Understanding Apella's Case & Turnover Forecasts

Overview

In the OR, days rarely run exactly as scheduled. Apella helps your team see a more accurate picture of how the day is likely to unfold by providing precise predictions for when cases and turnovers will start and how long they'll take.

How Forecasts Work

Apella uses machine learning to analyze historical patterns and real-time OR data. Apella's system recognizes patterns across dozens of factors that affect case duration and the likelihood of starting on time.

Apella learns from two main data sources:

1. Apella-detected events: Real-time OR signals captured through Apella's ambient video, such as back table open, patient wheeled in, or patient draped. These signals help Apella learn surgeon- and procedure-specific trends for both cases and turnovers, and the models improve their understanding over time.
2. EHR data: Contextual case details from your EHR that impact case times, including time of day, day of week, whether the patient is inpatient or outpatient, case sequencing, and more.

By analyzing these factors together, Apella forecasts:

• Whether a case is likely to start on time
• How long each case is likely to take
• How long each turnover is likely to take

Dynamic Forecast Updates

Forecasts are not static. As the day progresses, Apella continuously updates predictions based on what is actually happening in the OR.

Case and turnover start and end times are dynamically adjusted, so the forecast always reflects the most accurate view of the day.

Understanding Specific Forecasts

In Schedule view, you can hover over or click into any specific case or turnover to view details on Apella's wheels in, wheels out, or duration forecasts.

Forecasted Wheels In and Wheel Out Times

To learn why a case is predicted to start or end at a different time than the EHR schedule: hover over or click the case to view details, and hover over the forecasted wheels in or wheels out times.

These forecasts depend on:

1. Pattern recognition: Apella's machine learning model has identified that cases in similar situations typically run this far ahead or behind schedule
2. Day-of factors: The prior case is running long or the turnover is taking longer than planned
3. Duration: How long the case is predicted to take (for wheels out only)

Note: Apella's case start time is based on patient wheels in and case end time is based on patient wheels out. This may differ from how cases are scheduled in your EHR.

Forecasted Case Durations

To learn why a case is predicted to run longer or shorter than scheduled in your EHR, hover over or click into the case for details. The forecast reflects what Apella has learned about similar procedures in comparable circumstances.

Forecasted Turnover Times

Hover over or click into a turnover to view details on a forecasted turnover duration. Predictions are segmented into three phases:

• Cleaning: Patient wheels-out to mop-out
• Cleaned: Mop-out to back table open
• Opening: Back table open to next patient wheels-in

🙋FAQs

Why is Apella’s schedule more valuable than the EHR schedule? 

Apella uses more data to make better predictions. It combines historical EHR data with real-time information captured by our cameras, resulting in more accurate forecasts. Unlike the EHR schedule that remains static even as your day unfolds, Apella continuously adjusts the schedule so you always have the most current picture of what to expect and can plan accordingly.

How is Apella’s schedule different from the EHR?

What Apella shows is slightly different from the EHR. Before a case starts, Apella shows when we predict the patient will wheel in and wheel out for each case, along with how long we predict cleaning and opening to take between cases.

Depending on how your site schedules cases, your EHR schedule may include some of these turnover components as part of the previous or following case, or not include them.

What data does Apella use?

It combines real-time Apella-detected events as well as EHR contextual case data. By combining these two data sources, Apella's machine learning models account for variables such as surgeon, procedure type, procedure phase lengths, the day of the week and time of day, and whether the case is inpatient or outpatient.  

How does Apella help during surgery?

Apella continuously updates predictions as cases progress, giving you an accurate, real-time view of how your day is unfolding and what's coming next.

What about future schedules?

Apella provides more accurate forecasts for tomorrow and beyond than your EHR schedule, and these predictions improve continuously as the system learns from your facility's patterns.