AI Researcher. Medical imaging, machine learning, retrieval-augmented generation, knowledge graphs.

I am pursuing master's studies in Data Science and Artificial Intelligence at Campbellsville University. My earlier work in IT and healthcare administration involved data records, reporting, and information handling. My current focus is on research and practical tools for machine learning, medical imaging, and retrieval-augmented systems.

Portfolio and interactive demonstrations: https://mizanai88.github.io

Each repository runs offline on CPU from authored synthetic fixtures, carries tests and CI, documents its data access, and states exactly how it relates to research. None bundles a dataset. Implementation was AI-assisted on commission and reviewed before release; see each repository's NOTICE.md.

Longitudinal reporting and evidence

• longitudinal-report-workbench: interval-aware retrieval, patient-only verification, source-aware admission, one rewrite, template fallback, as an inspectable state machine.
• patient-evidence-verifier: the patient-only boundary as a type that cannot carry retrieved content.
• clinical-tuple-eval: exact clinical tuples, false content, omissions and report states with explicit denominators.
• temporal-retrieval-bench: does retrieval change appropriately with the comparison interval?

Retrieval and graphs

• graph-provenance-explorer: typed graph over a fictional corpus, BM25 and one-hop retrieval with provenance paths; companion to a coauthored article on knowledge-graph retrieval-augmented clinical summaries.

Medical vision

• medical-vision-baselines: multi-label chest image baselines on a MedMNIST task with strict split discipline and a real layer inspector.
• attribution-sanity-lab: when does an attribution map actually change?

Agricultural vision

• rice-leaf-ensembles, potato-leaf-transfer, coconut-leaf-benchmark: small benchmarks with duplicate checks, corruption suites, calibration and an audit trail from dataset to prediction.

Structured data and forecasting

• heart-risk-calibration: discrimination and calibration on a public structured dataset with in-fold preprocessing.
• mobility-demand-ml: hourly taxi demand with rolling-origin evaluation and labelled hypothetical scenarios.

• ClinicGraphRAG knowledge graph and LLM retrieval-augmented generation for reliable clinical decision support summaries. Discover Artificial Intelligence 6, 833 (2026). https://doi.org/10.1007/s44163-026-01492-w
• Medical Image Classification and Enhancement Using Machine Learning: A Focus on Fingerprint Colorized Data. Journal of Neonatal Surgery 14(32S), 415-431 (2025). https://www.jneonatalsurg.com/index.php/jns/article/view/7382
• Strategic Insights through Machine Learning: A Comparative Study of Uber and Lyft with a Subscription-Based Model for Lyft (2017-2024). IJSAT 16(2) (2025). https://doi.org/10.71097/IJSAT.v16.i2.4206
• Coconut Leaf Disease Detection using Deep Learning Techniques. IJSAT 16(1) (2025). https://doi.org/10.71097/IJSAT.v16.i1.1751
• Hybrid Deep Learning Framework for Enhanced Heart Disease Prediction: Integrating XGBoost and Capsule Networks with CNN-Transformer Architectures. Journal of Computer Science and Technology Studies 3(2), 116-123 (2021). https://doi.org/10.32996/jcsts.2021.3.2.9

Full author lists and neutral summaries are on the portfolio's research page.