Primary brain tumours make up less than 2% of cancers and statistically occur in around 250,000 people a year globally. Medical resonance imaging (MRI) plays a pivotal role in the diagnosis of brain tumours and advanced imaging techniques can precisely detect brain tumours. On this note, Dr Sudhakar Tummala, Assistant Professor, Department of Electronics and Computer Engineering, has published a paper titled, “Classification of Brain Tumour from Magnetic Resonance Imaging using Vision Transformers Ensembling” in the journal Current Oncology having an impact factor of 3.1. The paper highlights the pioneering breakthrough made in the development of vision transformers (ViT) in enhancing MRI for efficient classification of brain tumours, thus reducing the burden on radiologists.

Abstract of the paper

The automated classification of brain tumours plays an important role in supporting radiologists in decision making. Recently, vision transformer (ViT)-based deep neural network architectures have gained attention in the computer vision research domain owing to the tremendous success of transformer models in natural language processing. Hence, in this study, the ability of an ensemble of standard ViT models for the diagnosis of brain tumours from T1-weighted (T1w) magnetic resonance imaging (MRI) is investigated. Pretrained and fine tuned ViT models (B/16, B/32, L/16, and L/32) on ImageNet were adopted for the classification task. A brain tumour dataset from figshare, consisting of 3064 T1w contrast-enhanced (CE) MRI slices with meningiomas, gliomas, and pituitary tumours, was used for the cross-validation and testing of the ensemble ViT model’s ability to perform a three-class classification task. The best individual model was L/32, with an overall test accuracy of 98.2% at 384 × 384 resolution. The ensemble of all four ViT models demonstrated an overall testing accuracy of 98.7% at the same resolution, outperforming individual model’s ability at both resolutions and their ensemble at 224 × 224 resolution. In conclusion, an ensemble of ViT models could be deployed for the computer-aided diagnosis of brain tumours based on T1w CE MRI, leading to radiologist relief.

A brief summary of the research in layperson’s terms

Brain tumours (BTs) are characterised by the abnormal growth of neural and glial cells. BTs causes several medical conditions, including the loss of sensation, hearing and vision problems, headaches, nausea, and seizures. There exist several types of brain tumours, and the most prevalent cases include meningiomas (originate from the membrane surrounding the brain), which are non-cancerous; gliomas (start from glial cells and the spinal cord); and glioblastomas (grow from the brain), which are cancerous. Sometimes, cancer can spread from other parts of the body, which is called brain metastasis. A pituitary tumour is another type of brain tumour that develops in the pituitary gland in the brain, and this gland primarily regulates other glands in the body. Magnetic resonance imaging (MRI) is a versatile imaging method that enables one to noninvasively visualise inside the body, and is in extensive use in the field of neuroimaging.

There exist several structural MRI protocols to visualise inside the brain, but the prime modalities include T1-weighted (T1w), T2-weighted, and T1w contrast-enhanced (CE) MRI. BTs appear with altered pixel intensity contrasts in structural MRI images compared with neighbouring normal tissues, enabling clinical radiologists to diagnose them. Several previous studies have attempted to automatically classify brain tumours using MRI images, starting with traditional machine learning classifiers, such as support vector machines (SVMs), k-nearest-neighbour (kNN), and Random Forest, from hand-crafted features of MRI slices. With the rise of convolutional neural network (CNN) deep learning model architectures since 2012, in addition to emerging advanced computational resources, such as GPUs and TPUs, during the past decade, several methods have been proposed for the classification of brain tumours based on the finetuning of the existing state-of-the-art CNN models, such as AlexNet, VGG16, ResNets, Inception, DenseNets, and Xception, which had already been found to be successful for various computer vision tasks.

Despite the tremendous success of CNNs, they generally have inductive biases, i.e., the translation equivariance of the local receptive field. Due to these inductive biases, CNN models have issues when learning long-range information; moreover, data augmentation is generally required for CNNs to improve their performance due to their dependency on local pixel variations during learning.Therefore, in this work, the ability of pretrained and fine tuned ViT models, both individually and in an ensemble manner, is evaluated for the classification of meningiomas, gliomas, and pituitary tumours from T1w CE MRI at both 224 × 224 and 384 × 384 image resolutions.

Dr Sudhakar Tummala has mentioned the social implications of the research by expounding that the computer-aided diagnosis of brain tumours from T1w CE MRI using an ensemble of fine tuned ViT models can be an alternative to manual diagnoses, thereby reducing the burden on clinical radiologists. He also explains the future prospects of his research, which is to add explainability to the ensemble model predictions and to develop methods for precise contouring of tumour boundaries.

Details of Collaborations

Prof Seifedine Kadry, Department of Applied Data Science, Noroff University College, Kristiansand, Norway.

Dr Syed Ahmad Chan Bukhari, Division of Computer Science, Mathematics and Science, Collins College of Professional Studies, St. John’s University, New York, USA.

The Department of Electronics and Communication Engineering is glad to announce that Mr Garikapati Anith Chowdary, a BTech passed-out student has published a paper in collaboration with Assistant Professor Dr M Durga Prakash. The paper titled Tunnel Field Effect Transistor Design and Analysis for Biosensing Applications was published in the Q2 journal Silicon having an Impact Factor 2.941.

The physical modelling of the tunnel field effect transistor (TFET) is done in this study. The Silvaco TCAD tool is used to design and simulate the TFET structure. The FET device has attracted a lot of attention as the ideal tool for creating biosensors because of its appealing properties such as ultra-sensitivity, selectivity, low cost, and real-time detection capabilities in a sensing point of view.

These devices have a lot of potential as a platform for detecting biomolecules. Short channel effects, specificity, and nano-cavity filling have all been improved in FET-based biosensors. FET-based biosensors are appropriate for label-free applications. Random dopant variations and a thermal budget are seen during the construction of a JLFET. To overcome this problem, the charge-plasma-based concept was established in FETs in this study.

Different metallurgical functions for electrodes were employed in this biosensor to behave as a p-type source and n-type drain. To alleviate the short channel effects, a dual material gate work function for the gate electrode was devised, as well as a double gate architecture. Biomolecules can be neutral or charge-based, and both types of biomolecules can be identified using a proof-of-concept FET-based biosensor. Changes in the drain current (Id) of the device were achieved by varying dielectric values and charges in the cavity region with variable cavity lengths.

Research at the Department of Electronics and Communication Engineering is currently developing defect detection algorithms. Assistant Professor Dr V Udaya Sankar, Professor Dr Yellampalli Siva Sankar, and their BTech student Ms Gayathri Lakshmi have published a paper, A Review of various defects in PCB, in the Journal of Electronic Testing: Theory and Applications with an impact factor of 0.795.

Abstract

Printed Circuit Boards (PCBs) are the building blocks for all electronic products. Fabrication of a PCB involves various mechanical and chemical processes. As obtaining accuracy in the mechanical and chemical processes is very difficult, various defects/faults are formed during PCBs fabrication. These fabrication defects lead to performance degradation of electronic products. This paper describes multiple defects present in PCBs under the Through-hole and SMD categories. To understand the frequency of occurrence and reason for defects in both manual and machine, PCB fabrication data was collected and analysed from April 2017 to July 2020 as a part of industry collaboration.

The research is a review done on the defects present in PCB. Researchers surveyed various papers on PCB defects and their detection. Based on the literature review and information obtained from Efftronics systems Pvt. Ltd, they classified the defects, gave a detailed explanation for each, and provided some analysis of their occurrences.

While doing the literature review, researchers observed that no paper mentioned all the defects that can occur in the case of PCB fabrication. For this reason, they came up with this paper which provides detailed information regarding the defects. Information is also obtained from the industry. Comparing the defects can help focus on the critical defects for future research on defect detection methodology.

The project is done in collaboration with Efftronics Systems Pvt. Ltd. Through the partnership, the company supported sharing images, insights information related to defects and involved in discussions. Also, the company allowed visiting their premises to understand more about PCB defects. Researchers look forward to creating a prototype that detects all the defects mentioned in this paper for a given PCB.

The Department of Electronics and Communication Engineering is delighted to announce that Assistant Professor Dr Anirban Ghosh and Research Scholar Mr Naga Srinivasarao Chilamkurthy have published their article titled “Low-Power Wide-Area Networks: A Broad Overview of its different aspects” in IEEE Access, a Q1 Journal, having an Impact Factor of 3.476. The work was published in collaboration with Dr Om Jee Pandey from the Indian Institute of Technology BHU, Dr Cenkeramaddi Linga Reddy from the University of Agder, and Dr Hong-Ning Dai from Hong Kong Baptist University, Hong Kong.

This is a survey article on Low-power Wide-area networks which provides a detailed description of LPWAN technologies in the context of IoT applications. In this survey article, they review and provide an overarching description of LPWAN in terms of design goals, techniques to improve design objectives, and system architecture. They have also evaluated several existing and non-standardized LPWAN technologies and the market opportunities of LPWAN. With the help of this article, the researchers can choose the best LPWAN technology for their specific applications.

The practical implementation of the article can be found in various social and commercial applications such as smart healthcare, intelligent transportation, climate-smart agriculture, rescue operations, logistics, smart cities, industries, utilities, smart buildings, consumer electronics, security, asset tracking, smart waste management systems, cognitive manufacturing, and Machine-to-Machine (M2M) communications. Their future research plans include working on Wireless Sensor Networks, Low-Power Wide-Area Networks, Small-World Networks, and applying machine learning and reinforcement learning techniques in the context of wireless networks for cyber-physical systems and IoT applications.

Abstract of the Research

Low-power wide-area networks (LPWANs) are gaining popularity in the research community due to their low power consumption, low cost, and wide geographical coverage. LPWAN technologies complement and outperform short-range and traditional cellular wireless technologies in a variety of applications, including smart city development, machine-to-machine (M2M) communications, healthcare, intelligent transportation, industrial applications, climate-smart agriculture, and asset tracking. This review paper discusses the design objectives and the methodologies used by LPWAN to provide extensive coverage for low-power devices. We also explore how the presented LPWAN architecture employs various topologies such as star and mesh. We examine many current and emerging LPWAN technologies, as well as their system architectures and standards, and evaluate their ability to meet each design objective. In addition, the possible coexistence of LPWAN with other technologies, combining the best attributes to provide an optimum solution is also explored and reported in the current overview. Following that, a comparison of various LPWAN technologies is performed, and their market opportunities are also investigated. Furthermore, an analysis of various LPWAN use cases is performed, highlighting their benefits and drawbacks. This aids in the selection of the best LPWAN technology for various applications. Before concluding the work, the open research issues, and challenges in designing LPWAN are presented.

Continue reading →

The Department of Electronics and Communication Engineering is glad to announce that Assistant Professor Dr Sunil Chinnadurai and his research scholar Mr Shaik Rajak have published a paper titled “Energy Efficient MIMO-NOMA aided IoT Network in B5G Communications” in the Q1 journal Computer Networks having an Impact Factor of 5.5. With an intent to accelerate the development of future intelligence wireless systems, the paper proposes an energy-efficient massive multiple-input-multiple-output (MIMO)- non-orthogonal multiple access (NOMA) aided internet of things (IoT) network to support the massive number of distributed users and IoT devices with seamless data transfer and connectivity.

Abstract of the research

Massive MIMO has been identified as a suitable technology to implement the energy efficient IoT network beyond 5G (B5G) communications due to its distinct characteristics with a large number of antennas. However, providing fast data transfer and maintaining hyperconnectivity between the IoT devices in B5G communications will bring the challenge of energy deficiency. Hence, they considered a massive MIMO-NOMA aided IoT network considering imperfect channel state information and practical power consumption at the transmitter. The far users of the base stations are selected to investigate the power consumption and quality of service. Then, they calculated the power consumption which is a non-convex function and non-deterministic polynomial problem. To solve the above problem, fractional programming properties are applied which converted the polynomial problem into the difference of convex function. And then they employed the successive convex approximation technique to represent the non-convex to convex function. Effective iterative-based branches and the reduced bound process are utilized to solve the problem. Numerical results observed that their implemented approach surpasses previous standard algorithms on the basis of convergence, energy efficiency, and user fairness.

Explanation of the research in layman’s terms

• A cost-effective (i.e., energy efficient) maximization problem for the multiple cells NOMA heterogeneous network scheme is explored when meeting the transmission power and data necessity of far users. The singular value uncertainty model (SVUM) is deliberated to add the errors with the transmitted signal. Since it’s a non-convex problem and challenging to solve, they used the properties of fractional programming to convert it into its corresponding mathematical terms. ITS needs higher data rate and seamless connectivity to operate with maximum speed and safety.
• SCA methods are then applied to change the optimisation problem. After that, an effective iterative scheme is employed based on Branch and Reduced Bound (BRB) that resolves the energy-efficient SVUM problem and satisfies the convergence criteria.
• The proposed iterative BRB method enhances user fairness and decreases inter-tier interference (ITI). IRS has been recognised as the key enabling technology to provide the data required by the ITS with less power consumption.
• Energy efficiency achieved by the proposed BRB method is examined with the help of numerical results and found that the proposed algorithm provides better efficacy than the majorisation minimisation (MM) method and the well-known OMA scheme.

Practical implementations of the research

• To provide high data rates to wireless sensors and the internet of things (IoT), future communication systems can ultimately be advanced by implementing NOMA, small cell, and heterogeneous networks (HetNets) along with MIMO.
• An energy-efficient massive MIMO-NOMA aided IoT network to support the massive number of distributed users and IoT devices with seamless data transfer and connectivity between them in B5G communications.

Future research plans

• To explore the energy efficiency of AI-driven IoT networks for applications such as intelligent health care and intelligent vehicular communications.
• MIMO-NOMA with IRS elements to reduce power consumption and improve the connectivity between the users.