Search

Abstract Chest Radiography is a non-invasive imaging modality for diagnosing and managing chronic lung disorders, encompassing conditions such as pneumonia, tuberculosis, and COVID-19. While it is crucial for disease localization and severity assessment, existing computer-aided diagnosis (CAD) systems primarily focus on classification tasks, often overlooking these aspects. Additionally, prevalent approaches rely on class activation or saliency maps, providing only a rough localization. This research endeavors to address these limitations by proposing a comprehensive multi-stage framework. Initially, the framework identifies relevant lung areas by filtering out extraneous regions. Subsequently, an advanced fuzzy-based ensemble approach is employed to categorize images into specific classes. In the final stage, the framework identifies infected areas and quantifies the extent of infection in COVID-19 cases, assigning severity scores ranging from 0 to 3 based on the infection’s severity. Specifically, COVID-19 images are classified into distinct severity levels, such as mild, moderate, severe, and critical, determined by the modified RALE scoring system. The study utilizes publicly available datasets, surpassing previous state-of-the-art works. Incorporating lung segmentation into the proposed ensemble-based classification approach enhances the overall classification process. This solution can be a valuable alternative for clinicians and radiologists, serving as a secondary reader for chest X-rays, reducing reporting turnaround times, aiding clinical decision-making, and alleviating the workload on hospital staff.

In traditional machine learning, the training and testing data are assumed to come from the same independent and identical distributions. This assumption, however, does not hold up in real-world applications, as differences between the training and testing data may have different distributions. Domain adaptation has emerged as a solution that enables the transfer of knowledge between domains with distinct distributions. In this paper, we primarily utilize domain adaptation in the context of visual recognition tasks despite its growing application in diverse domains. Earlier studies have mainly aimed at minimizing the differences in global distributions between the domains and failed to capture the local, pertinent features crucial for domain alignment. Furthermore, models struggle to perform well and generalize to target data when outliers or noise exist in the datasets. This work addresses these problems and provides unique strategies for unsupervised domain adaptation using RDAOT (Robust Deep Adaptation via Optimal Transport). To capture local information by utilizing LMMD (Local Maximum Mean Discrepancy) to minimize the divergence of the feature distributions between the domains. We examine label noise robustness in the source domain and ROT (Robust Optimal Transport) loss to preserve robustness in domain adaptation, which lessens the cost of transporting source distributions to the target distributions. The significance of our presented technique was assessed through extensive experiments on six different visual recognition domain adaptation datasets. The results demonstrate that our method outperforms the current state-of-the-art techniques, indicating superior performance. Our approach was evaluated against several baselines, and the results significantly improved average accuracy across various datasets. Specifically, the average accuracy improved from on the OfficeCaltech10 (91.8% to 96.85%), OfficeHome (67.7% to 68.10%), Office31 (88.17% to 88.92%), IMAGECLEF-DA (87.9% to 90.24%), PACS (69.08% to 85.72%), and VisDA-2017 (80.2 % to 89.43%) datasets, respectively.

Cervical cancer is a severe and pervasive disease that poses a significant health threat to women globally. The Pap smear test is an efficient and effective method for detecting cervical cancer in its early stages. However, manual screening is labor-intensive and requires expert cytologists, leading to potential delays and inconsistencies in diagnosis. Deep Learning-based Computer-Aided Diagnosis (CAD) has shown significant results and can ease the problem of manual screening. However, one single model is sometimes insufficient to capture the complex data pattern for accurate disease prediction. In this work, we develop an end-to-end architecture utilizing three pre-trained models for the initial cervical cancer prediction. To aggregate the outcomes of these models, we propose a novel fuzzy rank-based ensemble considering two non-linear functions for the final level prediction. Unlike simple fusion techniques, the proposed architecture provides the final predictions on the test samples by considering the base classifier’s confidence in the predictions. To further enhance the classification capabilities of these models, we integrate advanced augmentation techniques such as CutOut, MixUp, and CutMix. The proposed model is evaluated on two benchmark datasets, SIPaKMeD and Mendeley LBC, using a 5-fold cross-validation approach. On the SIPaKMeD dataset, the proposed ensemble architecture achieves a classification accuracy of 97.18% and an F1 score of 97.16%. On the Mendeley LBC dataset, the accuracy reaches 99.22% with an F1 score of 99.19%. Experimental results demonstrate the proposed architecture’s effectiveness and potential in cervical Pap smear image classification. This could aid medical professionals in making more informed treatment decisions, improving overall effectiveness in the testing process.

In the domain of password security, research has made significant progress in handling different kinds of threats which require human intelligence factor to fix the vulnerabilities. In spite of having strong theoretical establishments, most of these defense mechanisms cannot be used in practice as humans have limitations in processing complex information. The little bit of good news is that very few research proposals in this field have shown the promises to be deployable in practice. This paper focuses on such one method - proposed by Roth et al. back in 2004, which provides adequate user-friendliness to enter Personal Identification Number (PIN) securely in the presence of human shoulder surfers. Surprisingly, the background algorithm of this method for validating users' responses runs in linear time on a search space of cardinality 5 and hence, the validation process does not put much load on the authenticating device. Therefore, such human identification protocol can also be integrated into the IoT infrastructure for conducting a more secured login from the client-side. Having such advantages, though remained secure for almost ten years after its release in 2004, recently, few proposals revealed some serious vulnerable aspects of the Roth et al.'s proposal. In this paper, we have taken an attempt to save this user-friendly form of authentication. Firstly, we have made a critical discussion on the importance of the targeted PIN entry method in the domain of usable security and then given a brief overview of the identified limitations of this protocol. Followed by this, a few initiatives have been taken to fix the identified vulnerabilities of Roth et al.'s proposal by revising its working principle, while the login procedure and the usability standard of this method stay unaffected.