Data Analytics and AI Pathway

If you are interested in learning more about data analytics and artificial intelligence, check out these sessions at the 2021 AACC Annual Scientific Meeting & Clinical Lab Expo.

Sunday

11001or 62001 Artificial Intelligence in the Clinic: Strengths, Weaknesses, and Opportunities

By nature, many of the traditional clinical tasks such as risk assessment, prediction of treatment efficacy, and forecasting patient trajectory can be thought of as prediction problems. Given sufficient amounts of patient data with outcomes, a machine learning model can make predictions which often exceed in accuracy human experts. However, to make these tools more applicable in the clinical setting, we need to augment artificial intelligence models with the ability to explain their decisions to humans, and assess their uncertainty. Join Dr. Regina Barzilay, the 2021 Wallace H. Coulter Lectureship Award winner in the opening plenary session to learn about artificial intelligence and its application to clinical sciences. Learn more.

193009 Doing More with R: Create Your Own Automated Reports and Dashboards

Dr. Shannon Haymond and colleagues lead this hands-on AACC University course aimed to develop skills in the use of R and RStudio for (1) reproducible data analysis and visualization workflows, (2) producing highly effective, publication-quality graphics, and (3) generating reports and dashboards that can be automated and easily shared. The focus will be on analyses and reports commonly used for clinical laboratory operational and quality assurance monitoring activities. Attendees will create a beginning-to-end workflow using provided templates and example data sets. The flexibility of visualization and report outputs from R makes this session applicable to anyone who wants to enhance their ability to reproducibly create figures and communicate data analyses. Learn more.

Monday

32226 How Artificial Intelligence and Machine Learning Will Help with Patient Diagnosis: Application to Autoimmune Testing

The development of artificial intelligence (AI) and machine learning (ML)-based technologies in medicine and pathology is advancing rapidly, but real-world clinical laboratory implementation has not yet become a reality. In this Scientific Session, Drs. Vincent Ricchiuti and Michael Mahler review some of the key practical issues surrounding the implementation of AI/ML. Learn more.

Tuesday

33102 Data Aggregation and Integration in Laboratory Medicine: How to Build Prediction Models and Learn from Multi-Institutional Data

As laboratorians, we are well aware of the richness of the data we report and of challenges of sharing and harmonizing laboratory testing data across practices. Broad sharing of laboratory data could catalyze secondary uses of these data that lead to improvements in quality. There are several challenges, including sharing of sufficient meta information to describe the testing context, adoption and implementation of rich communication standards, privacy concerns, and the necessary cost and effort. In this Scientific Session, we review the current landscape of data sharing and ongoing efforts to advance these practices. We highlight lessons learned from rapid efforts to share and learn from multi-institutional SARS-CoV-2 testing data, including the National COVID Cohort Collaborative (N3C). Finally, we demonstrate novel approaches and methods for building prediction models and learning from multi-institutional data. Learn more.

Wednesday

34223 Machine Learning Analysis of Laboratory Test Results Supports Clinical Decision-Making and Patient Care

Machine learning provides a powerful tool for integrating clinical laboratory data, developing novel clinical insight and driving intelligent clinical decision support. This Scientific Session will use an interactive, case-based approach to provide an intuitive overview of machine learning and its applications to laboratory medicine. The first talk will offer an overview of the value of using machine learning to support laboratorians and clinicians in patient and context-specific laboratory test result interpretation. We will also discuss strategies and barriers to the implementation of machine learning-based clinical decision support. The second talk will provide a practical explanation of commonly used machine learning algorithms and approaches. The third talk will offer an in-depth case study, using a project from the speakers’ research to illustrate how to undertake a machine learning-based analysis from start to finish. Learn more.

Thursday

35104 Mind the App: Application Development as a Solution to Unmet Needs in Laboratory Workflows

This Scientific Session will demonstrate the value of investing resources in producing in-house developed software applications (“lab apps”) to support laboratory workflows as well as discuss the important factors which need to be considered before adopting this approach. Virtually all modern clinical laboratories depend on a laboratory information system (LIS) for workflow management. LIS functionality has matured to the point where best-of-breed systems handle functional requirements in both routine and reference laboratory settings. Nevertheless, laboratory workflows vary across laboratory medical disciplines, hospital environments, and municipal boundaries. This results in gaps between what a generic LIS can provide and what is operationally required. Learn more.

Register for 2021 AACC

Join us in Atlanta, Georgia, September 26-30 to explore the cutting-edge science and technology shaping the future of laboratory medicine.

AACC’s COVID-19 Safety Plan

Check out the latest updates as of September 13 on onsite COVID-19 testing at the meeting and how to mitigate your COVID-19 risk while traveling.