EnsoSleep is an AI scoring software solution that automates sleep event detection with 90 percent agreement. AI scored sleep studies are ready for a clinician’s overview and a discussion with a patient in about 15 minutes.
As an artificial intelligence (AI) solution in medicine, we’re often asked a lot of general questions about our AI scoring software, and how it works. Some folks ask about our AI’s capabilities: how fast is it? Is it consistent? What does it actually do? How does it work? Many in the sleep industry wonder what’s different between the autoscoring products they’ve seen before and our EnsoSleep AI scoring. To answer these questions, we compiled an AI Scoring FAQ with the most common questions we hear from sleep professionals interested in our AI technology.
What is AI?
Artificial Intelligence (AI) is the general field of study at the intersection of statistics, software development and theoretical computer science. AI developers ideate different applicable algorithmic approaches using large data sets to create machines with a certain level of intelligence. AI is often thought of as a robot, but a calculator, a scheduling assistant, an electric thermostat, and even your old Tamagotchi could be considered basic AI. The goal of many AI technologies is to solve problems the same way humans might, but with more accuracy and efficiency.
What is Machine Learning (ML)?
Machine Learning (ML) refers to a collection of methods that enables an algorithm to execute specific tasks without being explicitly programmed. This is achieved by taking advantage of previously observed data, and through a training process, allows the ML algorithm to learn relationships within the observed data in order to make predictions on newly collected data. As a result, an ML-powered algorithm will continue to improve over time.
What is Waveform AI?
Waveforms are used in healthcare to monitor, diagnose, and treat patients. The electric signal pulsing up and down on the monitor next to a patient’s bed is waveform data. Heartbeats on an EKG, eye movements through an EOG, and brain waves through an EEG all output as waveform data, with over 1.5 billion waveforms run per year globally across specialties. EnsoData’s Waveform AI leads the world in reading and understanding these waveforms. The first waveform AI product on the market was EnsoSleep, a sleep scoring AI built on over 500,000 clinical studies. You can see the other AI products in healthcare down below.
What is AI Sleep Scoring?
AI sleep scoring is the automated event detection and analysis of PSG or HSAT data performed by an AI scoring software, like EnsoSleep. The AI does this by analyzing the waveform data of sleep tests and providing an AI scored study for a registered clinician to review, all in a fraction of the time it takes to perform the task manually.
How effective is AI scoring?
People also ask, how consistent is AI? How accurate is it? Our EnsoSleep AI scoring requires a clinician’s review, and during those reviews we consistently agree with the overscorers at an average rate of 90 percent, with some clinicians agreeing at higher rates.
How fast can an RPSGT review an AI scored PSG test?
EnsoSleep identifies patterns and events in polysomnography tests (PSGs) for clinical diagnosis, replacing a workflow that currently requires hours from clinicians to manually mark reams of complex data by hand. This results in very efficient turnaround times. For most customers, clinicians can review an EnsoSleep AI scored PSG in around 15 minutes, compared to an hour on average for a test scored from scratch.
Can your AI score HSATs?
Yes, EnsoSleep works for home sleep apnea tests (HSATs). In fact, AI scoring makes at-home testing more scalable and profitable for providers as it helps open up precious time for other activities.
How fast can AI score an HSAT?
For all tests, the connection levels on the user end impact speeds, but EnsoSleep can score HSATs in about 15 minutes, including upload, scoring, and download times. Click on the image below to read a case study of how this was implemented at a 4-location sleep clinic.
What’s the difference between autoscoring technologies and AI scoring technology?
AI scoring technology is not like the old autoscoring, which was based on algorithms written by programmers to follow “if-this-then-that” rules on an event by event basis. With EnsoSleep AI scoring, we’re leveraging machine learning and deep learning in our algorithms. We’ve trained our system on hundreds of thousands of studies that were scored by RPSGTs and interpreted by board certified physicians. As a result, the AI scoring is a much more sophisticated, consistent, and adaptable solution than the other autoscoring solutions.
How does AI sleep scoring compare to outsourcing sleep study scoring?
Sleep scoring solutions powered by AI drive better turnaround times for understaffed sleep centers. In this case study, see how a 4-location sleep practice reduced patient times from weeks to same-day results: “Leveraging Cutting Edge Sleep scoring AI Saves iSleep Time.”
Can AI provide sleep staging?
Yes, our EnsoSleep AI does provide sleep stage scoring.
That’s all great, but is there a video I can share with my team that easily explains EnsoSleep’s AI scoring?
How can AI help the field of Sleep Medicine?
In a world in which 80% of sleep apnea patients have yet to find out if they are suffering from the condition the field of healthcare must choose to adopt necessary technology to speed up the process. EnsoSleep is a sleep scoring software designed to fill that immense need.
What is the sleep apnea scoring criteria for the AI?
Our AI scoring helps identify many sleep issues, including obstructive sleep apnea (OSA), periodic limb movements (PLMS) and other sleep disorders that are identified during a PSG or HSAT. Our AI reduces manual sleep analysis time allowing technologists to focus on the parts of the study that can lead to important diagnosis, such as heart arrhythmias or abnormal EEG.
Is the AI scoring FDA cleared?
The FDA cleared EnsoSleep in 2017. This early clearance was also one of the earliest AI-based devices in medicine, earning accolades in the Benjamens, et. al. study: “The state of artificial intelligence-based FDA-approved medical devices and algorithms: an online database.”
Do you partner with anyone in the sleep industry?
EnsoData partners with medical device, wearable, and EMR companies to further streamline integration and adoption of sleep scoring AI amongst the sleep community. For specific partnership queries, send us a message via our contact page.
What are the key benefits of using AI scoring?
The key benefits to using AI scoring are simple: consistency, simplicity, and speed.
- EnsoSleep provides more consistent results, faster turnaround for patients, and a better patient experience.
- Our solution minimizes downstream costs of untreated conditions and increases access to personalized treatment.
- EnsoSleep is interoperable with the leading medical and wearable devices, so there is no change to the workflows clinicians use today when implementing EnsoData’s technology.
What do your customers think of your AI scoring?
Customers love working with EnsoData: the company has 96% customer retention and has scaled to work with more than 400 clinics in the past 2 years.
What is the cost of your AI scoring?
Great question! Our pricing is designed to help you maximize your team’s time. If you’d like to know more about our tiered pricing options, please request a demo with one of our team’s account executives.
Do you have a sleep scoring manual?
We do have a user manual to help our users navigate our software, but we pride ourselves on fitting seamlessly into your workflow.
Do you have any other resources on AI scoring?
There are a handful of other great pages on our site to check out:
- Check out our EnsoSleep White Paper, the first major asset outlining our waveform AI scoring solution for sleep clinics.
- Dive into our educational sleep webinars. We have a variety of topics designed to help everyone in the sleep field.
- AI-Powered software has transformed a number of sleep clinics. Dive into our EnsoSleep case studies to read each story.
- Want to see our published scientific papers in sleep research? Check out our publications page for articles dating back to our inception in 2015.
We hope this helps you understand artificial intelligence sleep scoring a bit more, but if we didn’t answer your question, please, let us know what it is in the form below. Thank you.