Whether you're exploring data analytics for the first time or refining advanced machine learning models, ADLM's data science challenges offer hands-on opportunities to build practical skills using authentic laboratory datasets.
Note: In July 2023, AACC (American Association for Clinical Chemistry) became the Association for Diagnostics & Laboratory Medicine (ADLM). Some historical references may use our former name.
Artery learning challenges
The ADLM Data Analytics Steering Committee has developed a progressive series of challenges on the Artery platform where you can develop foundational to intermediate data skills at your own pace. These challenges encourage peer learning and knowledge sharing within the laboratory medicine community.
Available challenges include
- Basic Data Management
- Histograms
- Evaluating Correlations
- Calculation of the Anion Gap
- Subgroup Analyses
- Dirty Data
- Data Joining
- Data Joining 2
Submit solutions, ask questions, and learn from colleagues tackling the same problems. ADLM membership required to access Artery content.
Competitive data science challenges
2025 Challenge: LabDocs Unlocked
The Challenge: Build an AI tool that rapidly extracts and presents information from complex laboratory document repositories.
Why It Matters: Laboratory professionals spend countless hours searching through policies, procedures, and technical documentation. An intelligent extraction tool would free up time for high-impact work while ensuring quick access to critical information.
The Winner: Jonathan Montgomery from Indigo BioAutomation presented their winning solution at the ADLM. Watch the recording here.
Explore the Solutions: Review all competition entries and review the code for your own AI tool development on GitHub.
2024 Challenge: FairLabs
The Challenge: Create an accessible, shareable tool that visualizes fairness metrics and provides actionable insights to advance health equity in laboratory medicine.
What Made It Unique: This challenge encouraged participants to partner with local stakeholders and apply their tools to real institutional datasets, driving tangible equity improvements.
The Winners: Nathan Breit, Jing Zhang, Joyce Liao, and Kate Crawford from the University of Washington presented their solution at the ADLM 2024 Health Equity & Access breakfast.
Explore the Solutions: Review all competition entries and access the dataset for your own dashboard development on GitHub.
Developed in collaboration with the Informatics Section at Washington University School of Medicine and ADLM's Health Equity & Access and Informatics divisions.
2023 Challenge: Help with Hemolysis
The Challenge: Develop an algorithm to identify which phlebotomists would benefit most from hemolysis prevention training by predicting potential cost savings over the following year.
The Impact: Sample hemolysis leads to specimen rejection, patient redraw, delayed results, and increased costs. This challenge addressed a universal laboratory pain point through predictive analytics.
The Winners: Eric Olson (Babson Diagnostics), Dave DeCaprio (ClosedLoop), and Ethan Olson shared their winning approach at the 2023 ADLM Annual Scientific Meeting.
Test Your Skills: Access the competition description and dataset on Kaggle.
Developed in collaboration with the Informatics Section at Washington University School of Medicine.
2022 Challenge: Predicting PTHrP Results
The Challenge: Build a predictive model for parathyroid hormone-related protein (PTHrP) results using laboratory data available at the time of order.
Why It Mattered: Accurate prediction of PTHrP results could optimize test utilization and reduce unnecessary testing while maintaining diagnostic quality.
The Winners: Yingheng Wang, Weishen Pan, He Sarina Yang, and Fei Wang from Cornell University presented their machine learning approach at the 2022 AACC Annual Scientific Meeting.
Learn More:
- Access the competition and dataset on Kaggle
- Read about the competition and winners in CLN
- Watch Yingheng Wang explain the winning methodology in the video above
Developed in collaboration with the Informatics Section at Washington University School of Medicine.
Why participate?
Skill Development: Progress from basic data manipulation to advanced machine learning using real laboratory datasets
Networking: Connect with data-savvy colleagues across clinical chemistry, informatics, and laboratory medicine
Recognition: Winning teams present their solutions at ADLM's Annual Scientific Meeting and related events
Real-World Impact: Address genuine challenges facing laboratory professionals today
Get started
Ready to enhance your data science skills? Explore the Artery challenges or data science competitions. All experience levels are welcome.