Gaussian Mixture with Max Expectation Guide for Stacked Architecture of Denoising Autoencoder and DRBM for Medical Chest Scans and Disease Identification
Artificial intelligence (AI), in particular deep learning, has proven to be efficient in medical diagnosis. This paper introduces a new hybrid deep learning model for pneumonia diagnosis based on chest CT scans. At the core of the model, a Gaussian mixture is combined with the expectation-maximizati...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/1/105 |
| _version_ | 1797625983964020736 |
|---|---|
| author | Mona Jamjoom Abeer M. Mahmoud Safia Abbas Rania Hodhod |
| author_facet | Mona Jamjoom Abeer M. Mahmoud Safia Abbas Rania Hodhod |
| author_sort | Mona Jamjoom |
| collection | DOAJ |
| description | Artificial intelligence (AI), in particular deep learning, has proven to be efficient in medical diagnosis. This paper introduces a new hybrid deep learning model for pneumonia diagnosis based on chest CT scans. At the core of the model, a Gaussian mixture is combined with the expectation-maximization algorithm (EMGMM) to extract the regions of interest (ROI), while a convolutional denoising autoencoder (DAE) and deep restricted Boltzmann machine (DRBM) are combined for the classification. In order to prevent the model from learning trivial solutions, stochastic noises were added as an input to the unsupervised learning phase. The dataset used in this work is a publicly available dataset of chest X-rays for pneumonia on the Kaggle website; it contains 5856 images with 1583 normal cases and 4273 pneumonia cases, with an imbalance ratio (IR) of 0.46. Several operations including zooming, flipping, shifting and rotation were used in the augmentation phase to balance the data distribution across the different classes, which led to enhancing the IR value to 0.028. The computational analysis of the results show that the proposed model is promising as it provides an average accuracy value of 98.63%, sensitivity value of 96.5%, and specificity value of 94.8%. |
| first_indexed | 2024-03-11T10:04:07Z |
| format | Article |
| id | doaj.art-a38a0b1d3de44d3ca1881344c08f8535 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T10:04:07Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-a38a0b1d3de44d3ca1881344c08f85352023-11-16T15:11:11ZengMDPI AGElectronics2079-92922022-12-0112110510.3390/electronics12010105Gaussian Mixture with Max Expectation Guide for Stacked Architecture of Denoising Autoencoder and DRBM for Medical Chest Scans and Disease IdentificationMona Jamjoom0Abeer M. Mahmoud1Safia Abbas2Rania Hodhod3Department of Computer Sciences, College of Computer and Information Sciences, Princess Nourah Bint Abdulrahman University, Riyadh 11671, Saudi ArabiaDepartment of Computer Science, Faculty of Computer and Information Sciences, Ain Shams University, Cairo 11566, EgyptDepartment of Computer Science, Faculty of Computer and Information Sciences, Ain Shams University, Cairo 11566, EgyptTSYS School of Computer Science, Turner College of Business, Columbus State University, Columbus, GA 31907, USAArtificial intelligence (AI), in particular deep learning, has proven to be efficient in medical diagnosis. This paper introduces a new hybrid deep learning model for pneumonia diagnosis based on chest CT scans. At the core of the model, a Gaussian mixture is combined with the expectation-maximization algorithm (EMGMM) to extract the regions of interest (ROI), while a convolutional denoising autoencoder (DAE) and deep restricted Boltzmann machine (DRBM) are combined for the classification. In order to prevent the model from learning trivial solutions, stochastic noises were added as an input to the unsupervised learning phase. The dataset used in this work is a publicly available dataset of chest X-rays for pneumonia on the Kaggle website; it contains 5856 images with 1583 normal cases and 4273 pneumonia cases, with an imbalance ratio (IR) of 0.46. Several operations including zooming, flipping, shifting and rotation were used in the augmentation phase to balance the data distribution across the different classes, which led to enhancing the IR value to 0.028. The computational analysis of the results show that the proposed model is promising as it provides an average accuracy value of 98.63%, sensitivity value of 96.5%, and specificity value of 94.8%.https://www.mdpi.com/2079-9292/12/1/105deep learningpneumonia predictionGaussian mixtureconvolution autoencoderBoltzmann machine |
| spellingShingle | Mona Jamjoom Abeer M. Mahmoud Safia Abbas Rania Hodhod Gaussian Mixture with Max Expectation Guide for Stacked Architecture of Denoising Autoencoder and DRBM for Medical Chest Scans and Disease Identification Electronics deep learning pneumonia prediction Gaussian mixture convolution autoencoder Boltzmann machine |
| title | Gaussian Mixture with Max Expectation Guide for Stacked Architecture of Denoising Autoencoder and DRBM for Medical Chest Scans and Disease Identification |
| title_full | Gaussian Mixture with Max Expectation Guide for Stacked Architecture of Denoising Autoencoder and DRBM for Medical Chest Scans and Disease Identification |
| title_fullStr | Gaussian Mixture with Max Expectation Guide for Stacked Architecture of Denoising Autoencoder and DRBM for Medical Chest Scans and Disease Identification |
| title_full_unstemmed | Gaussian Mixture with Max Expectation Guide for Stacked Architecture of Denoising Autoencoder and DRBM for Medical Chest Scans and Disease Identification |
| title_short | Gaussian Mixture with Max Expectation Guide for Stacked Architecture of Denoising Autoencoder and DRBM for Medical Chest Scans and Disease Identification |
| title_sort | gaussian mixture with max expectation guide for stacked architecture of denoising autoencoder and drbm for medical chest scans and disease identification |
| topic | deep learning pneumonia prediction Gaussian mixture convolution autoencoder Boltzmann machine |
| url | https://www.mdpi.com/2079-9292/12/1/105 |
| work_keys_str_mv | AT monajamjoom gaussianmixturewithmaxexpectationguideforstackedarchitectureofdenoisingautoencoderanddrbmformedicalchestscansanddiseaseidentification AT abeermmahmoud gaussianmixturewithmaxexpectationguideforstackedarchitectureofdenoisingautoencoderanddrbmformedicalchestscansanddiseaseidentification AT safiaabbas gaussianmixturewithmaxexpectationguideforstackedarchitectureofdenoisingautoencoderanddrbmformedicalchestscansanddiseaseidentification AT raniahodhod gaussianmixturewithmaxexpectationguideforstackedarchitectureofdenoisingautoencoderanddrbmformedicalchestscansanddiseaseidentification |