A5101916117- Explainable Artificial Intelligence (XAI)
- Ethics and Social Impacts of AI
- Adversarial Robustness in Machine Learning
- Video Surveillance and Tracking Methods
- Photonic and Optical Devices
- Photonic Crystal and Fiber Optics
- AI in cancer detection
- Brain Tumor Detection and Classification
- Medical Image Segmentation Techniques
- Lubricants and Their Additives
- Glycosylation and Glycoproteins Research
- Big Data and Business Intelligence
- Biodiesel Production and Applications
- Advanced Malware Detection Techniques
- Color Science and Applications
- Medical Imaging and Analysis
- Advanced Combustion Engine Technologies
- Digital Imaging for Blood Diseases
- Advanced Vision and Imaging
- Neural Networks and Applications
- Network Security and Intrusion Detection
- Digital Media Forensic Detection
- Video Analysis and Summarization
- Image and Video Quality Assessment
- Tribology and Lubrication Engineering
Indian Institute of Technology Madras
2021-2024
Delhi Technological University
2014-2024
International Institute of Information Technology
2024
Chandigarh University
2024
Qingdao University of Science and Technology
2024
Qingdao University of Technology
2024
Lebanese American University
2024
Maharishi Markandeshwar University, Mullana
2024
Indian Institute of Information Technology, Nagpur
2024
Sardar Vallabhbhai National Institute of Technology Surat
2023
Explainable machine learning offers the potential to provide stakeholders with insights into model behavior by using various methods such as feature importance scores, counterfactual explanations, or influential training data. Yet there is little understanding of how organizations use these in practice. This study explores view and explainability for stakeholder consumption. We find that, currently, majority deployments are not end users affected but rather engineers, who debug itself. There...
The most frequently occurring cancer among Indian women is breast cancer. There a chance of fifty percent for fatality in case as one two diagnosed with die the cases [1]. This paper aims to present comparison largely popular machine learning algorithms and techniques commonly used prediction, namely Random Forest, kNN (k-Nearest-Neighbor) Naïve Bayes. Wisconsin Diagnosis Breast Cancer data set was training compare performance various terms key parameters such accuracy, precision. results...
Concerns within the machine learning community and external pressures from regulators over vulnerabilities of algorithms have spurred on fields explainability, robustness, fairness. Often, issues in fairness are confined to their specific sub-fields few tools exist for model developers use simultaneously build modeling pipelines a transparent, accountable, fair way. This can lead bottleneck developer's side as they must juggle multiple methods evaluate algorithms. In this paper, we present...
Machine learning models are prone to biased decisions due biases in the datasets they trained on. In this paper, we introduce a novel data augmentation technique create fairer dataset for model training that could also lend itself understanding type of bias existing i.e. if arises from lack representation particular group (sampling bias) or it because human reflected labels (prejudice based bias). Given involving protected attribute with privileged and unprivileged group, an "ideal world''...
Abstract A solid‐core (SC) octagonal photonic crystal fiber (O‐PCF) designed for operation in the terahertz regime aims to enhance its performance. The effective material loss, confinement mode area, and nonlinear coefficient have been investigated compared two different media: silica glass Teflon. frequency range from 0.7 3.0 THz has utilized analyze all results using a full‐vectorial finite element method, with COMSOL Multiphysics software employed simulations. With suitable selection of...
Background: Local anesthetic injection is one of the most anxiety- provoking procedure for both children and adult patients in dentistry. A computerized system slow delivery local has been developed as a possible solution to reduce pain related injection. Study design: The present study was conducted evaluate compare perception rates pediatric with traditional methods, objectively subjectively. It randomized controlled hundred aged 8-12 years healthy physical mental state, assessed being...
Cancer cells need as much 40-times more sugar than their normal cell counterparts. This demand is attained by the excessive expression of inimitable transporters on surface cancer cells, driven voracious appetite for carbohydrates. Nanotechnological advances drive research utilizing ligand-directed therapeutics and diverse carbohydrate analogs. The precise delivery these therapeutic cargos not only mitigates toxicity associated with chemotherapy but also reduces grim toll mortality morbidity...
Oral Cancer, a worldwide health concern, highlights the urgent need for accurate and swift detection cure. Current diagnosing strategies primarily involve pathologists analyzing tissue biopsy samples, method that is time-consuming heavily driven by pathologists' experience. To address these drawbacks, this study proposes novel technique incorporates machine vision cancer detection, aiming to enhance diagnostic accuracy. Given intricate nature of histopathological images, we adopt an...
The efficiency optimization methods for natural coagulants are often restricted due to non-scientific trial-and-error approaches. They inaccurate in predicting the complex interactions of jet mixing parameters, coagulant dosage, and environmental conditions. To overcome these obstacles, this research paper proposes advanced hybrid models machine learning enhance flocculation efficiency. We use CatBoost model with NTK learn intricate nonlinear among velocity, time, dose, pH, turbidity. is...
The knowledge base in the field of vaccination research and development has greatly improved. In present era, utilizing a novel vaccine design optimization strategy improves efficiency activity. this regard, drug delivery system produces nanosized, vaccine-loaded carrier molecules, which is called Nanovaccine. advancement nanovaccine improved technology comes under preclinical clinical trials, whereas different routes administration are applied against infectious diseases. high...
Breast cancer is the most prevalent form of and can occur in both men women, although it common among women. The problem with manual pathology examination breast that time-consuming as requires scanning through images tissue under various distinct magnification levels to obtain accurate diagnoses. advancement computer assisted diagnosis system help improve early process using machine learning algorithms. This paper proposes hybrid architecture CLAHE deep convolutional network for...
Fairness and robustness in machine learning are crucial when individuals subject to automated decisions made by models high-stake domains. To promote ethical artificial intelligence, fairness metrics that rely on comparing model error rates across subpopulations have been widely investigated for the detection mitigation of bias. However, measures ability achieve recourse relatively unexplored. In this paper, we present a novel formulation training neural networks considers distance data...
We theoretically examine two designs of single-mode (i) Er-doped tellurite and (ii) undoped photonic crystal fiber (PCF) for generation slow light with tunable features based on stimulated Brillouin scattering. obtained gain up to 91 dB time delay ∼145 ns at maximum allowable pump power ∼775 mW in a 2 m PCF ∼88 ∼154 ∼21 100 fiber. Simulated results clearly indicate that the doped Er enhances comparable can be even reduced length. believe carried out examination simulation have potential...
As a result of boom in technological advancements witnessed over the past few decades, we are racing forward quest contriving more innovative, and efficient means production manufacturing. These innovations assessed on basis their complexity, cost-effectiveness among other parameters. Industry 4.0 gives us base to understand new ways with additive manufacturing (AM) being its pillar. The aerospace sector requires higher strength-to-weight ratio materials as they would be fuel-efficient....