A Comparative Study on the Detection of Pneumonia in Chest X-Ray Images Utilizing Deep Learning Models (Published)
Pneumonia continues to pose a notable public health issue on a global scale, highlighting the crucial need for precise and prompt identification to lessen its effects on patient prognosis. Chest X-ray imaging is a common diagnostic tool for identifying pneumonia due to its non-invasiveness and wide availability. Recently, convolutional neural networks (CNNs), a type of deep learning method, have shown promise in automating the detection of pneumonia from X-ray images. In this paper, we present a comprehensive comparative study of three popular deep learning models—VGG16, Inception V3, and ResNet 50V2—for pneumonia detection in X-ray datasets. The dataset used consists of 5,863 chest X-ray images collected from Kaggle, which are classified into two main categories: pneumonia and normal. The ResNet 50V2 model did very well in the experiments; it was able to correctly identify pneumonia 95.97% of the time, which was better than both VGG16 (93.58%) and Inception V3 (93.38%). We also conduct an analysis of performance metrics such as validation loss, validation accuracy, recall, and precision, and calculate the area under the receiver operating characteristic (ROC) curve (AUC) for each model. We also talk about how to compare ROC curves and precision-recall curves, which lets you see how well the models do at telling the difference between things and how well they do across a number of evaluation metrics. Our study contributes to the development of efficient and high-performance deep learning models for improving the diagnosis and treatment processes for pneumonia patients in the medical field.
Keywords: Chest X-ray, Comparative Analysis, Inception V3, Pneumonia, ResNet 50V2, VGG16, convolutional neural networks, deep learning
Using Deep Learning Convolutional Neural System for Discrete Minke Whale Appreciation (Published)
The only known predictable aggregation of dwarf minke whales .occurs in the Australian offshore waters of the northern Great Barrier Reef in May-August each year. The identification of individual whales is required for research on the whales’ population characteristics and for monitoring the potential impacts of tourism activities, including commercial swims with the whales. At present, it is not cost-effective for researchers to manually process and analyze the tens of thousands of underwater images collated after each observation/tourist season, and a large data base of historical non-identified imagery exists. This study reports the first proof of concept for recognizing individual dwarf minke whales using the Deep Learning Convolutional Neural Networks (CNN).The “off-the-shelf” Image net-trained VGG16 CNN was used as the feature-encoder of the per-pixel sematic seg-mentation Automatic Minke Whale Recognizer (AMWR). The most frequently photographed whale in a sample of 76 individual whales (MW1020) was identified in 179 images out of the total 1320 images provided. Training and image augmentation procedures were developed to compensate for the small number of available images. The trained AMWR achieved 93% prediction accuracy on the testing subset of 36 positive/MW1020 and 228 negative/not-MW1020 images, where each negative image contained at least one of the other 75 whales. Furthermore on the test subset, AMWR achieved 74% precision, 80% recall, and 4% false-positive rate, making the presented approach comparable or better to other state-of-the-art individual animal recognition results.
Keywords: convolutional neural networks, deep learning, dwarf minke whales, image recognition, photo-identification, population biology