Distribution-Sensitive Unbalanced Data Oversampling Method for Medical Diagnosis

Han, Weihong; Huang, Zizhong; Li, Shudong; Jia, Yan

doi:10.1007/s10916-018-1154-8

Cited by 52 publications

(28 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the method proposed by [12], the authors tackled data imbalance in a medical diagnosis dataset by introducing a distribution sensitive oversampling approach. In the proposed method, the minority samples were divided into noise samples, unstable samples, boundary samples, and stable samples according to their location in the distribution.…”

Section: Methods Used For the Data Imbalance Problemmentioning

confidence: 99%

Predicting No-show Medical Appointments Using Machine Learning

Alshaya

McCarren

Al-Rasheed

2019

Communications in Computer and Information Science

View full text Add to dashboard Cite

Health care centers face many issues due to the limited availability of resources, such as funds, equipment, beds, physicians, and nurses. Appointment absences lead to a waste of hospital resources as well as endangering patient health. This fact makes unattended medi-cal appointments both socially expensive and economically costly. This research aimed to build a predictive model to identify whether an appointment would be a no-show or not in order to reduce its consequences. This paper proposes a multi-stage framework to build an accurate predictor that also tackles the imbalanced property that the data exhibits. The first stage includes dimensionality reduction to compress the data into its most important components. The second stage deals with the imbalanced nature of the data. Different machine learning algorithms were used to build the classifiers in the third stage. Various evaluation metrics are also discussed and an evaluation scheme that fits the problem at hand is described. The work presented in this paper will help decision makers at health care centers to implement effective strategies to reduce the number of no-shows.

show abstract

Section: Methods Used For the Data Imbalance Problemmentioning

confidence: 99%

Predicting No-show Medical Appointments Using Machine Learning

Alshaya

McCarren

Al-Rasheed

2019

Communications in Computer and Information Science

View full text Add to dashboard Cite

show abstract

“…In [15], Han and coworkers proposed Distribution-Sensitive (DS). This is an oversampling algorithm for Medical Diagnosis for imbalanced data.…”

Section: Related Workmentioning

confidence: 99%

Classification of Guillain–Barré Syndrome Subtypes Using Sampling Techniques with Binary Approach

et al. 2020

View full text Add to dashboard Cite

Guillain–Barré Syndrome (GBS) is an unusual disorder where the body’s immune system affects the peripheral nervous system. GBS has four main subtypes, whose treatments vary among them. Severe cases of GBS can be fatal. This work aimed to investigate whether balancing an original GBS dataset improves the predictive models created in a previous study. purpleBalancing a dataset is to pursue symmetry in the number of instances of each of the classes.The dataset includes 129 records of Mexican patients diagnosed with some subtype of GBS. We created 10 binary datasets from the original dataset. Then, we balanced these datasets using four different methods to undersample the majority class and one method to oversample the minority class. Finally, we used three classifiers with different approaches to creating predictive models. The results show that balancing the original dataset improves the previous predictive models. The goal of the predictive models is to identify the GBS subtypes applying Machine Learning algorithms. It is expected that specialists may use the model to have a complementary diagnostic using a reduced set of relevant features. Early identification of the subtype will allow starting with the appropriate treatment for patient recovery. This is a contribution to exploring the performance of balancing techniques with real data.

show abstract

“…In the training set, oversampling techniques are used to increase the number of minority class so that when a model fitting is carried out, a balanced dataset can be obtained. • Random oversampling method (ROS) performs duplication by selecting a random set of minority classes for random data replication [22]. Because the sampling process is carried out randomly, random oversampling has a disadvantage, namely that it requires a long training time, and there is the possibility of overfitting.…”

Section: Introductionmentioning

confidence: 99%

An in-depth performance analysis of the oversampling techniques for high-class imbalanced dataset

Wibowo

Fatichah

2021

regist. j. ilm. teknol. sist. inf.

View full text Add to dashboard Cite

Class imbalance occurs when the distribution of classes between the majority and the minority classes is not the same. The data on imbalanced classes may vary from mild to severe. The effect of high-class imbalance may affect the overall classification accuracy since the model is most likely to predict most of the data that fall within the majority class. Such a model will give biased results, and the performance predictions for the minority class often have no impact on the model. The use of the oversampling technique is one way to deal with high-class imbalance, but only a few are used to solve data imbalance. This study aims for an in-depth performance analysis of the oversampling techniques to address the high-class imbalance problem. The addition of the oversampling technique will balance each class’s data to provide unbiased evaluation results in modeling. We compared the performance of Random Oversampling (ROS), ADASYN, SMOTE, and Borderline-SMOTE techniques. All oversampling techniques will be combined with machine learning methods such as Random Forest, Logistic Regression, and k-Nearest Neighbor (KNN). The test results show that Random Forest with Borderline-SMOTE gives the best value with an accuracy value of 0.9997, 0.9474 precision, 0.8571 recall, 0.9000 F1-score, 0.9388 ROC-AUC, and 0.8581 PRAUC of the overall oversampling technique.

show abstract

Distribution-Sensitive Unbalanced Data Oversampling Method for Medical Diagnosis

Cited by 52 publications

References 12 publications

Predicting No-show Medical Appointments Using Machine Learning

Predicting No-show Medical Appointments Using Machine Learning

Classification of Guillain–Barré Syndrome Subtypes Using Sampling Techniques with Binary Approach

An in-depth performance analysis of the oversampling techniques for high-class imbalanced dataset

Contact Info

Product

Resources

About