Dr. Bhavanishankar K

Scopus Publications

Multi-scale Attention Feature Map based USegNet for Lung Nodule Segmentation and DCFL-CNN for Classification Using Computed Tomography Images
The Intelligent Networks and Systems Society

Volumetric ResNet for Enhanced Lung Nodule Classification: A High-Accuracy Approach using Deep Learning
V Asha and K Bhavanishankar
IEEE
Lung cancer remains a leading cause of mortality, affecting both men and women, has posed significant challenges for accurate identification for decades. Recent advancements in Deep Learning (DL) have revolutionized the field of Computer-Aided Diagnosis (CAD) for detection and classification of lung cancer. This study introduces a novel approach leveraging Volumetric ResNet architecture for accurate classification of candidate Lung Nodules into Nodule and Non-Nodule categories. The pipeline begins with histogram equalization for enhanced Computed Tomography (CT) scan contrast, followed by Volumetric ResNet for classification with false positive reduction. Evaluation on the LUNA16 dataset demonstrates the superiority of the proposed model in terms of accuracy (99.49%), specificity (99.74%), sensitivity (100%), precision (99.75%) with augmentation and of accuracy (99.26%), specificity (98.89%), sensitivity (99.63%), precision (98.90%), False Positive Rate (FPR) 1.11% and False Negative Rate (FNR) 0.37% without augmentation, it surpasses the existing state-of-the-art methods.

Towards Efficient Lung Cancer Detection: V-Net-basSegmentation of Pulmonary Nodules
Asha V and Bhavanishankar K
International Association of Online Engineering (IAOE)
The novel approach uses the V-Net architecture to segment pulmonary nodules from computed tomography (CT) scans, enhancing lung cancer detection’s efficiency. Addressing lung cancer, a major global mortality cause, underscores the urgency for improved diagnostic methods. The aim of this research is to refine segmentation, a critical step for early cancer detection. The study leverages V-Net, a three-dimensional (3D) convolutional neural network (CNN) tailored for medical image segmentation, applied to lung nodule identification. It utilizes the LUNA16 dataset, containing 888 annotated CT images, for model training and evaluation. This dataset’s variety of pulmonary conditions allows for a comprehensive method of assessment. The tailored V-Net architecture is optimized for lung nodule segmentation, with a focus on data preprocessing to elevate input image quality. Outcomes reveal significant progress in segmentation precision, achieving a loss score of 0.001 and a mIOU of 98%, setting new standards in the domain. Visuals of segmented lung nodules illustrate the method’s effectiveness, indicating a promising avenue for early lung cancer detection and potentially better patient prognoses. The study contributes significantly to enhancing lung cancer diagnostic methodologies through advanced image analysis. An improved segmentation method based on V-Net architecture surpasses current techniques and encourages further deep learning exploration in medical diagnostics.

Strategic Teaching Enhancement through Predictive Analysis for Individuals (STEP.AI)
Harsha S, Sreevidya Rampura Chandrappa, Priyanga P, and Bhavanishankar K
IEEE
In the evolving landscape of educational technology, predictive assessment using learning level classification has emerged as a pivotal tool for enhancing personalized learning experiences. This research paper delves into the methodologies and efficacy of predictive assessment models that classify learners' proficiency levels to forecast their future academic performance. By leveraging machine learning algorithms and extensive educational data, our study develops a robust framework capable of dynamically assessing student capabilities and predicting their learning trajectories. The proposed regression-based model integrates a variety of features including prior academic records, engagement metrics, and cognitive skills assessments to create a comprehensive learning profile for each student. The research findings demonstrate that predictive assessment models can significantly improve the accuracy of proficiency level classification, thus enabling educators to tailor instructional strategies to individual student needs. The implementation of these models in real-world classroom settings shows a marked improvement in student outcomes, as the predictions allow for timely interventions and support. Moreover, this research highlights the potential of predictive assessments to identify at-risk students early, providing a proactive approach to educational support. In conclusion, the integration of predictive assessment and learning level classification represents a transformative approach in education, promising enhanced educational experiences and outcomes through data-driven insights. Future work will focus on refining these models to accommodate diverse learning environments and further validating their effectiveness across different educational contexts.

Hand Gesture based Computer Screen Control
A N Ramya Shree, R C Sreevidya, K Bhavanishankar, Aryan Prasad, G Aishwarya, and Akash Devappa
IEEE
The hand gesture-based Computer Screen Control approach has had a lot of popularity in these years. This article examines recent advancements in hand gesture recognition technology and its applications across diverse fields such as gaming, robotics, virtual reality, and sign language recognition. The challenges encountered by researchers in developing precise and dependable gesture recognition systems by considering variability in hand shapes, varying lighting conditions, and occlusion, are explored. The document offers an overview of different methodologies and techniques employed in gesture recognition and computer screen control. This search also supports real-time gesture recognition, multi-modal interaction, and integrating language processing with gestures as further enhancements.

Lung Cancer Detection using CT Scans: Image Processing through Deep Learning - A Review
Asha V and Bhavanishankar K
IEEE
Lung cancer being one of the catastrophic diseases is haunting mankind from past seven decades. Unfortunately, early detection of lung cancer is unlikely, hence leading to highest mortality rates. However, various imaging modalities including Computed Tomography (CT) helps in detecting the lung cancer possible at the earliest. Processing such huge data of CT scans is highly time demanding and Computed Aided Diagnosis system (CAD)does a great job from image acquisition till the detection/classification of lung nodules through series of processing stages. This research study covers all the processing stages and major contributions in those stages. This study also summarizes various methods used in basic image processing through deep learning algorithms. A tabulation of various datasets and metrics descriptions is also discussed.

Model Design to Analyse Coronary Artery Disease Using Machine Learning Techniques (MDACADMLT)
R. C. Sreevidya, K. Bhavanishankar, and G. Jalaja
Springer Nature Singapore

Tailor-Made Teller: A Hybrid Screen Reader
Anagha Naga Krishna, Bhamini N Kashyap, Jahnavi T A, Pooja K Bhat, and Bhavanishankar K
IEEE
The Tailor-Made Teller (TMT) serves as a file screen reader, which enables the user to upload a file or text, to have it read aloud along with text highlights; which is relevant for students, people with learning disabilities and the visually impaired. Existing screen readers enable users to experience the Text-to-Speech functionality as an accessibility tool for web pages and on-screen text. However, the currently available systems pertain to specific operating systems and do not support various file formats for a free cost. The Tailor-made teller extracts text from various file programs and leverages the Google Text-to-Speech API to obtain an audio file, which contains the converted speech. The text from the uploaded file gets displayed on the user's screen, where the audio and text highlights run in synchrony. TMT has been tested on Image, PDF, Text and docx files of various sizes and possesses an average accuracy of 98.9%.

Classification of lung nodules into benign or malignant and development of a CBIR system for lung CT scans
K. Bhavanishankar and M. V. Sudhamani
Springer International Publishing

Novel techniques for classification of lung nodules using deep learning approach
K. Bhavanishankar and M. V. Sudhamani
Bentham Science Publishers Ltd.
Objective: Lung cancer is proving to be one of the deadliest diseases that is haunting mankind in recent years. Timely detection of the lung nodules would surely enhance the survival rate. This paper focusses on the classification of candidate lung nodules into nodules/non-nodules in a CT scan of the patient. A deep learning approach –autoencoder is used for the classification. Investigation/Methodology: Candidate lung nodule patches obtained as the results of the lung segmentation are considered as input to the autoencoder model. The ground truth data from the LIDC repository is prepared and is submitted to the autoencoder training module. After a series of experiments, it is decided to use 4-stacked autoencoder. The model is trained for over 600 LIDC cases and the trained module is tested for remaining data sets. Results: The results of the classification are evaluated with respect to performance measures such as sensitivity, specificity, and accuracy. The results obtained are also compared with other related works and the proposed approach was found to be better by 6.2% with respect to accuracy. Conclusion: In this paper, a deep learning approach –autoencoder has been used for the classification of candidate lung nodules into nodules/non-nodules. The performance of the proposed approach was evaluated with respect to sensitivity, specificity, and accuracy and the obtained values are 82.6%, 91.3%, and 87.0%, respectively. This result is then compared with existing related works and an improvement of 6.2% with respect to accuracy has been observed.

Filter based approach for automated detection of candidate lung nodules in 3D computed tomography images
K. Bhavanishankar and M. V. Sudhamani
Springer Singapore

Dr. Bhavanishankar K

EDUCATION

RESEARCH, TEACHING, or OTHER INTERESTS

FUTURE PROJECTS

Scopus Publications

RECENT SCHOLAR PUBLICATIONS

Publications

Dr. Bhavanishankar K

EDUCATION

RESEARCH, TEACHING, or OTHER INTERESTS

FUTURE PROJECTS

Am Alert

Scopus Publications

RECENT SCHOLAR PUBLICATIONS

Publications