Hi, I'm Farjana Yesmin.

Machine Learning Researcher · Trustworthy AI for Healthcare

I am a Machine Learning Researcher dedicated to building AI systems that are as equitable as they are intelligent. My work sits at the intersection of Privacy-Preserving Federated Learning, Explainable AI, and Trustworthy Healthcare AI. I am the author and maintainer of FairHealth, an open-source Python library for fairness and privacy in clinical AI. I believe technology truly serves people only when it is grounded in transparency and clinical validation. Outside research, I find balance through the piano, the outdoors, and stories discovered while traveling.

Farjana Yesmin

Research

Privacy-Preserving Federated Learning

MedHE: CKKS homomorphic encryption combined with adaptive gradient sparsification — achieving 97.5% communication reduction without sacrificing accuracy or HIPAA compliance.

Fairness & Explainability

Intersectional bias diagnosis in image classifiers, fairness-aware ECG models across demographics, and Hybrid Fuzzy-XGBoost frameworks validated directly with clinicians (N=14).

Healthcare Applications

Dengue symptom triage chatbots, wearable fall detection for Parkinson's, maternal health risk prediction, and post-disaster equitable aid allocation in Bangladesh.

News

October 2026Paper accepted for an ORAL PRESENTATION at EAI BODYNETS 2026 (Florence, Italy) — "Generative AI for Synthetic Multimodal Body Sensor Data: A Rigorous Multi-Metric Evaluation with Fairness-Aware Generation."Accepted
September 2026Poster accepted at WiML Workshop @ NeurIPS 2026 (Sydney).Accepted
August 2026Paper accepted for an ORAL PRESENTATION at RTIS 2026 (Universitas Islam Negeri Sunan Kalijaga, Yogyakarta, Indonesia) — "Constraint-Aware Neuro-Symbolic Learning with Domain Knowledge Injection: A Type-2 Fuzzy Framework for Interpretable Analytics."Accepted
August 2026Paper accepted for an ORAL PRESENTATION at SIMBig 2026 (Universidad Católica San Pablo, Arequipa, Perú) — "Diagnosing and Mitigating Compound Class-Environment Bias in Image Classification: A Bias-Weighted Augmentation Framework."Accepted
August 2026Paper accepted for an ORAL PRESENTATION at SIMBig 2026 (Universidad Católica San Pablo, Arequipa, Perú) — "Joint Denoising and 3D Point Cloud Reconstruction from Single CT Images via Shared U-Net Representations."Accepted
July 2026Invited as Reviewer — Deep Learning Indaba 2026 Start-up Pitch Competition.Service
June 2026Invited to Program Committee — NeSy 2026 (20th Conference on Neurosymbolic Learning and Reasoning).Service
June 2026Book chapter proposal accepted — "Rule-Augmented Explainable AI for Multimodal Biosignal Monitoring in Low-Resource Healthcare", CRC Press.Accepted
June 2026Paper accepted for an ORAL PRESENTATION at CIBB 2026 (Sapienza University of Rome, Italy) — "MedHE: Communication-Efficient Privacy-Preserving Federated Learning for Healthcare Text Classification."Accepted
June 2026Paper accepted at ECAF 2026 — "Beyond the Black Box: Claim-Level Interpretability for Accountable Retrieval-Augmented Generation."Accepted
May 2026Abstract accepted at WiML Symposium @ ICML 2026 (Seoul).Accepted
May 2026Paper accepted at EIML@ICML 2026 — "When Symbolic Rules Cannot Find Their Own Triggers: An Epistemic Diagnosis of Predicate Grounding Collapse in Neuro-Symbolic Fraud Detection."Accepted
May 2026Invited to Program Committee — AIES 2026 (AAAI/ACM).Invited
May 2026 FairHealth datasets are now on Hugging Face: Bangladesh Dengue, PDNA 2022, and Maternal Health. [Hugging Face] Data
May 2026 Released FairHealth — An open-source library for Trustworthy Healthcare AI. pip install fairhealth · [GitHub] · [PyPI] · [Docs] · [arXiv] Software
May 2026Invited as Reviewer — ECAF 2026, Ghent, Belgium.Invited
Apr 2026Book chapter proposal accepted — "Clinician-in-the-Loop Agentic Systems for Maternal Healthcare", Springer Nature.Accepted
Apr 2026Paper accepted at MobiHealth 2026 (EAI, Heraklion) — Fairness-Aware ECG Disease Prediction in Wearable Systems.Accepted
Apr 2026Op-ed in Prothom Alo English — "When the algorithm doesn't listen: AI and justice in Bangladesh's clinics."Press
Apr 2026Paper accepted at ICAIHE 2026, Waseda University, Tokyo — Explainable AI for Maternal Health Risk Prediction.Accepted
Mar 2026Paper accepted at CCAI 2026 (IEEE, Nanjing) — Fairness-Aware AI for Post-Flood Aid.Accepted
Mar 2026PC Member — DravidianLangTech & LTEDI at ACL 2026.Service
Mar 2026Confirmed ACM Certified Reviewer — ACM HEALTH, TIST, ACMJRC, TOMM.Service
Oct 2025Two papers accepted at DASGRI 2026 (Springer LNNS, London) — Air Quality Prediction & Health Bulletin Analysis.Accepted
Oct 2025Paper accepted at EAI Healthwear 2025, Isparta, Turkey — ECG Disease Prediction in Wearable Systems.Accepted

Publications

Peer-Reviewed
AI Chatbots for Dengue Symptom Triage in Bangladesh: A Decision Tree Classifier Approach
Farjana Yesmin
Springer LNNS · DASGRI 2026, London
↗ ResearchGate
Optimized Hybrid ML Framework for Real-Time Air Quality Prediction in IoT-Enabled Smart Cities
Farjana Yesmin
Springer LNNS · DASGRI 2026, London
↗ ResearchGate
Evolving Health Indicators in Bangladesh: A Comparative Analysis Using Data Science Tools
Farjana Yesmin, Nusrat Shirmin
Springer LNNS · DASGRI 2026, London
↗ ResearchGate
Cross-Domain Evaluation for Multi-Task Learning in NLP: A Unified Framework for Generalization and Robustness
Farjana Yesmin
Artificial Intelligence eJournal, 8(4), 2025
↗ SSRN
Accepted · Forthcoming
Generative AI for Synthetic Multimodal Body Sensor Data: A Rigorous Multi-Metric Evaluation with Fairness-Aware Generation
Farjana Yesmin
EAI BODYNETS 2026 · Florence, Italy · October 2026 (Oral Presentation)
↗ Code
Constraint-Aware Neuro-Symbolic Learning with Domain Knowledge Injection: A Type-2 Fuzzy Framework for Interpretable Analytics
Farjana Yesmin
RTIS 2026 · Universitas Islam Negeri Sunan Kalijaga, Yogyakarta, Indonesia · October 2026 (Oral Presentation)
↗ Code
Diagnosing and Mitigating Compound Class-Environment Bias in Image Classification: A Bias-Weighted Augmentation Framework
Farjana Yesmin
SIMBig 2026 · Universidad Católica San Pablo, Arequipa, Perú · October 2026 (Oral Presentation)
↗ Code
Joint Denoising and 3D Point Cloud Reconstruction from Single CT Images via Shared U-Net Representations
Farjana Yesmin
SIMBig 2026 · Universidad Católica San Pablo, Arequipa, Perú · October 2026 (Oral Presentation)
↗ Code
MedHE: Communication-Efficient Privacy-Preserving Federated Learning for Healthcare Text Classification
Farjana Yesmin
CIBB 2026 · Sapienza University of Rome, Italy · September 2026 (Oral Presentation)
↗ arXiv↗ Code
Beyond the Black Box: Claim-Level Interpretability for Accountable Retrieval-Augmented Generation
Farjana Yesmin, Nusrat Shirmin
ECAF 2026
When Symbolic Rules Cannot Find Their Own Triggers: An Epistemic Diagnosis of Predicate Grounding Collapse in Neuro-Symbolic Fraud Detection
Farjana Yesmin, Nusrat Shirmin
EIML@ICML 2026 · Epistemic Intelligence in Machine Learning Workshop · Seoul, South Korea, July 2026
↗ OpenReview
Explainable AI for Maternal Health Risk Prediction: A Hybrid Fuzzy-XGBoost Framework with Clinician Validation
Farjana Yesmin, Nusrat Shirmin, Suraiya Shabnam Bristy
ICAIHE 2026 · Waseda University, Tokyo · July 2026
↗ Preprint
Fairness-Aware Representation Learning for ECG-Based Disease Prediction in Wearable Systems
Farjana Yesmin, Nusrat Shirmin
MobiHealth 2026 (EAI) · Heraklion, Greece · June 2026
↗ Preprint
Toward Equitable Recovery: A Fairness-Aware AI Framework for Prioritizing Post-Flood Aid in Bangladesh
Farjana Yesmin, Romana Akter
CCAI 2026 (IEEE) · Nanjing, China · May 2026
↗ arXiv
Forthcoming Book Chapters
Clinician-in-the-Loop Agentic Systems for Maternal Healthcare in Low-Resource Settings
Farjana Yesmin
In: Sustainable Agentic AI: Building Autonomous Systems for Social Good (Eds. M. Mehrotra et al.) · Springer Nature · Chapter Proposal Accepted, Full Chapter Under Review
Rule-Augmented Explainable AI for Multimodal Biosignal Monitoring in Low-Resource Healthcare
Farjana Yesmin
In: Explainable AI in Multimedia and Multimodal Applications · CRC Press / Taylor & Francis Group · Chapter Proposal Accepted, Full Chapter Under Review
Under Review
Equity by Design: Fairness-Aware Explainable AI for Maternal Health Risk Assessment in Low-Resource Bangladesh
Farjana Yesmin, Nusrat Shirmin, Suraiya Shabnam Bristy
SIMBig 2026 · 13th International Conference on Information Management and Big Data
Geopolitical Crisis Economics Forecasting in the Small-Data Regime: A Comparative Study with the Russia-Ukraine War Dataset
Farjana Yesmin, Romana Akter
DASFAA 2027 · The 32nd International Conference on Database Systems for Advanced Applications
Wearable Sensor-Based Freezing of Gait Detection in Parkinson's Disease: A Multi-Model Evaluation with Attention-Based Explainability
Farjana Yesmin
ISAS 2026 · 10th International Symposium on Innovative Approaches in Smart Technologies

Projects

📦
FairHealth: Trustworthy AI Library
Open-source toolkit for fairness auditing, hybrid explainability (Fuzzy-XGBoost), and private federated learning.
pip install fairhealth
PythonPyPIDatasets
↗ GitHub ↗ PyPI ↗ Docs ↗ arXiv ↗ HuggingFace
🔒
MedHE: Privacy-Preserving Federated Learning
CKKS homomorphic encryption + adaptive gradient sparsification for secure collaborative learning across medical institutions.
97.5% comm. reduction · 89.5% accuracy · ε ≤ 1.0
Federated LearningCKKSDifferential Privacy
↗ Paper ↗ Code
🤖
AI Chatbots for Healthcare in Bangladesh
Decision-tree chatbot for dengue fever symptom triage in low-bandwidth clinical settings across Bangladesh.
Published · DASGRI 2026 · Springer LNNS
NLPDecision TreesHealthcare
↗ Paper ↗ Code
🏃
Wearable Fall Detection — Parkinson's
Multi-sensor fusion with transformer encoding for real-time Freezing of Gait detection and attention-based clinical explainability.
Under Review · BHI 2026
TransformersSensor FusionXAI

Experience

2020 – Present
Graduate Research Assistant
University of Tulsa, OK
Multimodal deep learning for immune responses, neuropsychiatric disorders, and healthcare AI. Machine learning, network theory, and information theory.
Summer 2020
Summer Externship, AT&T
Leadership and technology strategy across HR, finance, and technology. Market trend analysis and business innovation.
Summer 2020
Summer Scholar, HP
Technology, sales, and analytics at a Fortune 500 company. Participated in the HP Bot-a-Thon RPA automation competition using UiPath. Certificate from HP Chief Transformation Officer.
Aug 2018 – Mar 2019
Graduate Teaching Assistant
Boise State University, ID
Taught Java and Data Structures programming labs. Mentored students in algorithms and debugging.

Skills

ML & AI

PyTorchTensorFlow Scikit-learnXGBoost Transformers / HuggingFaceFederated Learning Homomorphic EncryptionDifferential Privacy Neuro-Symbolic AISHAP / LIME Time Series AnalysisBioinformatics Computer VisionNLP LLMs & RAGFuzzy Logic

Programming

PythonR JavaC++ SQL / MySQLHTML / CSS / JS

Tools & Platforms

Git & GitHubDocker KubernetesJupyter Google ColabTableau MATLABSplunk UiPath / RPACKKS Libraries LaTeXMicrosoft Excel

Education

University of Tulsa
PhD in Computing
Boise State University
PhD in Computing (Transferred)
Jashore University of Science & Technology
B.Sc. in Computer Science & Engineering

Certifications

Nature Masterclasses: Focus on Peer Review · Springer Nature · Jan 2026
Elsevier Certified Peer Reviewer · Jan 2026
ACM Peer Reviewer Training · ACM
Google Data Analytics Professional
IBM Data Science & Applied AI Professional
UiPath RPA Certifications & MS Excel

Academic Service

2026Reviewer — Deep Learning Indaba 2026 Start-up Pitch Competition
2026PC Member — NeSy 2026 (20th Conference on Neurosymbolic Learning and Reasoning)
2026PC Member — AIES 2026 (AAAI/ACM Conf. on AI, Ethics, and Society)
2026Reviewer — ECAF 2026 (European Conf. on Algorithmic Fairness, Ghent)
2026Reviewer — EIML@ICML 2026 (Epistemic Intelligence in ML, Seoul)
2026Reviewer — PharML 2026 (ECML PKDD Workshop on Pharmacology and Machine Learning)
2026Reviewer — ACM ICMI 2026 (Multimodal Interaction, Naples)
2026PC Member — DravidianLangTech & LTEDI @ ACL 2026
2026Reviewer — IndabaX Nigeria 2026 (PMLR Vol. 319)
2026ACM Certified Reviewer — HEALTH · TIST · ACMJRC · TOMM
2026Reviewer — ACM IUI 2026 · PeerJ CS · ICICST 2026
2025PC Reviewer — ACM ICMI 2025 · AI for Math @ ICML · MUGen @ ICML · Deep Learning Indaba Ideathon

Beyond Research

Sports & Outdoors

🥾 Hiking 🏊 Swimming 🚴 Cycling 🎾 Tennis 🏃 Running ⛺ Camping

18 US states explored.

Music & Reading

Amateur pianist — watch a performance ↗

PhilosophySci-FiMystery ThrillerHistoryPsychology PoetryBiographyFantasyTravel

Travel Writing

Thanksgiving with an American Family Sunset at Camel's Back Park Summer in Salt Lake City First Ocean View in America Arizona Hot Springs Eid in Trump's America America's Most Beautiful Small Town BDNews24 — Expat Life

Other Information

Mobility: Open to relocation (Nordic/EU/North America) and fully remote work. Open to learning new languages for new opportunities.
Languages: Bengali (Native) · English (Fluent) · Norwegian (Elementary, Duolingo Score: 15, September 2026, Diamond League, 303-day streak).
Volunteer: Event Coordinator of the Bangladeshi Student Association at Boise State University(2018–2019), Blood Donor, Red Crescent Society (2014–2016) · JAGO Foundation · Sahosika — Women's Education · Education USA Bangladesh Panel Speaker.

Blog

I write about the philosophy and practice of trustworthy AI — from epistemic injustice in clinical algorithms to differential privacy trade-offs in global health. View all posts →

LLM Safety in Clinical Deployment: Why Prompt Engineering Is Not Enough Apr 2026
The Cost of Perfect Privacy: What Differential Privacy Hides from Global Health Apr 2026
From Black Box to Arguable System: A Practical Framework for Clinician-AI Collaboration Apr 2026
Epistemic Injustice in Clinical AI: When Marginalized Voices Are Silenced by Algorithms Mar 2026

Get In Touch

Open to research collaborations, speaking opportunities, and conversations about responsible AI.

Send an Email