Ali Raza Ansari

The evolving healthcare domain necessitates an upgrade through digitization, integrating patient data, and advanced medical results. In the last couple of decades, advances in information and storage technologies in healthcare have produced vast amounts of data. The remarkable increases in data volumes, along with the enticing prospects and potential inherent in data analysis, have contributed to the concept of Big Data. There is a pressing need within the research community to analyze these large volumes of Big Data. To address this challenge, Big Data Analytics (BDA), the systematic process of examining large and complex datasets to uncover hidden patterns, correlations, and insights for informed decision-making, has emerged. It employs various methodologies and techniques to enable informed decision-making. This study delves into using Machine Learning (ML) in big data environments, explicitly utilizing the MLib library in Apache Spark to derive meaningful insights from diabetic healthcare dataset. The CDC’s Behavioral Risk Factor Surveillance System (BRFSS) was used to empirically demonstrate the advantages of integrating BDA with ML for medical decision-making in Big Data environments. The research finding highlighted the superior performance of Logistic Regression (LR) models compared to other models like Naive Bayes (NB), providing valuable insights for healthcare applications.

Skin cancer is a common but preventable type of cancer. Observing the rising trend of cases being reported every year, accurate timely diagnose can stop the spread of disease in the body. Manual diagnosis requires clinical expertise and causes delays whereas automatic diagnosis can bring accuracy and save time. This research introduces an enhanced approach, Detectron2, a state-of-the-art modular framework for deep learning-based segmentation tasks that enables precise instance level segmentation, therefore, distinguishing overlapping lesions and accurately identifying individual regions of interests. This customized approach offers the ability to handle complex and irregular lesion structures, robustness against noise elements like hairlines and ink projections without any pre-processing. This paper utilized manually annotated International Skin Imaging Collaboration (ISIC) archive and PH ² datasets for training and testing. Detectron2 establishes itself as a promising tool by achieving Intersection over Union (IoU) scores of 0.917 for the 2016 ISIC dataset, 0.869 for the 2017 ISIC dataset, 0.942 for the 2018 ISIC dataset, and 0.915 for the PH ² dataset.

The present work re-evaluates the evaluation method for text summarization tasks. Two state-of-the-art assessment measures e.g., Recall-Oriented Understudy for Gisting Evaluation (ROUGE) and Bilingual Evaluation Understudy (BLEU) are discussed along with their limitations before presenting a novel evaluation metric. The evaluation scores are significantly different because of the length and vocabulary of the sentences, this suggests that the primary restriction is its inability to preserve the semantics and meaning of the sentences and consistent weight distribution over the whole sentence. To address this, the present work organizes the phrases into six different groups and to evaluate “text summarization” problems, a new hybrid approach (HybridEval) is proposed. Our approach uses a weighted sum of cosine scores from InferSent’s SentEval algorithms combined with original scores, achieving high accuracy. HybridEval outperforms existing state-of-the-art models by 10-15% in evaluation scores.

To improve mobility and rehabilitation, precise and adaptive control mechanisms have been developed for lower limb exoskeletons. Brain-computer interface (BCI) provides advanced and intuitive control of assistive and rehabilitation exoskeletons to aid the user. Functional near-infrared spectroscopy (fNIRS) is a non-invasive, and portable brain imaging modality, gained momentum in rehabilitation studies in the last decade. This study provides a novel approach to control a lower limb exoskeleton with enhanced classification accuracy using fNIRS-based BCI, the k-nearest neighbors (kNN) classifier, and optimal feature combination. The brain signals were acquired using fNIRS for walking vs rest for twenty healthy participants, having ten trials for each participant. The statistical measures: mean, peak, variance, skewness, kurtosis, and slope are extracted as features. Optimal feature combination was analyzed and selected for enhanced classification accuracy. kNN was analyzed and selected as an optimal classifier with optimal ‘k’ (number of nearest neighboring data points that the kNN considers while classifying a new data point) using elbow method to improve classification performance. The proposed method achieves an average classification accuracy of 88.19 ± 2.55 %, in offline configuration. In order to control exoskeleton in online settings, simulated online classification was performed using one unknown trial, fed as real-time signal. Sliding window of 2.5 sec is used and achieved average classification accuracy of 97.5%. This research represents a major advancement in user-centric assistive technologies and advances the field of neuro-powered exoskeletons. It also lays the groundwork for future advancements in the integration of neuroimaging, machine learning, and rehabilitation.

In brain-computer interfacing, the SSVEP (steady-state visual evoked potential) method serves to foster collaboration between humans and robots. SSVEP-based detection methods require complex multichannel data acquisition, making them difficult to deploy due to discomfort during extended use and the complexity of the algorithms involved. On the other hand, single-channel setup offers simplicity and ease of use. However, in a single channel, achieving encouraging performance in the SD (subject-dependent) scenario is challenging, and accuracy drops further in the SI (subject-independent) scenario. This requires the development of a generalized approach to improve performance in both scenarios. This study proposes (VMD-DNN) to detect SSVEP in single-channel setups for SD and SI scenarios. The novelty of the proposed method lies in utilizing VMD (Variational Mode Decomposition) as a preprocessor, leveraging harmonic information and Kurtosis of the cross-correlation function to select harmonics from VMD decomposed signal. The preprocessed reconstructed signal uses complex spectrum features as input to the DNN for classification. The results show an average accuracy of 93%, 95.3% in SD and 79%, 92.33% in SI scenarios tested on two publicly available datasets, respectively. The ITR (Information transfer rate) was 67.50 bit/min, 92.31 bit/min for SD, and 46.13 bit/min, 85.94 bit/min for SI for both datasets, respectively. In SD, accuracy is improved by 3.34% and 5%, and ITR by 8.87% and 12.91% over baseline methods for both datasets respectively. The proposed VMD-DNN model is effective, with improved performance and lower computational complexity. The robust single-channel approach makes it user-friendly for human-robot collaboration.

Personality analysis allows the experts to get insights into an individual’s conduct, vulnerabilities, and prospective capabilities. Some common methods employed for personality prediction include text analysis, social media data, facial expressions, and emotional speech extraction. Recently, some studies have utilized the big five model to predict personality traits using non-verbal cues (gaze score, body motion, head motion). However, these studies mostly target only three aspects of the big five mode. None of the studies so far have used non-verbal cues to target all five traits (extraversion, openness, neuroticism, agreeableness, and conscientiousness) of the Big Five model. In this paper, we propose a multi-modal system that predicts all five personality traits of the Big Five model using non-verbal cues (facial expressions, head poses, body poses), 44-item Big Five Inventory (BFI) questionnaire, and expert analysis. The facial expression module utilizes the Face Emotion Recognition Plus (FER+) dataset trained with Convolution Neural Network (CNN) model achieving 95.14% accuracy. Evaluating 16 subjects in verbal interaction with humanoid robot NAO, we combined questionnaire feedback, human-robot interaction data, and expert perspectives to deduce their Big Five traits. Findings reveal 100% accuracy in personality prediction via expert insights and the system, and 75% for the questionnaire-based approach.

This manuscript delves into the examination of the stochastic fractional derivative of Drinfel’d-Sokolov-Wilson equation, a mathematical model applicable in the fields of electromagnetism and fluid mechanics. In our study, the proposed equation is through examined through various viewpoints, encompassing soliton dynamics, bifurcation analysis, chaotic behaviors, and sensitivity analysis. A few dark and bright shaped soliton solutions, including the unperturbed term, are also examined, and the various 2D and 3D solitonic structures are computed using the Tanh-method. It is found that a saddle point bifurcation causes the transition from periodic behavior to quasi-periodic behavior in a sensitive area. Further analysis reveals favorable conditions for the multidimensional bifurcation of dynamic behavioral solutions. Different types of wave solutions are identified in certain solutions by entering numerous values for the parameters, demonstrating the effectiveness and precision of Tanh-methods. A planar dynamical system is then created using the Galilean transformation, with the actual model serving as a starting point. It is observed that a few physical criteria in the discussed equation exhibit more multi-stable properties, as many multi-stability structures are employed by some individuals. Moreover, sensitivity behavior is employed to examine perturbed dynamical systems across diverse initial conditions. The techniques and findings presented in this paper can be extended to investigate a broader spectrum of nonlinear wave phenomena.

This work studies the behavior of electrical signals in resonant tunneling diodes through the application of the Lonngren wave equation. Utilizing the method of Lie symmetries, we have identified optimal systems and found symmetry reductions; we have also found soliton wave solutions by applying the tanh technique. The bifurcation and Galilean transformation are found to determine the model implications and convert the system into a planar dynamical system. In this experiment, the equilibrium state, sensitivity, and chaos are investigated and numerical simulations are conducted to show how the frequency and amplitude of alterations affect the system. Furthermore, local conservation rules are demonstrated in more detail to unveil the whole system of movements.