Clinical and Bio Analytics Transplant Laboratory (CBATL)



Clinical and Bio-Analytics Transplant Laboratory (CBATL) of the University of Washington Department of Surgery provides analytical and interpretative expertise that facilitates discoveries with high clinical impact. CBATL is a “think tank” for improving transplant patient care. To improve transplant care, it is our belief that the best skill to have is “asking the right question”.  Our “team” of critical thinkers and working with other investigators is the best way to ask the right questions. Members of CBATL brainstorm research ideas to pursue, prioritize and select projects, and for those with expertise in a specific area of a project to provide project management skills to oversee that project to completion. 

Being able to ask the right question has never been more important with the burgeoning growth of new clinical and biologic data. This presents a seemingly insurmountable challenge to harness the data for improving patient care. CBATL was organized to meet this new challenge.

CBATL has expertise in machine learning, genomic evaluation, data science for large clinical repositories, microsimulation modeling, experimental design, text analytics, database design, and outcomes research. 

Vision: Improve transplant care

Specific Aims:

  1. Machine learning analytic support for grants and research studies
  2. Support trainees in developing data science skills
  3. Create machine learning analytic models to aid with patient care
  4. Facilitate deployment of analytic models by publications, applications, or website

Learn more about Transplant Surgery Research >>

Using Predictive Analytics to Predict Length of Stay in Transplant Patients


Dr. Jim Perkins developed predictive models helping transplant programs predict their patients length of stay depending on known information about their patients. Hospital length of stay (LOS) after liver transplantation correlates with liver disease severity, post-transplant survival rates, and transplant-associated cost. Using data for a national database, four predictive models for LOS were built for the various stages of the liver transplant process. These LOS prediction models can help guide patients care and counseling, predict outcomes, and direct research into mitigating factors that prolong LOS.

To see the predictive models click here >>

Dr. Perkins is also working with the University of Washington Medical Center's Center for Clinical Excellence on various projects including predicting patients risk for developing sepsis, readmissions to the hospital, and developing delirium in the ICU.



CBATL investigators are available for consultations with faculty and post-graduate research trainees working in all other topic areas. For more information, please contact Dr. James Perkins, CBATL Director, at

Disclaimer: Clinical models are not to replace physician insight. These are to be used as an aid for the physician’s thought process.