Integrating Enhanced Sparse Autoencoder-Based Artificial Neural Network Technique and Softmax Regression for Medical Diagnosis
In recent times, several machine learning models have been built to aid in the prediction of diverse diseases and to minimize diagnostic errors made by clinicians. However, since most medical datasets seem to be imbalanced, conventional machine learning algorithms tend to underperform when trained w...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-11-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/9/11/1963 |
| _version_ | 1797547271581073408 |
|---|---|
| author | Sarah A. Ebiaredoh-Mienye Ebenezer Esenogho Theo G. Swart |
| author_facet | Sarah A. Ebiaredoh-Mienye Ebenezer Esenogho Theo G. Swart |
| author_sort | Sarah A. Ebiaredoh-Mienye |
| collection | DOAJ |
| description | In recent times, several machine learning models have been built to aid in the prediction of diverse diseases and to minimize diagnostic errors made by clinicians. However, since most medical datasets seem to be imbalanced, conventional machine learning algorithms tend to underperform when trained with such data, especially in the prediction of the minority class. To address this challenge and proffer a robust model for the prediction of diseases, this paper introduces an approach that comprises of feature learning and classification stages that integrate an enhanced sparse autoencoder (SAE) and Softmax regression, respectively. In the SAE network, sparsity is achieved by penalizing the weights of the network, unlike conventional SAEs that penalize the activations within the hidden layers. For the classification task, the Softmax classifier is further optimized to achieve excellent performance. Hence, the proposed approach has the advantage of effective feature learning and robust classification performance. When employed for the prediction of three diseases, the proposed method obtained test accuracies of 98%, 97%, and 91% for chronic kidney disease, cervical cancer, and heart disease, respectively, which shows superior performance compared to other machine learning algorithms. The proposed approach also achieves comparable performance with other methods available in the recent literature. |
| first_indexed | 2024-03-10T14:41:58Z |
| format | Article |
| id | doaj.art-664cc7dbfa164fe189618d62ca13bf4f |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T14:41:58Z |
| publishDate | 2020-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-664cc7dbfa164fe189618d62ca13bf4f2023-11-20T21:42:35ZengMDPI AGElectronics2079-92922020-11-01911196310.3390/electronics9111963Integrating Enhanced Sparse Autoencoder-Based Artificial Neural Network Technique and Softmax Regression for Medical DiagnosisSarah A. Ebiaredoh-Mienye0Ebenezer Esenogho1Theo G. Swart2Center for Telecommunication, Department of Electrical and Electronic Engineering Science, University of Johannesburg, Johannesburg 2006, South AfricaCenter for Telecommunication, Department of Electrical and Electronic Engineering Science, University of Johannesburg, Johannesburg 2006, South AfricaCenter for Telecommunication, Department of Electrical and Electronic Engineering Science, University of Johannesburg, Johannesburg 2006, South AfricaIn recent times, several machine learning models have been built to aid in the prediction of diverse diseases and to minimize diagnostic errors made by clinicians. However, since most medical datasets seem to be imbalanced, conventional machine learning algorithms tend to underperform when trained with such data, especially in the prediction of the minority class. To address this challenge and proffer a robust model for the prediction of diseases, this paper introduces an approach that comprises of feature learning and classification stages that integrate an enhanced sparse autoencoder (SAE) and Softmax regression, respectively. In the SAE network, sparsity is achieved by penalizing the weights of the network, unlike conventional SAEs that penalize the activations within the hidden layers. For the classification task, the Softmax classifier is further optimized to achieve excellent performance. Hence, the proposed approach has the advantage of effective feature learning and robust classification performance. When employed for the prediction of three diseases, the proposed method obtained test accuracies of 98%, 97%, and 91% for chronic kidney disease, cervical cancer, and heart disease, respectively, which shows superior performance compared to other machine learning algorithms. The proposed approach also achieves comparable performance with other methods available in the recent literature.https://www.mdpi.com/2079-9292/9/11/1963sparse autoencoderunsupervised learningSoftmax regressionmedical diagnosismachine learningartificial neural network |
| spellingShingle | Sarah A. Ebiaredoh-Mienye Ebenezer Esenogho Theo G. Swart Integrating Enhanced Sparse Autoencoder-Based Artificial Neural Network Technique and Softmax Regression for Medical Diagnosis Electronics sparse autoencoder unsupervised learning Softmax regression medical diagnosis machine learning artificial neural network |
| title | Integrating Enhanced Sparse Autoencoder-Based Artificial Neural Network Technique and Softmax Regression for Medical Diagnosis |
| title_full | Integrating Enhanced Sparse Autoencoder-Based Artificial Neural Network Technique and Softmax Regression for Medical Diagnosis |
| title_fullStr | Integrating Enhanced Sparse Autoencoder-Based Artificial Neural Network Technique and Softmax Regression for Medical Diagnosis |
| title_full_unstemmed | Integrating Enhanced Sparse Autoencoder-Based Artificial Neural Network Technique and Softmax Regression for Medical Diagnosis |
| title_short | Integrating Enhanced Sparse Autoencoder-Based Artificial Neural Network Technique and Softmax Regression for Medical Diagnosis |
| title_sort | integrating enhanced sparse autoencoder based artificial neural network technique and softmax regression for medical diagnosis |
| topic | sparse autoencoder unsupervised learning Softmax regression medical diagnosis machine learning artificial neural network |
| url | https://www.mdpi.com/2079-9292/9/11/1963 |
| work_keys_str_mv | AT sarahaebiaredohmienye integratingenhancedsparseautoencoderbasedartificialneuralnetworktechniqueandsoftmaxregressionformedicaldiagnosis AT ebenezeresenogho integratingenhancedsparseautoencoderbasedartificialneuralnetworktechniqueandsoftmaxregressionformedicaldiagnosis AT theogswart integratingenhancedsparseautoencoderbasedartificialneuralnetworktechniqueandsoftmaxregressionformedicaldiagnosis |