Pneumonia Image Classification Using CNN with Max Pooling and Average Pooling

Nurjannah, Annisa Fitria; Kurniasari, Andi Shafira Dyah; Sari, Zamah; Azhar, Yufis

doi:10.29207/resti.v6i2.4001

Cited by 7 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Max pooling is a commonly used pooling method employed to reduce the dimensions of feature maps. In the 3D context, max pooling operates across three dimensionswidth, height, and depth/time [34][35][36][37].…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

Emotion Recognition and Management in the Tourism Industry During Emergency Events Using Improved Convolutional Neural Network

Qi,

Han

2024

IEEE Access

View full text Add to dashboard Cite

The purpose is to optimize emotion recognition capabilities in the tourism industry during emergency events and enhance management efficiency. Hence, based on the principles of emergency events and emotion recognition, this work has outlined the fundamental emotion recognition process. Additionally, an emotion recognition model is constructed based on a Convolutional Neural Network (CNN), and the process of extracting emotions using a three-dimensional (3D) CNN is proposed. Finally, an attention mechanism is employed to optimize the 3D CNN model, and a comparative analysis of accuracy and precision is conducted using the Bimodal Face and Body Gesture Database (FABO). The research findings reveal that the optimized 3D CNN model exhibits lower error rates in recognizing anxiety emotions, with only 3 samples going unrecognized. Its overall recognition performance is superior to other models. There are variations in the recognition accuracy among different models, but in general, the optimized 3D CNN model performs relatively well across various datasets, achieving recognition accuracies of 83.92%, 82.33%, and 88.81%. Compared to other models, the optimized 3D CNN model demonstrates higher precision in recognizing different emotions, particularly excelling in identifying anger, disgust, and happiness, with precision rates of 97%, 91%, and 94%, respectively. This work has improved the accuracy and efficiency of emotion recognition, providing more intelligent and effective support for emergency event management in the tourism industry.

show abstract

Section: Convolutional Neural Networkmentioning

confidence: 99%

Emotion Recognition and Management in the Tourism Industry During Emergency Events Using Improved Convolutional Neural Network

Qi,

Han

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The primary components of the CNN architecture were built using a series of convolution, normalization, and pooling layers [30] [31]. The convolution layer was in charge of feature extraction and normalization in the interim [32].…”

Section: Convolutional Neural Network (Cnns)mentioning

confidence: 99%

Expert System for Diagnosis of Lung Disease from X-Ray Using CNN and SVM

Zulkifli,

Soeprihatini,

Sfenrianto

et al. 2023

Int. J. Artif. Intell. Res

View full text Add to dashboard Cite

The lung disease diagnosis expert system utilizes human knowledge to diagnose various conditions affecting the lung. Diseases caused by fungal or bacterial infection in the organ can cause inflammation as well as death when it is not detected on time. A standard method to diagnose these conditions is the use of a chest X-ray (CXR), which requires careful examination of the image by an expert. In this study, several CNN and SVM architectural models were proposed to classify CXR images to diagnose whether a person has COVID-19, Viral Pneumonia, Bacterial Pneumonia, Tuberculosis (TB), and Normal. The experiment showed that InceptionV3 had the best results compared to other CNN architectures and SVM. Classification accuracy, precision, recall, and f1-score of CXR images for COVID-19, Viral Pneumonia, Bacterial Pneumonia, TB, and Normal were 0.86, 0.91, 0.91, and 0.91, respectively. This study was based on a deep learning system with different CNN and SVM architectures that can work well on the CXR images dataset for diagnosing lung disease.

show abstract

“…In the modeling process, the hyperparameter tuning process is added to find the model with the best parameters, so that The pooling layer is the division of the feature map of an image into several sub-sections and reducing the parts in new sub-sections. The MaxPool layer will take the maximum value from the feature map, while the AveragePool will take the average value from the feature map to be able to retrieve all image information [21]. Use the GlobalAveragePooling2D layer here to select the average feature value of an image.…”

Section: Model Architecturementioning

confidence: 99%

Classification of Brain Tumors on MRI Images Using Convolutional Neural Network Model EfficientNet

Wibowo

Fayyadl

Azhar

et al. 2022

J. RESTI (Rekayasa Sist. Teknol. Inf.)

View full text Add to dashboard Cite

A brain tumor is a lump caused by an imperfect cell turnover cycle in the brain and can affect all ages. Brain tumors have 4 grades, namely grades 1 to 2 are benign tumor grades, and grades 3 to 4 are malignant tumor grades. Therefore, early identification of brain tumor disease is very important in providing appropriate treatment and treatment. This study uses a dataset obtained through the Kaggle website titled Brain Tumor Classification (MRI). The number of data is 3264 images with details of Glioma tumors (926 images), Meningioma tumors (937 images), pituitary tumors (901 images), and without tumors (500 images). In this study, there are 4 scenarios with different testers. This study proposes the classification of brain tumors using Hyperparameter Tuning and EfficientNet models on MRI images. The EfficientNet model used is the EfficientNetB0 and EfficientNetB7 models with the architecture used are the input layer, GlobalAveragePooling2D layer, dropout layer, and dense layer as well as adding augmentation data to the dataset to manipulate the data in order to improve the results of the proposed model. After building the model, the results of accuracy, precision, recall, and f1-score will be obtained in each scenario. Accuracy results in Scenario 1 are 91%, scenario 2 is 95% accurate, scenario 3 is 95%, and scenario 4 is 98%.

show abstract

Pneumonia Image Classification Using CNN with Max Pooling and Average Pooling

Cited by 7 publications

References 11 publications

Emotion Recognition and Management in the Tourism Industry During Emergency Events Using Improved Convolutional Neural Network

Emotion Recognition and Management in the Tourism Industry During Emergency Events Using Improved Convolutional Neural Network

Expert System for Diagnosis of Lung Disease from X-Ray Using CNN and SVM

Classification of Brain Tumors on MRI Images Using Convolutional Neural Network Model EfficientNet

Contact Info

Product

Resources

About