Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
7 results
Search Results
Item Categorizing Relations via Semi-supervised Learning Using a Hybrid Tolerance Rough Sets and Genetic Algorithm Approach(Springer Science and Business Media Deutschland GmbH, 2022) Agrawal, S.; Ahmed, R.; Anand Kumar, M.; Ramanna, S.In the last few decades, we have seen a tremendous increase in the amount of data available on the web. There have been significant advances in constructing knowledge bases consisting of relations from the text data. These relations are words in the text often represented as pairs (Noun, Context), for example (Disease, Symptom), which can be classified into some predefined category to give us some useful information. Categorization of relations using tolerance-rough set based semi-supervised learning algorithm (TPL) have been successfully demonstrated in several works. However, an unexplored problem is the automatic selection of hyper parameters of the TPL algorithm. This paper proposes a genetic algorithm-based approach (TPL-GA) for optimizing the hyper-parameters that are fundamental to the TPL algorithm. The proposed approach was tested on two standard datasets drawn from different domains representing two different languages: English and Hindi text. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.Item Identifying Similar Questions in the Medical Domain Using a Fine-tuned Siamese-BERT Model(Institute of Electrical and Electronics Engineers Inc., 2022) Merchant, A.; Shenoy, N.; Bharali, A.; Anand Kumar, A.M.A large number of people search about their health related problems on the web. However, the number of sites with qualified and verified people answering their queries is quite low in comparison to the number of questions being put up. The rate of queries being searched on such sites has further increased due to the COVID-19 pandemic. The main reason people find it difficult to find solutions to their queries is due to ineffective identification of semantically similar questions in the medical domain. For most cases, answers to the queries people ask would be present, the only caveat being the question may be present in a different form than the one asked by the particular user. In this research, we propose a Siamese-based BERT model to detect similar questions using a fine-tuning approach. The network is fine-tuned with medical question-answer pairs and then with question-question pairs to get a better question similarity prediction. © 2022 IEEE.Item Semi-supervised Semantic Segmentation of Effusion Cytology Images Using Adversarial Training(Springer Science and Business Media Deutschland GmbH, 2023) Rajpurohit, M.; Aboobacker, S.; Vijayasenan, D.; Sumam David, S.; Suresh, P.K.; Sreeram, S.In pleural effusion, an excessive amount of fluid gets accumulated inside the pleural cavity along with signs of inflammation, infections, malignancies, etc. Usually, a manual cytological test is performed to detect and diagnose pleural effusion. The deep learning solutions for effusion cytology include a fully supervised model trained on effusion cytology images with the help of output maps. The low-resolution cytology images are harder to label and require the supervision of an expert, the labeling process time-consuming and expensive. Therefore, we have tried to use some portion of data without any labels for training our models using the proposed semi-supervised training methodology. In this paper, we proposed an adversarial network-based semi-supervised image segmentation approach to automate effusion cytology. The semi-supervised methodology with U-Net as the generator shows nearly 12% of absolute improvement in the f-score of benign class, 8% improvement in the f-score of malignant class, and 5% improvement in mIoU score as compared to a fully supervised U-Net model. With ResUNet++ as a generator, a similar improvement in the f-score of 1% for benign class, 8% for the malignant class, and 1% in the mIoU score is observed as compared to a fully supervised ResUNet++ model. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.Item Semi-supervised Semantic Segmentation for Effusion Cytology Images(Springer Science and Business Media Deutschland GmbH, 2023) Aboobacker, S.; Vijayasenan, D.; Sumam David, S.; Suresh, P.K.; Sreeram, S.Cytopathologists analyse images captured at different magnifications to detect the malignancies in effusions. They identify the malignant cell clusters from the lower magnification, and the identified area is zoomed in to study cell level details in high magnification. The automatic segmentation of low magnification images saves scanning time and storage requirements. This work predicts the malignancy in the effusion cytology images at low magnification levels such as 10 × and 4 ×. However, the biggest challenge is the difficulty in annotating the low magnification images, especially the 4 × data. We extend a semi-supervised learning (SSL) semantic model to train unlabelled 4 × data with the labelled 10 × data. The benign F-score on the predictions of 4 × data using the SSL model is improved 15% compared with the predictions of 4 × data on the semantic 10 × model. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.Item Semantic segmentation of low magnification effusion cytology images: A semi-supervised approach(Elsevier Ltd, 2022) Aboobacker, S.; Vijayasenan, D.; Sumam David, S.; Suresh, P.K.; Sreeram, S.Cytopathologists examine microscopic images obtained at various magnifications to identify malignancy in effusions. They locate the malignant cell clusters at a low magnification and then zoom in to investigate cell-level features at a high magnification. This study predicts the malignancy at low magnification levels such as 4X and 10X in effusion cytology images to reduce scanning time. However, the most challenging problem is annotating the low magnification images, particularly the 4X images. This paper extends two semi-supervised learning (SSL) models, MixMatch and FixMatch, for semantic segmentation. The original FixMatch and MixMatch algorithms are designed for classification tasks. While performing image augmentation, the generated pseudo labels are spatially altered. We introduce reverse augmentation to compensate for the effect of the spatial alterations. The extended models are trained using labelled 10X and unlabelled 4X images. The average F-score of benign and malignant pixels on the predictions of 4X images is improved approximately by 9% for both Extended MixMatch and Extended FixMatch respectively compared with the baseline model. In the Extended MixMatch, 62% sub-regions of low magnification images are eliminated from scanning at a higher magnification, thereby saving scanning time. © 2022 Elsevier LtdItem A Dual-Stage Semi-Supervised Pre-Training Approach for Medical Image Segmentation(Institute of Electrical and Electronics Engineers Inc., 2024) Aralikatti, R.C.; Pawan, S.J.; Rajan, J.Deep neural networks have played a vital role in developing automated methods for addressing medical image segmentation. However, their reliance on labeled data impedes the practicability. Semi-Supervised learning is gaining attention for its intrinsic ability to extract valuable information from labeled and unlabeled data with improved performance. Recently, consistency regularization methods have gained interest due to their efficient learning procedures. They are, however, confined to data or network-level perturbations, negating the benefit of having both forms in a single framework. In light of this, we ask an intriguing but unexplored question: Can we have both network-level and data-level perturbation in the semi-supervised framework? To this end, we present a holistic approach that integrates data-level perturbation in the model pre-training stage, followed by implicit network-level perturbation in the fine-tuning stage. Furthermore, we incorporate networks with manifold learning paradigms throughout the training to facilitate the formation of robust data representations by ensuring local and global semantic affinities adhering to the theory of consensus. Notably, this may be the first attempt in the semi-supervised medical image segmentation archetype to use data and network-level perturbation with a model pre-training strategy. We extensively validated the efficacy of the proposed framework on three benchmark datasets, namely the Automated Cardiac Diagnosis Challenge, ISIC-2018, and Left Atrial Segmentation Challenge datasets, subjected to severely low-sampled labeled data. Notably, in ACDC (4%), ISIC-2018 (5%), and LA (6%) labeled cases, the proposed method outperforms the second-best method by 2.95%, 1.31%, and 0.71% in the Dice Similarity Metric. © 2023 IEEE.Item Multistage Image Reconstruction and Attention-Based Semi-Supervised Learning for Medical Image Segmentation(SAGE Publications Ltd, 2025) Gawas, P.; Kamath S, S.; Singh, A.; Gurupur, V.Automated segmentation of medical images is critical in detecting and diagnosing various conditions. In recent years, supervised deep learning (DL) techniques have been widely researched. However, their application is often limited by the availability of annotated data in the medical domain. To address this, recent studies have explored semi-supervised techniques, though very few of these works focus on skin-lesion segmentation. In addition, they struggle to effectively capture contextual features to delineate the region of interest from the surrounding tissues in the image, which is crucial for accurate segmentation. In this article, a semi-supervised approach for medical image segmentation called MIRA (Medical Image Reconstruction and Analysis) is proposed, which uses adaptive-attention U-Net (AA-U-Net) trained on pseudo-labels generated with a lightweight feature-consistent encoder-decoder network (FCED-Net) to address these challenges. A case study focusing on the precise segmentation of malignant skin lesions is considered for our experiments, as the scarcity of extensive annotated dermatology data limits the effectiveness of traditional DL models. The proposed pipeline is validated and tested using two standard datasets, ISIC2016 and PH2. With only 50% annotated samples, the proposed approach demonstrated promising performance with DSC, IoU, and accuracy of 0.96, 0.92, and 0.85 on ISIC2016 and 0.93, 0.88, and 0.93 on cross-data testing with PH 2 dataset. When benchmarked against leading edge models trained on 100% labeled data, MIRA achieved promising results and even outperformed in some cases. These findings show that it can significantly reduce manual annotation requirements while achieving segmentation performance comparable to models trained on fully annotated skin lesion data. © The Author(s) 2025