Kentucky Open Science
A statewide hub that connects people who have research challenges with the data, models, and computing to address them, and makes every resource in the Commonwealth usable from one place.
Search the hub
Ask a question in plain language, or search models, datasets, projects, people, and publications.
Define a challenge once. The hub keeps working on it.
Agents keep evaluating new methods against each challenge, PanAtlas links what they find, and the people working on related challenges are brought together.
Projects
Applied research projects across healthcare, agriculture, public health, and education.
smell-datasets
Projectsmell-datasets Unless otherwise noted, this work is derived from the Institute for Biomedical Informatics Innovation Core at the University of Kentucky. Introduction Electronic nose (e-nose) technology is a type of…
Clinical Trial Matching
ProjectAI-powered patient-to-clinical-trial matching using reasoning models for eligibility decisions.
llm-basics
ProjectLLM Basics Simple, self-contained examples showing how to use an OpenAI-compatible API for various AI tasks. The scripts work with OpenAI directly or any other OpenAI-compatible endpoint (vLLM, LLM Factory,…
kentucky-open-science.github.io
ProjectKentucky Open Science — wiki and task board This repository is two things: The KOS wiki at kentucky-open-science.github.io: a searchable page for every public repository in the Kentucky-Open-Science organization, generated…
free-claude-code
ProjectQuick Start: LLM Factory (OpenAI-compatible) This is a fork of Alishahryar1/free-claude-code, maintained by Evan Damron. It adds an openai_compatible provider that points FCC at any OpenAI-compatible Chat Completions endpoint —…
native_research_llms
ProjectThe Academic Foundation Model Index (AFMI) An index of foundation models trained natively, from scratch, on their own data and tokenizer, by academic and research institutions. Universities come first; research…
AI Models
Pre-trained and fine-tuned models for computer vision, NLP, and medical AI.
KOS-V5
ModelVariants: Kentucky-Open-Science/KOS-V5-Base, Kentucky-Open-Science/KOS-V5-Instruct license: apache-2.0 library_name: transformers pipeline_tag: text-generation language: en developed_by: University of Kentucky (College of Medicine Office for Research; Center for Clinical and Translational Science) and University of…
DALE-CT-2S
ModelKOS-V4-Instruct-GGUF
ModelDALE-CT
ModelVariants: Kentucky-Open-Science/DALE-CT-1S, Kentucky-Open-Science/DALE-CT-2S, Kentucky-Open-Science/DALE-CT-0, Kentucky-Open-Science/DALE-CT-1S-v2 license: cc-by-nc-sa-4.0 datasets: ibrahimhamamci/CT-RATE rajpurkarlab/ReXGroundingCT library_name: timm pipeline_tag: image-feature-extraction tags: university-of-kentucky medical radiology chest-ct vision lejepa language: en Model Card for DALE-CT-1S Authors: Evan W.…
DALE-CT-0-L
ModelModel Card for DALE-CT-0-L Authors: Evan W. Damron (ORCID 0009-0009-1104-3202) · Mahmut S. Gokmen · Mitchell A. Klusty · Caroline N. Leach · Emily B. Collier · V. K. Cody…
Datasets
Curated datasets for training and evaluation.
Recent Publications
Peer-reviewed papers and preprints from Kentucky researchers.
Abstract Nontuberculous mycobacteria (NTM) are ubiquitous bacteria that cause a spectrum of diseases, most notably pulmonary disease (NTMPD). The host factors contributing to the heightened susceptibility and severity of NTMPD…
Abstract Alzheimer's disease neuropathological changes (ADNC)-operationalized with semi-quantitative parameters-represent the consensus-based gold standard for diagnostic evaluation of disease severity. Although useful, ADNC diagnostic frameworks have limitations, particularly in advanced disease…
Abstract Coronary artery calcium (CAC) scoring is a key predictor of cardiovascular risk, but it relies on ECG-gated CT scans, restricting its use to specialized cardiac imaging settings. We introduce…
Abstract Automated radiology report generation from 3D computed tomography (CT) volumes is challenging due to extreme sequence lengths, severe class imbalance, and the tendency of large language models (LLMs) to…
Abstract Coronary artery disease (CAD), one of the leading causes of mortality worldwide, necessitates effective risk assessment strategies, with coronary artery calcium (CAC) scoring via computed tomography (CT) being a…
Abstract Histopathological image analysis plays a critical role in modern medical diagnostics, particularly in the detection and classification of various types of cancer. This study proposes a method called HistoDARE…
Grants & Funding
Federal and institutional grants supporting Kentucky research.
Agency: NIH/NIDAAward Number: 1U54DA058256-01Program: AppalTRuST Project 3
Agency: NSFAward Number: OCI-1246332
Agency: NIH/HHSAward Number: R01 DK124774
Expanding access to AI training resources across Kentucky through the NAIRR pilot program.
Platform
Self-service tools for researchers — no programming expertise required.
Astral
PlatformThe agent layer: an orchestrator plans each task and calls specialist agents, each exposing one capability through the Model Context Protocol, under the participant’s own delegated authority.
Audio Foundry
PlatformSecure, web-based transcription platform with speaker labeling, timestamps, summarization, and theme extraction. Previously CAT-Talk.
CLASSify
PlatformNo-code web platform for training, comparing, and explaining classification and clustering models on tabular data, with synthetic data generation.
Cresco
PlatformThe transport layer: a distributed agent framework whose agents link isolated resources into one secure mesh and move data or computation under each site’s access controls.
Forecaster
PlatformWeb platform for time-series forecasting with multiple models and language-model-assisted interpretation, no programming required.