Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Author: search.filters.author.Kumar, S.Subject: search.filters.subject.Deep Learning

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Detecting COVID-19 Infection Using Customized Convolutional Neural Network
    (Institute of Electrical and Electronics Engineers Inc., 2024) Kumar, S.; Kisku, B.; Vardhan K, S.H.; Kumar, S.; Bhowmik, B.
    The COVID-19 pandemic has affected 775 million people globally, with an estimated death toll of 7 million. Detection methods like reverse transcription polymer chain reaction (RT-PCR) face multiple challenges, including false positive cases, time-consuming, and high cost. A rapid, precise, affordable screening alternative is essential to expedite COVID-19 detection. Various efforts have focused on expediting COVID-19 detection due to the high costs and logistical challenges associated with traditional methods. This paper proposes a customized deep-learning framework architecture for automatically identifying COVID-19 infection in chest X-ray (CXR) images. Multiple neural networks extract deep features from the CXR images, including popular models such as VGG19, DenseNet201, EfficientNet, MobileNetV2, and InceptionV3. The proposed model undergoes training and testing using the QaTa-COVID-19 dataset. The proposed model achieves classification accuracy of 97.06%, with precision, recall, and F1 score rates for COVID-19 cases recorded at 97.34%, 96.36%, and 97.01%, respectively, for the 4-class cases (COVID vs. Normal vs. Pediatric Bacterial Pneumonia vs. Pediatric Viral Pneumonia). © 2024 IEEE.
  • No Thumbnail Available
    Item
    Automated Segmentation of COVID-19 Infected Lungs via Modified U-Net Model
    (Institute of Electrical and Electronics Engineers Inc., 2024) Kumar, S.; Bhowmik, B.
    The COVID-19 pandemic has led to significant outbreaks in more than 220 countries worldwide, profoundly impacting the public health and lives. As of February 2024, over 774 million cases have been reported, with more than 7,035,337 deaths recorded. Therefore, there is a significant need for automated image segmentation to serve as clinical decision support. This paper presents a novel automated segmentation framework that dynamically generates distinct and randomized image patches for training using preprocessing techniques and extensive data augmentation. The proposed architecture employs a semantic segmentation approach, ensuring accuracy despite limited data availability. Experimental assessment comprises a visual inspection of the predicted segmentation outcomes. Quantitative evaluation of segmentation includes standards performance metrics such as precision, recall, Dice score, and Intersection over Union (IoU). The results exhibit a remarkable Dice coefficient score of 98.3% and an IoU rate surpassing 96.8%, demonstrating the model's robustness in identifying COVID-19-infected lung regions. © 2024 IEEE.