Faculty Publications

Now showing 1 - 3 of 3

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 Ltd
The use of cloud based machine learning to predict outcome in intracerebral haemorrhage without explicit programming expertise
(Springer Science and Business Media Deutschland GmbH, 2024) Hegde, A.; Vijayasenan, D.; Mandava, P.; Menon, G.
Machine Learning (ML) techniques require novel computer programming skills along with clinical domain knowledge to produce a useful model. We demonstrate the use of a cloud-based ML tool that does not require any programming expertise to develop, validate and deploy a prognostic model for Intracerebral Haemorrhage (ICH). The data of patients admitted with Spontaneous Intracerebral haemorrhage from January 2015 to December 2019 was accessed from our prospectively maintained hospital stroke registry. 80% of the dataset was used for training, 10% for validation, and 10% for testing. Seventeen input variables were used to predict the dichotomized outcomes (Good outcome mRS 0–3/ Bad outcome mRS 4–6), using machine learning (ML) and logistic regression (LR) models. The two different approaches were evaluated using Area Under the Curve (AUC) for Receiver Operating Characteristic (ROC), Precision recall and accuracy. Our data set comprised of a cohort of 1000 patients. The data was split 8:1 for training & testing respectively. The AUC ROC of the ML model was 0.86 with an accuracy of 75.7%. With LR AUC ROC was 0.74 with an accuracy of 73.8%. Feature importance chart showed that Glasgow coma score (GCS) at presentation had the highest relative importance, followed by hematoma volume and age in both approaches. Machine learning models perform better when compared to logistic regression. Models can be developed by clinicians possessing domain expertise and no programming experience using cloud based tools. The models so developed lend themselves to be incorporated into clinical workflow. © The Author(s) 2024.
Generalizable DNN model for brain tumor sub-structure segmentation from low-resolution 2D multimodal MR Images
(Elsevier Ltd, 2025) Bhaskaracharya, B.; Nair, R.P.; Prakashini, K.; R, G.M.; Litvak, P.; Mandava, P.; Vijayasenan, D.; Sumam David, S.D.
Segmenting subregions within gliomas are critical for effective treatment planning of brain tumors. However, traditional methods of analyzing these regions using multiple MRI modalities are time-consuming, tedious, and subjective. To address these challenges, automatic segmentation models have been developed but are often built with complex 3D architecture using 3D MRI data. Also, brain tumor substructure segmentation is a highly class-imbalanced problem. To overcome these limitations, we propose two models that work on low-resolution 2D MRI data, widely used in resource-constrained countries. One model employs training a 2D U-NeT model using proposed hard sampling approach, demonstrating its effectiveness in segmenting gliomas, especially in datasets with extreme class imbalance. Another model incorporates pointwise and depthwise convolutions in each convolutional layer, enabling efficient information processing and feature learning. By ensembling the prediction maps of these models, we further improve overall segmentation performance. Our models were evaluated on the BraTS2018 dataset, achieving dice scores of 0.78 for Enhancing Tumor (ET), 0.82 for Tumor Core (TC), and 0.87 for Whole Tumor (WT). On a tertiary care hospital dataset, dice scores of 0.68 (ET), 0.75 (TC), and 0.84 (WT) were obtained, demonstrating their robustness and proximity to state-of-the-art methods. In summary, the proposed models offer efficient and reliable segmentation of glioma subregions. Their high dice scores, and computational efficiency, make them valuable tools for treatment planning and advancements in brain tumor segmentation. © 2024 Elsevier Ltd

Faculty Publications

Browse

Filters

Settings

Sort By

Results per page

Search Results