NoShow-Predict: An AI-Powered Software for Patient Scheduling Management

UTHealth creators have utilized artificial intelligence (AI) to create a software tool that is integrated with existing electronic health records (EHR) systems to analyze historical patient attendance patterns and non-sensitive patient attributes, and assigns a no-show probability to every upcoming appointment. Staff can then trigger targeted actions—extra reminders, wait-list fills, or strategic overbooking—potentially translating to multimillion-dollar annual savings with only a marginal rise in total visits.

Current Challenges with Patient No-Shows

Unexpected patient no-shows continue to undermine scheduling efficiency and inflate operating costs in medical clinics despite sophisticated EHR systems.

Introducing: NoShow-Predict

UTHealth Houston creators have developed NoShow-Predict, a software tool that tackles the chronic problem of unexpected patient no-shows by using advanced machine-learning algorithms to forecast no-show risk before it disrupts the day’s schedule.

Technology Differentiators

NoShow-Predict’s architecture pushes the frontier of AI by tightly integrating three complementary strengths:

• temporal reasoning that captures how events unfold over time

• spatial context that links related entities within each moment

• augmentation components that create enriched, synthetic features.

Working together, these elements give the model a holistic view of both sequence and structure while expanding the data landscape from which it can learn. This tri-layered design drives performance well beyond conventional approaches and establishes a new standard for state-of-the-art predictive modeling.

Intended Users and Potential Benefits

Front-desk schedulers, clinic managers, and health-system operations teams would benefit directly from the use of NoShow-Predict, as the software cuts the frequency and impact of missed appointments, reduces manual rescheduling workload, improves resource allocation, and ultimately lowers overall clinic operating costs while enhancing patient access. Clinicians are also direct beneficiaries as their schedules are maximized for efficiency. Patients can also benefit from a clinic’s use of the software, as earlier appointment times may become available due to the more efficient management/allocation of schedules.

Intellectual Property Status

Available for licensing

About the Creators

Martin J. Citardi, MD

Vice Dean for Clinical Technology

Chair of Otorhinolaryngology-Head & Neck Surgery

Xiaoqian Jiang, Ph.D.

Associate VP for Medical AI, Chair of Department of Health Data Science and Artificial Intelligence

Professor in Biomedical Informatics and Bioengineering

Jiajie Zhang, Ph.D.

Dean, Professor, Chair of McWilliams School of Biomedical Informatics