Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
14 results
Search Results
Item Security issues and challenges in Healthcare Automated Devices(Institute of Electrical and Electronics Engineers Inc., 2020) Jangid, A.; Dubey, P.K.; Chandavarkar, B.R.Automated devices can be seen everywhere be at home, office, medical devices, Mobiles, etc. This paper presents some of the healthcare-related automated devices with their shortcomings related to security. We are addicted to automated devices and in near future, we will be watching new emergence of devices with the increasing power of automation devices and their security is a big concern as the credibility of a machine is questionable and it's related to automated devices and we are left with many challenges to resolve those security threats. This paper reviews the automation devices primarily in the healthcare field and their security-related issues along with the challenges that we might face in the future while using them. Some already available solutions are presented to try to come up with possible new solutions. © 2020 IEEE.Item TestQuBE: A Testbench Enhancement Methodology for Universal Serial Interfaces in Complex SoCs(IEEE Computer Society, 2021) Kulkarni, A.; Singh, A.; Waje, S.A.; Kashide, S.S.; Choi, S.B.With the increasing computational requirements of complex Systems on Chips (SoCs), the number of Universal Serial Interface (USI) instances have been scaled up, for handling data from ever greater number of peripheral components. This has increased the testbench (TB) complexity during verification, requiring numerous test vectors across multiple iterations to validate the design logic, negatively affecting turn-around-time. In this paper, we propose the TestQuBE (Testbench Quality Benchmark Enhancement) methodology which targets four TB quality benchmarking metrics: total TB development time, resource utilization, functional coverage development efficiency. TestQuBE generates an enhanced testbench with automated components that target the aforementioned TB quality metrics. Simulation results for a real-time use-case scenario show 76.2% improvement in functional coverage for up to 75% faster TB development time, resulting in a 91.6% reduction in resource utilization and approximately 50x greater development efficiency, compared to deployed solutions. © 2021 IEEE.Item A Comprehensive Review on Scaling Machine Learning Workflows Using Cloud Technologies and DevOps(Institute of Electrical and Electronics Engineers Inc., 2025) Ramesh, G.; Vaikunta Pai, T.; Birǎu, R.; Poojary, K.K.; Abhay; Shingad, A.R.; Sowjanya, N.; Popescu, V.; Mitroi, A.T.; Nioata, R.M.; Kiran Raj, K.M.Scaling Machine Learning (ML) workflows in cloud environments presents critical challenges in ensuring reproducibility, low-latency inference, infrastructure reliability, and regulatory compliance. This review addresses the lack of a comprehensive synthesis of how integrated DevOps practices and cloud-native technologies enable scalable, production-grade ML systems. We analyze the convergence of MLOps with tools such as Kubernetes, Jenkins, and Terraform, detailing their role in automating CI/CD pipelines, infrastructure provisioning, and model lifecycle management. The main highlights strategies for optimizing resource utilization, minimizing inference latency, and managing data versioning across hybrid and multi-cloud architectures (AWS, Azure, GCP). We also examine serverless computing, container orchestration, and monitoring practices to enhance scalability and governance. By categorizing challenges chronologically and evaluating emerging practices such as federated learning and security-by-design, this work bridges a key gap in existing literature. It offers a unified perspective on building reliable, reproducible, and compliant ML workflows, thereby advancing the state of scalable AI system engineering. © IEEE. 2013 IEEE.Item Epileptic EEG detection using neural networks and post-classification(2008) Patnaik, L.M.; Manyam, O.K.Electroencephalogram (EEG) has established itself as an important means of identifying and analyzing epileptic seizure activity in humans. In most cases, identification of the epileptic EEG signal is done manually by skilled professionals, who are small in number. In this paper, we try to automate the detection process. We use wavelet transform for feature extraction and obtain statistical parameters from the decomposed wavelet co-efficients. A feed-forward backpropagating artificial neural network (ANN) is used for the classification. We use genetic algorithm for choosing the training set and also implement a post-classification stage using harmonic weights to increase the accuracy. Average specificity of 99.19%, sensitivity of 91.29% and selectivity of 91.14% are obtained. © 2008 Elsevier Ireland Ltd. All rights reserved.Item An adoption model describing clinician’s acceptance of automated diagnostic system for tuberculosis(Springer Verlag service@springer.de, 2016) Panicker, R.O.; Soman, B.; Gangadharan, K.V.; Sobhana, N.V.Computerised medical diagnosing systems are very important to all healthcare professionals, especially clinicians who help in clinical decision-making in complex situations. The acceptance of automated or computerised medical diagnosing system for Tuberculosis (TB) among clinicians is very essential for its effective implementation and usage. This paper proposes a framework that aims to examine factors that influence clinician’s acceptance and use of computerised TB detection system. An extended Unified Theory of Acceptance and Use of Technology (UTAUT) model is adopted in the healthcare context of a developing country for this purpose. The proposed framework is expected to help researchers and clinicians to assess the uptake of modern technology by health care professionals and the tool could be used in other healthcare contexts also. This paper also reviewed previous research adopting UTAUT model, for identifying the constructs promoting the adoption of technology acceptance in health care context. © 2016, IUPESM and Springer-Verlag Berlin Heidelberg.Item Fully automatic ROI extraction and edge-based segmentation of radius and ulna bones from hand radiographs(PWN-Polish Scientific Publishers bbe@ibib.waw.pl, 2017) Simu, S.; Lal, S.; Nagarsekar, P.; Naik, A.Bone age is a reliable measure of person's growth and maturation of skeleton. The difference between chronological age and bone age indicates presence of endocrinological problems. The automated bone age assessment system (ABAA) based on Tanner and Whitehouse method (TW3) requires monitoring the growth of radius, ulna and short bones (phalanges) of left hand. In this paper, a detailed analysis of two bones in the bone age assessment system namely, radius and ulna is presented. We propose an automatic extraction method for the region of interest (ROI) of radius and ulna bones from a left hand radiograph (RUROI). We also propose an improved edge-based segmentation technique for those bones. Quantitative and qualitative results of the proposed segmentation technique are evaluated and compared with other state-of-the-art segmentation techniques. Medical experts have also validated the qualitative results of proposed segmentation technique. Experimental results reveal that these proposed techniques provide better segmentation accuracy as compared to the other state-of-the-art segmentation techniques. © 2017 Nalecz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of SciencesItem A benchmark study of automated intra-retinal cyst segmentation algorithms using optical coherence tomography B-scans(Elsevier Ireland Ltd, 2018) Girish, G.N.; Anima, V.A.; Kothari, A.R.; Sudeep, P.V.; Roychowdhury, S.; Rajan, J.(Background and objectives) Retinal cysts are formed by accumulation of fluid in the retina caused by leakages from inflammation or vitreous fractures. Analysis of the retinal cystic spaces holds significance in detection and treatment of several ocular diseases like age-related macular degeneration, diabetic macular edema etc. Thus, segmentation of intra-retinal cysts and quantification of cystic spaces are vital for retinal pathology and severity detection. In the recent years, automated segmentation of intra-retinal cysts using optical coherence tomography B-scans has gained significant importance in the field of retinal image analysis. The objective of this paper is to compare different intra-retinal cyst segmentation algorithms for comparative analysis and benchmarking purposes. (Methods) In this work, we employ a modular approach for standardizing the different segmentation algorithms. Further, we analyze the variations in automated cyst segmentation performances and method scalability across image acquisition systems by using the publicly available cyst segmentation challenge dataset (OPTIMA cyst segmentation challenge). (Results) Several key automated methods are comparatively analyzed using quantitative and qualitative experiments. Our analysis demonstrates the significance of variations in signal-to-noise ratio (SNR), retinal layer morphology and post-processing steps on the automated cyst segmentation processes. (Conclusion) This benchmarking study provides insights towards the scalability of automated processes across vendor-specific imaging modalities to provide guidance for retinal pathology diagnostics and treatment processes. © 2017 Elsevier B.V.Item Automatic detection of tuberculosis bacilli from microscopic sputum smear images using deep learning methods(PWN-Polish Scientific Publishers bbe@ibib.waw.pl, 2018) Panicker, R.O.; Kalmady, K.S.; Rajan, J.; Sabu, M.K.An automatic method for the detection of Tuberculosis (TB) bacilli from microscopic sputum smear images is presented in this paper. According to WHO, TB is the ninth leading cause of death all over the world. There are various techniques to diagnose TB, of which conventional microscopic sputum smear examination is considered to be the gold standard. However, the aforementioned method of diagnosis is time intensive and error prone, even in experienced hands. The proposed method performs detection of TB, by image binarization and subsequent classification of detected regions using a convolutional neural network. We have evaluated our algorithm using a dataset of 22 sputum smear microscopic images with different backgrounds (high density and low-density images). Experimental results show that the proposed algorithm achieves 97.13% recall, 78.4% precision and 86.76% F-score for the TB detection. The proposed method automatically detects whether the sputum smear images is infected with TB or not. This method will aid clinicians to predict the disease accurately in a short span of time, thereby helping in improving the clinical outcome. © 2018 Nalecz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of SciencesItem An automated mathematical morphology driven algorithm for water body extraction from remotely sensed images(Elsevier B.V., 2018) Rishikeshan, C.A.; Ramesh, H.The detection and extraction of water bodies from satellite imagery is very important and useful for several planning and developmental activities such as shoreline identification, mapping riverbank erosion, watershed extraction and water resource management. Popular techniques for water body extraction like those based on the normalized difference water index (NDWI) require reflectance information in the green and near-infrared (NIR) bands of the light spectrum. Moreover, some commonly used approaches may perform differently according to the spatial resolution of the images. In this regard, mathematical morphological (MM) techniques for image processing have been employed for spatial feature extraction as they preserve edges and shapes. This study proposes a flexible MM driven approach which is very effective for the extraction of water bodies from several satellite images with different spatial resolution. MM provides effective tools for processing image objects based on size and shape and is particularly adapted for water bodies that have typically specific spatial characteristics. In greater details, the proposed extraction algorithm preserves the actual size and shape of the water bodies since it is based on morphological operators based on geodesic reconstruction. Moreover, the choice of the filter size (called structural element (SE) in MM) in the proposed algorithm is done dynamically allowing one to retain the most precise results from different set of inputs images of different spatial resolution and swath. The availability of more than one spectral band of satellite imagery is not necessary for the proposed algorithm as it utilizes only a single band for its computation. This makes it convenient to apply in single band imageries obtained from satellites such as Cartosat thereby making the proposed approach effective over commonly used methods. The accuracy assessment was carried out and compared with the maximum likelihood (ML) classifier and methods based on spectral indices. In all the five test datasets, extraction accuracy of the proposed MM approach was significantly higher than that of spectral indices and ML methods. The results drawn from visual and qualitative assessments indicated its capability and efficiency in water body extraction from different satellite images. © 2018 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)Item Automatic detection and localization of Focal Cortical Dysplasia lesions in MRI using fully convolutional neural network(Elsevier Ltd, 2019) Bijay Dev, K.M.; Pawan, P.S.; Niyas, S.; Vinayagamani, S.; Kesavadas, C.; Rajan, J.Focal cortical dysplasia (FCD) is the leading cause of drug-resistant epilepsy in both children and adults. At present, the only therapeutic approach in patients with drug-resistant epilepsy is surgery. Hence, the quantification of FCD via non-invasive imaging techniques helps physicians to decide on surgical interventions. The properties like non-invasiveness and capability to produce high-resolution images makes magnetic resonance imaging an ideal tool for detecting the FCD to an extent. The FCD lesions vary in size, shape, and location for different patients and make the manual detection time consuming and sensitive to the experience of the observer. Automatic segmentation of FCD lesions is challenging due to the difference in signal strength in images acquired with different machines, noise, and other kinds of distortions such as motion artifacts. Most of the methods proposed in the literature use conventional machine learning and image processing techniques in which their accuracy relies on the trained features. Hence, feature extraction should be done more precisely which requires human expertise. The ability to learn the appropriate features/representations from the training data without any human interventions makes the convolutional neural network (CNN) the suitable method for addressing these drawbacks. As far as we are aware, this work is the first one to use a CNN based model to solve the aforementioned problem using only MRI FLAIR images. We customized the popular U-Net architecture and trained the proposed model from scratch (using MRI images acquired with 1.5T and 3T scanners). FCD detection rate (recall) of the proposed model is 82.5 (33/40 patients detected correctly). © 2019