Abstract

Deep learning based methods have shown great promise in achieving accurate automatic detection of Coronavirus Disease (COVID) - 19 from Chest X-Ray ( CXR) images.However, incorporating explainability in these so- lutions remains relatively less explored. We present a hi- erarchical classification approach for separating normal, non- COVID pneumonia (NCP ) and COVID cases using CXR im- ages. We demonstrate that the proposed method achieves clinically consistent explainations. We achieve this using a novel multi-scale attention architecture called Multi-scale Attention Residual Learning (MARL ) and a new loss function based on conicity for training the proposed architecture. The proposed classification strategy has two stages. The first stage uses a model derived from DenseNet to sepa- rate pneumonia cases from normal cases while the second stage uses the MARL architecture to discriminate between COVID and NCP cases. With a five-fold cross validation the proposed method achieves 93%, 96.28%, and 84.51% accu- racy respectively over three large, public datasets for nor- mal vs. NCP vs. COVID classification. This is competitive to the state-of-the-art methods. We also provide explanations in the form of GradCAM attributions, which are well aligned with expert annotations. The attributions are also seen to clearly indicate that MARL deems the peripheral regions of the lungs to be more important in the case of COVID cases while central regions are seen as more important in NCP cases. This observation matches the criteria described by radiologists in clinical literature, thereby attesting to the utility of the derived explanations.

Abstract

Segmentation and analysis of sub-cortical structures is of interest in diagnosing some neurological diseases. Segmentation is a challenging task because of brain tissue ambiguity and data scarcity. Deep learning (DL) solutions are widely used for this purpose by considering the problem as a semantic segmentation of the brain. In general, DL approaches exhibit a bias towards larger structures when training is done on the whole brain. We propose a method to address this problem wherein a pre-training step is used to learn tissue characteristics and a rough ROI extraction step aids focusing on local context. We use a Residual U-net for demonstrating the proposed method. Experiments on the IBSR and MICCAI datasets show that our proposed solution leads to an improvement in segmentation performance in general with medium and small size structures benefiting significantly. The performance with the proposed method is also marginally better than a more complex, state of art sub-cortical structure segmentation method. A strength of the proposed method is that it can also be applied as a modification to any existing segmentation solution.

Abstract

Deep neural networks suffer from Catastrophic Forgetting (CF) on old tasks when they are trained to learn new tasks sequentially, since the parameters of the model will change to optimize on the new class. The problem of alleviating CF is of interest to Computer aided diagnostic (CAD) sys- tems community to facilitate class incremental learning (IL): learn new classes as and when new data/annotations are made available and old data is no longer accessible. However, IL has not been explored much in CAD development. We pro- pose a novel approach that ensures that a model remembers the causal factor behind the decisions on the old classes, while incrementally learning new classes. We introduce a common auxiliary task during the course of incremental training, whose hidden representations are shared across all the classification heads. Since the hidden representation is no longer task-specific, it leads to a significant reduction in CF. We demonstrate our approach by incrementally learning 5 different tasks on Chest-Xrays and compare the results with the state-of-the-art regularization methods. Our approach performs consistently well in reducing CF in all the tasks with almost zero CF in most of the cases unlike standard regularisation-based approaches.

Abstract

Due to the increasing number of Internet of Things (IoT) devices, a large amount of data is being generated. However, factors such as hardware malfunctions, network failures, or cyber-attacks affect data quality and result in inaccurate data generation. Therefore, to facilitate the data usage, we propose a novel 5D-IoT framework for heterogeneous IoT systems that provides uniform data quality assessment with meaningful data descriptions. Based on the quality assessment result, a data consumer can directly access data from any IoT source, which ultimately speeds up the analysis process and helps gain important insights in less time. The framework relies on semantic descriptions of sensor observations and SHACL shapes assessing the quality of such data. Evaluations carried out on real-time data show the added value of such a framework.

Abstract

nd optimizing an electromechanical system, instinctively affecting its performance. In this study, design optimization, finite element analysis, and experimental evaluation of capacitive pressure sensors were conducted. The air pressure sensing application was demonstrated to characterize different sensors, which include a combination of multiple rectangular cantilevers and diaphragms (square and circular-shaped). After the design improvement, we found that the square and circular diaphragms each with two trapezoidal cantilevers exhibited highest sensitivity to air pressure monitoring among the different investigated designs which combine the square and circular diaphragms with cantilevers. These designs were then selected for further analysis for acoustic pressure monitoring. The sensors were fabricated using the do- it-yourself technique with household materials such as post-it paper, posted tape, and foil. Our approach offers an alternative to the conventional cleanroom fabrication technique and uses easily available materials to fabricate affordable sensors. Therefore, this is the first step toward the development of democratized and sustainable electronic devices that are affordable and available to everyone on the internet.

Abstract

This paper presents a fast and efficient optimization engine with multi-directional, multi-objective algorithms based on a robust transistor sizing approach to improve digital circuit performance. However, such optimization processes are highly simulator-dependent and computationally expensive tasks. Therefore, we propose developing machine learning-based reliable models considering process and operating variations to speed up the optimization procedure by running them on developed Residual Neural Network (ResNN) models instead of running expensive circuit simulations. Results on 22nm Metal Gate HighK digital cells show a reduction in delay and leakage up to 36.7% and 18.8%, respectively improving computational efficiency by several orders.

Abstract

Information Security in the cyber world is a major cause for concern, with significant increase in the number of attack surfaces. Existing information on vulnerabilities, attacks, controls, and advisories available on web provides an opportunity to represent knowledge and perform security analytics to mitigate some of the concerns. Representing security knowledge in the form of ontology facilitates anomaly detection, threat intelligence, reasoning and relevance attribution of attacks, and many more. This necessitates dynamic and automated enrichment of information security ontologies. However, existing ontology enrichment algorithms based on natural language processing and ML models have issues with contextual extraction of concepts in words, phrases and sentences. This motivates the need for sequential Deep Learning architectures that traverse through dependency paths in text and extract embedded vulnerabilities, threats, controls, products and other security related concepts and instances from learned path representations. In the proposed approach, Bidirectional LSTMs trained on a large DBpedia dataset and Wikipedia corpus of 2.8 GB along with Universal Sentence Encoder is deployed to enrich ISO 27001 [1] based information security ontology. The model is trained and tested on an high performance computing (HPC) environment to handle Wiki text dimensionality. The approach yielded a test accuracy of over 80% when tested with knocked out concepts from ontology and web page instances to validate the robustness. Index Terms—Ontology Enrichment, Information Security, Bidirectional LSTM, Universal Sentence Encoder

Abstract

Agricultural industry is one of the most vital industries that has a major contribution to the economy due to its share in the Gross Domestic Product (GDP) and as a source of employment. The past few decades have seen immense change in the operation of agricultural sector with the introduction of precision farming in conjunction with Internet of Things (IoT). The application of such advancements is highly based on exchange of messages between various devices in the farming. This paper aims to study the security scenarios applicable in husbandry through the analysis of possible attacks and threats. The testbeds available for agriculture based on IoT have been studied. An architecture for smart farming is proposed which is independent of the underlying technologies that may be used and the requirements of security have been laid out based on the proposed architecture. A literature survey of security protocols …

Abstract

Quadrotors have become extremely popular in various application scenarios which include disaster response, maintenance etc. A crucial challenge in terms of control design surfaces when a quadrotor has to operate under space constraints (e.g. pipeline inspection from inside, payload delivery at a precise location) under parametric perturbations and environmental disturbances (e.g. wind, gust). To deal with such scenarios, a suitable controller should simultaneously ensure tracking accuracy within predefined bounds to avoid constraint violation and tackle uncertainties. However, the state-of-the-art designs are either inapplicable for quadrotor system which is under-actuated in nature, or are unable to tackle system parametric uncertainties while being under full state (i.e. position, attitude angle, linear velocity and angular velocity) constraints. To address such issues in literature, a robust controller is introduced in …

Abstract

Table-top Rearrangement and Planning is a challenging problem that relies heavily on an excellent perception stack. The perception stack involves observing and registering the 3D scene on the table, detecting what objects are on the table, and how to manipulate them. Consequently, it greatly influences the system’s task-planning and motion-planning stacks that follow. We present a comprehensive overview and discuss different challenges associated with the perception module. This work is a result of our extensive involvement in the ICRA 2022 Open Cloud Robot Table Organization Challenge, in which we currently stand first on the leaderboard (as of 24th of April 2022, the final week of the challenge).