At Random (MCAR) mechanisms. Mean imputation, hot deck, k-nearest neighbour, multiple imputations via chained equation, expected-maximisation, and predictive mean matching were used to impute incomplete datasets. For each classifier, correct classification accuracy and area under the Receiver Operating Characteristic (ROC) curves under MAR and MCAR mechanisms were compared. The linear discriminant classifier attained the highest classification accuracy (73.9%) based on mean-imputed data at 45% of missing data under MCAR mechanism. As a classifier, the logistic regression based on predictive mean matching imputed-data yields the greatest areas under ROC curves (0.6418) at 30% missingness while k-nearest neighbour tops the value (0.6428) at 60% of missing data under MCAR mechanism. |
| first_indexed | 2024-04-11T06:10:07Z |
| format | Article |
| id | doaj.art-c8a912b8a2994b56b89c5cd6dd290cc9 |
| institution | Directory Open Access Journal |
| issn | 1756-0381 |
| language | English |
| last_indexed | 2024-04-11T06:10:07Z |
| publishDate | 2022-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BioData Mining |
| spelling | doaj.art-c8a912b8a2994b56b89c5cd6dd290cc92022-12-22T04:41:19ZengBMCBioData Mining1756-03812022-12-0115111310.1186/s13040-022-00316-8Classification of breast cancer recurrence based on imputed data: a simulation studyRahibu A. Abassi0Amina S. Msengwa1Department of Natural Sciences, State University of ZanzibarDepartment of Statistics, University of Dar es SalaamAbstract Several studies have been conducted to classify various real life events but few are in medical fields; particularly about breast recurrence under statistical techniques. To our knowledge, there is no reported comparison of statistical classification accuracy and classifiers’ discriminative ability on breast cancer recurrence in presence of imputed missing data. Therefore, this article aims to fill this analysis gap by comparing the performance of binary classifiers (logistic regression, linear and quadratic discriminant analysis) using several datasets resulted from imputation process using various simulation conditions. Our study aids the knowledge about how classifiers’ accuracy and discriminative ability in classifying a binary outcome variable are affected by the presence of imputed numerical missing data. We simulated incomplete datasets with 15, 30, 45 and 60% of missingness under Missing At Random (MAR) and Missing Completely At Random (MCAR) mechanisms. Mean imputation, hot deck, k-nearest neighbour, multiple imputations via chained equation, expected-maximisation, and predictive mean matching were used to impute incomplete datasets. For each classifier, correct classification accuracy and area under the Receiver Operating Characteristic (ROC) curves under MAR and MCAR mechanisms were compared. The linear discriminant classifier attained the highest classification accuracy (73.9%) based on mean-imputed data at 45% of missing data under MCAR mechanism. As a classifier, the logistic regression based on predictive mean matching imputed-data yields the greatest areas under ROC curves (0.6418) at 30% missingness while k-nearest neighbour tops the value (0.6428) at 60% of missing data under MCAR mechanism.https://doi.org/10.1186/s13040-022-00316-8Classification accuracyImputed dataMissing data mechanismsMissingness percentagesSimulation |
| spellingShingle | Rahibu A. Abassi Amina S. Msengwa Classification of breast cancer recurrence based on imputed data: a simulation study BioData Mining Classification accuracy Imputed data Missing data mechanisms Missingness percentages Simulation |
| title | Classification of breast cancer recurrence based on imputed data: a simulation study |
| title_full | Classification of breast cancer recurrence based on imputed data: a simulation study |
| title_fullStr | Classification of breast cancer recurrence based on imputed data: a simulation study |
| title_full_unstemmed | Classification of breast cancer recurrence based on imputed data: a simulation study |
| title_short | Classification of breast cancer recurrence based on imputed data: a simulation study |
| title_sort | classification of breast cancer recurrence based on imputed data a simulation study |
| topic | Classification accuracy Imputed data Missing data mechanisms Missingness percentages Simulation |
| url | https://doi.org/10.1186/s13040-022-00316-8 |
| work_keys_str_mv | AT rahibuaabassi classificationofbreastcancerrecurrencebasedonimputeddataasimulationstudy AT aminasmsengwa classificationofbreastcancerrecurrencebasedonimputeddataasimulationstudy |