Faculty Publications

Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736

Publications by NITK Faculty

Browse

Subject: search.filters.subject.binary classification

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    An automated deep learning pipeline for detecting user errors in spirometry test
    (Elsevier Ltd, 2024) Bonthada, S.; Pariserum Perumal, S.P.; Naik, P.P.; Mahesh, M.A.; Rajan, J.
    Spirometer is used as a major diagnostic tool for obstructive airway diseases and a monitoring tool for therapy response and disease staging over time. It is a sophisticated medical device employed to quantify flow and volume of air exhaled by a subject during a specific testing period. The essential metrics obtained from the spirometry test, play a crucial role in enabling healthcare professionals to thoroughly evaluate the respiratory health and condition of the individual under examination. Several spirometer measurements including Forced Vital Capacity (FVC) and Forced Expiratory Volume (FEV) serve as guidelines for diagnosis and prognosis of Chronic Obstructive Pulmonary Diseases (COPD) and asthma. However, user errors caused by different reasons, including improper handling of the equipment and poor performance during the maneuvers of the expiratory airflow, end up in incorrect treatment directions. To ensure accurate results, spirometry tests traditionally require the presence of a skilled professional to identify and address these errors promptly. A novel machine learning approach is proposed in this paper to automatically identify four such user errors based on Volume-Time and Flow-Volume graphs. By detecting specific errors and providing immediate feedback to patients, reliability and accuracy of spirometry results will be improved and the need for trained professionals will be reduced. The implementation facilitates the widespread adoption of spirometry, particularly in low-resource telemedicine settings. This work implements a binary classification model distinguishing between normal and error test samples, achieving a prediction accuracy of 93%. Additionally, a 4-way classification model is presented for identifying individual error sub-types, demonstrating a prediction accuracy of 94%. © 2023 Elsevier Ltd
  • No Thumbnail Available
    Item
    Improving the performance of multi-stage HER2 breast cancer detection in hematoxylin-eosin images based on ensemble deep learning
    (Elsevier Ltd, 2025) Pateel, G.P.; Senapati, K.; Pandey, A.K.
    Background: Breast cancer is the most frequently diagnosed cancer among women worldwide, and histopathology is the gold standard in diagnosing the disease. Hematoxylin and Eosin (HE) staining, routinely employed to observe the overall tissue structure, is an affordable and commonly practiced cancer diagnosis. In contrast, Immunohistochemistry (IHC), which detects the increased presence of particular antigens linked to the mutation, can require multiple tests to conduct and is relatively costly. Generally, in computer-aided diagnosis, the conventional methods rely on a single network to extract features. However, these methods have significant limitations and fail to generalize. Methods: In this study, we propose an automated novel weighted average algorithm called HER2-ETNET, which ensembles the chosen three pre-trained deep learning models, DenseNet 201, GoogLeNet, and ResNet-50, to classify breast histopathology HE images into multi-class Human Epidermal Growth Factor Receptor-2 (HER2) status (HER2-0+, HER2-1+, HER2-2+, HER2-3+). The proposed method has the potential to bypass the IHC laboratory test. In this study, we form a weight matrix by fusing together, the scores of False Positive Rate (FPR) and False Negative Rate (FNR) of both training and validation sets, and the computed weights are assigned to the three base learners. This is in contrast to the previous works, in which the weights were generally assigned empirically to the chosen deep learning models, which might be erroneous. Result: The proposed approach is evaluated on the unseen test set, and it achieves accuracy, precision, recall and AUC of 97.44%, 97.32%, 97.39%, and 99.75% respectively. Conclusion: The proposed framework outperforms all the existing methods on the same dataset and is proven to be the reliable method in detecting the HER2 status (HER2-0+, HER2-1+, HER2-2+, HER2-3+) from HE images. This also proves that, HE stained images contain adequate information for efficiently detecting the HER2 status in breast cancer. © 2024 Elsevier Ltd