Garima Shukla

Garima Shukla

Auto-generated profile, based on publicly available information

Garima Shukla

Amity University

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 February 2025
Verified
Book Chapter
Closed access
SSShipra Saraswat
Shipra Saraswat
ARAnupama Rajesh
Anupama Rajesh
GSGurinder Singh
Gurinder Singh
KTKiran Trehan
Kiran Trehan
VPVedant Patil
Vedant Patil
SSSofia Singh
Sofia Singh
During the COVID-19 pandemic authorities reported around 1500 missing patients/ tested positive in Delhi-NCR who were difficult to trace down. In times of calamities, natural disasters and pandemics like situations hospitals and medical centres are over occupied by patients and management, doctors and families have a hard time locating their loved ones in multiple hospitals. The human face is a unique biometric system that can determine the age, gender, mood, of an individual and even identity for verification purposes. Along with the power of deep learning and artificial intelligence, applications of machine learning is patient recognition. A state-of-the-art patient recognition is a combination of multiple machine learning and deep learning algorithms for face detection, facial recognition, facial feature extraction and finally validation. Trained based on a database, the software can provide us with information pointing out the identity of missing patients.

19 March 2024 in Scientific Reports
Verified
Journal Article
Open access
PDF file
GSGarima Shukla
Garima Shukla
SSSofia Singh
Sofia Singh
CDChetan Dhule
Chetan Dhule
RARahul Agrawal
Rahul Agrawal
SSShipra Saraswat
Shipra Saraswat
BSBen Othman Sofiene
Ben Othman Sofiene
Nanoparticulate systems have the prospect of accounting for a new making of drug delivery systems. Nanotechnology is manifested to traverse the hurdle of both physical and biological sciences by implementing nanostructures indistinct fields of science, particularly in nano-based drug delivery. The low delivery efficiency of nanoparticles is a critical obstacle in the field of tumor diagnosis. Several nano-based drug delivery studies are focused on for tumor diagnosis. But, the nano-based drug delivery efficiency was not increased for tumor diagnosis. This work proposes a method called point biserial correlation symbiotic organism search nanoengineering-based drug delivery (PBC-SOSN). The objective and aim of the PBC-SOSN method is to achieve higher drug delivery efficiency and lesser drug delivery time for tumor diagnosis. The contribution of the PBC-SOSN is to optimized nanonengineering-based drug delivery with higher r drug delivery detection rate and smaller drug delivery error detection rate. Initially, raw data acquired from the nano-tumor dataset, and nano-drugs for glioblastoma dataset, overhead improved preprocessed samples are evolved using nano variational model decomposition-based preprocessing. After that, the preprocessed samples as input are subjected to variance analysis and point biserial correlation-based feature selection model. Finally, the preprocessed samples and features selected are subjected to symbiotic organism search nanoengineering (SOSN) to corroborate the objective. Based on these findings, point biserial correlation-based feature selection and a symbiotic organism search nanoengineering were tested for their modeling performance with a nano-tumor dataset and nano-drugs for glioblastoma dataset, finding the latter the better algorithm. Incorporated into the method is the potential to adjust the drug delivery detection rate and drug delivery error detection rate of the learned method based on selected features determined by nano variational model decomposition for efficient drug delivery.

9 February 2026 in Engineering, Technology & Applied Science Research
Verified
Journal Article
Open access
GSGarima Shukla
Garima Shukla
VAVanshaj Awasthi
Vanshaj Awasthi
PDPrashant Dubey
Prashant Dubey
SNSakshi Nipane
Sakshi Nipane
SRSampurna Roy
Sampurna Roy
SYSumendra Yogarayan
Sumendra Yogarayan
The subtle transition from healthy cognition to very mild dementia often leaves even experienced clinicians uncertain, highlighting the need for computational frameworks that integrate heterogeneous data sources while providing clinically meaningful explanations. Existing Artificial Intelligence (AI) approaches frequently reduce Alzheimer's Disease (AD) prediction to binary classification, rely on a single modality, or lack interpretability, thereby limiting their translational impact. This study introduces a multimodal Deep Learning (DL) framework designed for fine-grained, four-stage staging of AD across Non-Demented, Very Mild, Mild, and Moderate states. The framework integrates structural Magnetic Resonance Imaging (MRI) and structured clinical biomarkers within a unified architecture, employing pre-trained convolutional and transformer-based networks for imaging alongside gradient-boosted decision trees for tabular features such as demographics, cognitive scores, and laboratory measures. To promote transparency and clinical trust, the framework incorporates complementary interpretability strategies: Gradient-weighted Class Activation Mapping (Grad-CAM) to identify discriminative neuroanatomical regions and attention mechanisms to highlight influential clinical variables. Evaluation was conducted on the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort, with diagnostic labels determined at the subject level using clinician consensus informed by Clinical Dementia Rating (CDR) and Mini-Mental State Examination (MMSE) scores. The dataset comprised 229 Non-Demented, 398 Very Mild, 192 Mild, and 176 Moderate cases. GhostNet achieved an F1-score of 0.888 for Non-Demented and 0.847 for Mild Dementia, whereas Very Mild remained the most challenging stage (best F1-score ≤ 0.628). On structured clinical features, Extreme Gradient Boosting (XGBoost) attained an accuracy of 95.35%. Despite reduced performance for intermediate stages due to class imbalance and overlapping phenotypes, model training remained stable with minimal overfitting. This work provides one of the first interpretable multimodal frameworks for four-stage AD staging, advancing beyond conventional binary models and demonstrating the value of integrating complementary imaging and clinical modalities. By combining robust diagnostic accuracy with transparent, clinician-facing explanations, the approach offers a scalable and trustworthy pathway toward AI-enabled dementia staging, with particular promise for deployment in resource-limited healthcare systems.


26 October 2025
Verified
Book Chapter
Closed access
PDF file
GSGarima Shukla
Garima Shukla
SBSoham Yogesh Bagul
Soham Yogesh Bagul
SDSiddharth Ganesh Devar
Siddharth Ganesh Devar
ACAmandeep Singh Channa
Amandeep Singh Channa
SSShubham Kumar Suran
Shubham Kumar Suran
SSSofia Singh
Sofia Singh

10 November 2025
Verified
Book Chapter
Open access
GSGarima Shukla
Garima Shukla
YCYash Chouksey
Yash Chouksey
AKAnmol Koundal
Anmol Koundal
SBSuraj Bhoumick
Suraj Bhoumick
DTDevansh Tanwar
Devansh Tanwar
MGMonali Gulhane
Monali Gulhane
The growing need for precise and up-to-the-minute analysis of sporting events has propelled the development of both computer vision and machine learning. This work addresses the need for requisite efficiency by proposing a proficient AI-based football analysis system that employs YOLO, OpenCV and DeepSORT to detect and analyse players and the ball in both live and recorded matches. With the ability to perform real time object identification and motion tracking, the system aids in analysis by offering better understanding of the players positions, ball ownership and the game in general. K -Means Clustering is utilized for team classification and subsequent player tracking across frames whereby players are recognized through segmentation of jersey colors. It is trained on datasets such as SoccerNet and FIFA 2023, supplemented with custom annotations and augmentations via platforms like Roboflow. Availability of robust datasets means that the model is well suited to make real world, real time accurate decisions in the face of multiple complex variables which include high density of players on the field, poor lighting and fast movements. Preliminary tests indicate high accuracy in player and ball detection in standard game conditions, with the system capable of real-time performance on standard hardware. Some problems like occlusion management, high density of people (penalty areas) and small objects tracking (like balls) are solved by the usage of appearance features in DeepSORT based re-identification features and YOLO’s speed and efficiency. Non-max suppression reduces overlapping detections and further boosts accuracy. This study also seeks approaches to improve the AI classifiers’ performance when applied to football analytics for the future work. For example, they propose the possibility of integrating AR into fan interaction or predicting the outcome of a match using pose estimation. For live tactical analysis, post-match analysis, or automated highlight generation, this system provides adequate solutions for coaches, analysts, TV broadcasters, and fans. Overall, a football analysis system proposed herein constitutes a novel approach to sports data analysis by incorporating real-time, accurate, competitive and expandable solutions aimed at helping and improving decision-making processes in football. This marks the first steps towards the application of the same technology in a broad range of sport industries based on clever and modern data-based approaches to football and other sports games. There is a clear trend of increasing accuracy and detection efficiency from YOLOv3 to YOLOv8, making YOLOv8 the most suitable for high-performance, real-time football analytics tasks. YOLOv8 outperforms all versions with the highest F1 Score (97.20%), Recall (96.80%), and Precision (97.50%), ensuring superior detection accuracy and minimal false positives.