A5006786315- Machine Learning in Healthcare
- Topic Modeling
- Biomedical Text Mining and Ontologies
- EEG and Brain-Computer Interfaces
- Context-Aware Activity Recognition Systems
- Explainable Artificial Intelligence (XAI)
- Meta-analysis and systematic reviews
- Time Series Analysis and Forecasting
- Obstructive Sleep Apnea Research
- Natural Language Processing Techniques
- Amyotrophic Lateral Sclerosis Research
- Sleep and Wakefulness Research
- Tuberculosis Research and Epidemiology
- Neurogenetic and Muscular Disorders Research
- Hand Gesture Recognition Systems
- Functional Brain Connectivity Studies
- Text Readability and Simplification
- Human Pose and Action Recognition
- Digital Imaging for Blood Diseases
- Speech and Audio Processing
- Video Analysis and Summarization
- Music and Audio Processing
- Music Technology and Sound Studies
- Genomics and Rare Diseases
- Cholinesterase and Neurodegenerative Diseases
Georgia Institute of Technology
2019-2025
Emory University
2024-2025
Bangladesh University of Engineering and Technology
2017-2018
Amyotrophic lateral sclerosis (ALS) has an interactive, multifactorial etiology that makes treatment success elusive. This study evaluates how regulatory dynamics impact disease progression and treatment. Computational models of wild-type (WT) transgenic SOD1-G93A mouse physiology were built using the first-principles-based first-order feedback framework dynamic meta-analysis with parameter optimization. Two in silico developed: a WT model to simulate normal homeostasis ALS pathology their...
Meta-analysis of randomized clinical trials (RCTs) plays a crucial role in evidence-based medicine but can be labor-intensive and error-prone. This study explores the use large language models to enhance efficiency aggregating results from at scale. We perform detailed comparison performance these zero-shot prompt-based information extraction diverse set RCTs traditional manual annotation methods. analyze for two different meta-analyses aimed drug repurposing cancer therapy...
Background: Datasets on rare diseases, like pediatric acute myeloid leukemia (AML) and lymphoblastic (ALL), have small sample sizes that hinder machine learning (ML). The objective was to develop an interpretable ML framework elucidate actionable insights from tabular disease datasets. Methods: comprehensive employed optimized data imputation sampling, supervised unsupervised learning, literature-based discovery (LBD). deployed assess treatment-related infection in AML ALL. Results: An...
This work presents SeizFt-a novel seizure detection framework that utilizes machine learning to automatically detect seizures using wearable SensorDot EEG data. Inspired by interpretable sleep staging, our approach employs a unique combination of data augmentation, meaningful feature extraction, and an ensemble decision trees improve resilience variations in increase the capacity generalize unseen Fourier Transform (FT) Surrogates were utilized sample size class balance between labeled...
This work introduces TrialSieve, a novel framework for biomedical information extraction that enhances clinical meta-analysis and drug repurposing. By extending traditional PICO (Patient, Intervention, Comparison, Outcome) methodologies, TrialSieve incorporates hierarchical, treatment group-based graphs, enabling more comprehensive quantitative comparisons of outcomes. was used to annotate 1609 PubMed abstracts, 170,557 annotations, 52,638 final spans, incorporating 20 unique annotation...
Clinical Cohort Studies (CCS), such as randomized clinical trials, are a great source of documented research. Ideally, expert inspects these articles for exploratory analysis ranging from drug discovery evaluating the efficacy existing drugs in tackling emerging diseases to first test newly developed drugs. However, more than 100 published daily on single prevalent disease like COVID-19 PubMed. As result, it can take days physician find and extract relevant information. Can we develop system...
Seizure detection using machine learning is a critical problem for the timely intervention and management of epilepsy. We propose SeizFt, robust seizure framework EEG from wearable device. It uses features paired with an ensemble trees, thus enabling further interpretation model's results. The efficacy underlying augmentation class-balancing strategy also demonstrated. This study was performed Detection Challenge 2023, ICASSP Grand Challenge.
Beat tracking from music signals has significant importance in multimedia information retrieval systems, especially cover song detection. A predictive real-time beat system can also be used to assist musicians performing live. In this paper we present a algorithm, fast enough implemented on an embedded system. The onset of note is detected using maximum filter approach that suppresses the effect vibrato. Beats are predicted second advance causal variant Dynamic Programming. We have employed...
The accuracy of recent deep learning based clinical decision support systems is promising. However, lack model interpretability remains an obstacle to widespread adoption artificial intelligence in healthcare. Using sleep as a case study, we propose generalizable method combine with high derived from black-box learning.
Sleep staging is a crucial task for diagnosing sleep disorders. It tedious and complex as it can take trained expert several hours to annotate just one patient's polysomnogram (PSG) from single night. Although deep learning models have demonstrated state-of-the-art performance in automating staging, interpretability which defines other desiderata, has largely remained unexplored. In this study, we propose via Prototypes Expert Rules (SLEEPER), combines with defined rules using prototype...
This work presents a new, original document classification dataset, BioSift, to expedite the initial selection and labeling of studies for drug repurposing. The dataset consists 10,000 human-annotated abstracts from scientific articles in PubMed. Each abstract is labeled with up eight attributes necessary perform meta-analysis utilizing popular patient-intervention-comparator-outcome (PICO) method: has human subjects, clinical trial/cohort, population size, target disease, study drug,...
In this paper, we propound a command processing mechanism for an autonomous arm manipulator using real-time speech and images. We propose novel two-stage recognition algorithm two-fold Dynamic Time Warping in each stage. Real-time wake-up word is followed by offline k-means. Since high precision paramount any control system activation mechanism, restrictive threshold set to gain of 1. This alleviates the problem accidental triggering system. Object classification performed matching features...
Lay summarization aims to simplify complex scientific information for non-expert audiences. This paper investigates the trade-off between readability and relevance in lay of long biomedical documents. We introduce a two-stage framework that attains best metrics first subtask BioLaySumm 2023, with 8.924 FleschKincaid Grade Level 9.188 DaleChall Readability Score. However, this comes at cost reduced factuality, emphasizing inherent challenges balancing content preservation summarization. The...
Recent advances in deep learning have led to the development of models approaching human level accuracy. However, healthcare remains an area lacking widespread adoption. The safety-critical nature results a natural reticence put these black-box into practice. This paper explores interpretable methods for clinical decision support system called sleep staging, essential step diagnosing disorders. Clinical staging is arduous process requiring manual annotation each 30s using physiological...
Clinical Cohort Studies (CCS), such as randomized clinical trials, are a great source of documented research. Ideally, expert inspects these articles for exploratory analysis ranging from drug discovery evaluating the efficacy existing drugs in tackling emerging diseases to first test newly developed drugs. However, more than 100 published daily on single prevalent disease like COVID-19 PubMed. As result, it can take days physician find and extract relevant information. Can we develop system...