A medical imaging scientist who turns AI research into clinical impact.

Associate Principal at Guidehouse Digital LLC · Contracted Senior Medical Imaging Deep Learning Image Scientist at NIH / National Library of Medicine (NLM) · Ph.D. Information & Communication Engineering, Anna University · 15+ years academic · 9+ years research expertise.

Dr. Sivarama Krishnan Rajaraman
Sivarama Krishnan Rajaraman, Ph.D.
Senior Medical Imaging Deep Learning Image Scientist
NIH / National Library of Medicine
Deep Learning Medical Imaging Vision Transformers Trustworthy AI Multimodal Learning CNNs Generative Modeling
4,026+Citations
28H-Index
47i10-Index
100+Journals Reviewed
15+Years Academic
9+Years Research
4,026+
Citations
100+
Journals Reviewed

Bio

"I build AI tools that can genuinely support clinical decision-making, especially where resources are limited, and I care deeply about making those systems reliable, fair, and useful in the real world."

Dr. Sivarama Krishnan Rajaraman is an Associate Principal at Guidehouse Digital LLC and works with the Division of Intramural Research at the National Library of Medicine, part of the U.S. National Institutes of Health, as a Senior Medical Imaging Deep Learning Image Scientist. He earned his Ph.D. in Information and Communication Engineering from Anna University in India and later completed postdoctoral research at NLM and NIH.

His work sits at the intersection of artificial intelligence and clinical medicine. Over the years, he has focused on problems such as computer-aided TB and COVID-19 screening, mobile malaria diagnostics, generative modeling for data scarcity, and rigorous evaluation for real-world deployment, including calibration, uncertainty, fairness, and domain shift.

His research is closely aligned with the broader push for trustworthy and responsible clinical AI in public health. In practice, that means building systems that are not only accurate, but also transparent, equitable, and dependable when they are used in high-impact settings.

Focus Areas

🫁 TB & COVID-19 Screening

Computer-aided detection and severity quantification from chest radiographs, including pediatric cohorts, dual-projection modeling, lesion segmentation, and uncertainty-aware triage pipelines deployed at scale.

🦟 Malaria Diagnostics Malaria Screener

Mobile and AI-enabled detection of malaria parasites via smartphone-based microscopy, designed for low-resource, field-deployable point-of-care settings with model interpretability.

🧬 Generative Modeling & Data Scarcity

Tackling limited labeled medical data using GANs, latent diffusion models, and multimodal learning with LLM-generated and expert-curated text for robust clinical translation.

🛡️ Trustworthy AI for Clinical Deployment

Systems research in model calibration, uncertainty quantification, inter-reader variability, algorithmic fairness, and cross-domain robustness building defensible clinical AI.

Expertise

Model Architectures
CNNsVision Transformers (ViT)Hybrid ModelsDeep EnsemblesU-Net
Techniques
Generative ModelingLatent DiffusionMultimodal LearningTransfer LearningData AugmentationSelf-Supervised Learning
Trustworthy AI
CalibrationUncertainty QuantificationFairness AuditingDomain AdaptationExplainability / XAI
Tools & Frameworks
PyTorchTensorFlowKerasScikit-learnOpenCVMONAIAWS SageMaker

Citation History

Google Scholar Annual Citations
Sivarama Krishnan Rajaraman · Data current as of April 2026
Open Profile ↗
4,026
All Citations
28
h-index
47
i10-index
657
Top Paper
0160320480640800Citations 2013: 8 citations 2013 2014: 14 citations 2014 2015: 22 citations 2015 2016: 38 citations 2016 2017: 58 citations 2017 2018: 145 citations 1452018 2019: 310 citations 3102019 2020: 560 citations 5602020 2021: 635 citations 6352021 2022: 710 citations 7102022 2023: 740 citations 7402023 2024: 610 citations 6102024 2025: 176 citations 1762025 2013: 8 citations 2014: 14 citations 2015: 22 citations 2016: 38 citations 2017: 58 citations 2018: 145 citations 2019: 310 citations 2020: 560 citations 2021: 635 citations 2022: 710 citations 2023: 740 citations 2024: 610 citations 2025: 176 citations Citations per year Trend line

Annual citation counts sourced from Google Scholar profile (bHPu9eIAAAAJ). Update the annualData array in the script each year to keep the chart current.

Highlighted Work

Pre-trained convolutional neural networks as feature extractors toward improved malaria parasite detection in thin blood smear images
Rajaraman S et al. · PeerJ · doi:10.7717/peerj.4568
Iteratively pruned deep learning ensembles for COVID-19 detection in chest X-rays
Rajaraman S, Siegelman J, Alderson PO et al. · IEEE Access · doi:10.1109/ACCESS.2020.3003810
Visualization and interpretation of CNN predictions in detecting pneumonia in pediatric chest radiographs
Rajaraman S, Candemir S, Kim I et al. · MDPI Applied Sciences · doi:10.3390/app8101715
Performance evaluation of deep neural ensembles toward malaria parasite detection in thin-blood smear images
Rajaraman S, Jaeger S, Antani SK · PeerJ · doi:10.7717/peerj.6977
Modality-specific deep learning model ensembles toward improving TB detection in chest radiographs
Rajaraman S, Antani SK · IEEE Access · doi:10.1109/ACCESS.2020.2971257
Deep learning model calibration for improving performance in class-imbalanced medical image classification tasks
Rajaraman S, Ganesan P, Antani S · PLOS ONE · doi:10.1371/journal.pone.0262838

Featured Projects

🦟

Malaria Screener

Smartphone-based AI malaria diagnostic tool converting commodity Android devices into field-deployable microscopy units for blood-smear screening at the point of care.

CNNsAndroidXAI
Learn More ↗
🫁

Pediatric TB Screening

Deep learning framework for pediatric chest X-ray analysis with domain adaptation, class-imbalance handling via latent diffusion, and failure-mode characterization.

U-NetViTDiffusion
Learn More ↗
🦠

COVID-19 Severity AI

AI models for radiographic severity scoring using self-supervised contrastive learning and large-vision-model knowledge adaptation for scalable triage beyond binary detection.

ContrastiveCXRLVM
Learn More ↗
🛡️

Trustworthy AI Framework

Calibration and fairness pipelines for medical imaging AI including uncertainty quantification, inter-annotator agreement analysis, and cross-domain robustness evaluation.

CalibrationFairnessUQ
Learn More ↗
❤️

AI Echocardiography Analysis

Real-time AI system for cardiac parameter estimation from echocardiography images, including open-world view classification U.S. Patent filed 2022.

Real-TimePatentCardiac
Learn More ↗
🤖

Multimodal Learning Pipeline

Combining real and synthetic image-text data to overcome limited training datasets in multimodal medical AI using LLM-generated and expert-curated radiology text.

LLMMultimodalSynthesis
Learn More ↗

Editorial & Peer Review Roles

Editorial Roles

Associate Editor / Editorial Board Member, PLOS ONE

Editorial Board Member, PLOS Digital Health

Editorial Board Member, PeerJ Computer Science

Organizer of special issues and workshops at international AI and medical imaging conferences

Top 1% reviewer recognition on Publons (2017–18, 2018–19)

Peer Review

Active peer reviewer for 100+ journals and conferences including The Lancet Digital Health, Nature Scientific Reports, IEEE Transactions on Medical Imaging, PLOS ONE, Medical Image Analysis, and more.

Best Employee Award (2018, 2020); Special Act or Service Group Award, NLM/NIH (2018)

IEEE Best Project Award (AT&T Bell Laboratories, 2000)

Memberships

Senior Member, IEEE
Senior Member, IEEE Engineering in Medicine & Biology Society (EMBS)
Life Member, SPIE

Open to research collaborations

I welcome partnerships in medical AI, dataset sharing initiatives, joint publications, and co-organization of workshops and special journal issues. If you are working at the intersection of AI and clinical medicine, let's connect.

Get in Touch

Contact Information

LocationBethesda, MD / New York, NY, USA
Emaildr.shiv.srk@gmail.com
Phone+1-202-322-0984

Send a Message