autoSCORE™ AI-powered EEG Analysis

autoSCORE™ is the world’s first analysis tool that leverages advanced deep learning to streamline EEG review and interpretation. With expert-level analysis, autoSCORE flags clinically relevant abnormalities and provides study-level insights across long-term monitoring (LTM), ambulatory and routine studies.

By providing AI-assisted assessments and highlighting key findings, autoSCORE is designed to save save valuable time, allowing neuro care teams to focus more on patient care and less on manual data review.

Not available in all countries. Please contact your local sales representative for availability.

Expert-Level
EEG
Interpretation

Accelerates
Study
Reviews

Reporting
Consistency

Contact your Natus Rep

Book a demo

Explore more details

Resources

AI-assisted, expert-level EEG review

autoSCORE delivers advanced EEG abnormality classification that goes beyond flagging individual events. It automatically classifies abnormalities, provides a quick visual summary at the study level, and highlights segments of clinical significance — facilitating high quality, effective workflows, and consistency in reporting.

Automated abnormality categorization. Classifies detected abnormalities into four clinically relevant subtypes: focal epileptiform, generalized epileptiform, focal non-epileptiform, and diffuse non-epileptiform.
Study-level visual indicators. Get a quick visual summary of abnormality prevalence before opening a study, supporting workday planning.
Trace-level navigation. Quickly identify and navigate to traces or segments of clinical significance by abnormality type and probability score.
Confidence-weighted classifications. Each patient exam receives a study-level classification and color indicator that summarize the result at a glance.

Service and support

Looking for neuro product service or technical support?

Natus service and Support

Expert-level EEG review across study types

For LTM studies

AI-powered insights to help physicians reduce review burden and focus on critical study segments.

For Ambulatory EEG

Objective, AI-assisted analysis provides critical support when reviewing recordings more prone to artifacts, with results available while the study runs in the patient’s home.

For Routine EEG

Quick, confident routine EEG interpretation support—and help identifying normal studies.

For EEG service providers

Designed for faster, more consistent interpretations across large volumes of studies to enhance efficiency.

Peer-reviewed, proven performance

Developed using a deep-learning model trained on a dataset of over 30,000 expertly labeled EEG recordings*, autoSCORE has been extensively validated by experts, with results published in peer-reviewed publications.

autoSCORE performance is on par with leading human experts, outperforming similar AI models on the market.*
Consistently performs with accuracy, sensitivity and specificity near or above 90%.*

90%

Sensitivity*

87%

Specificity*

88%

Overall accuracy (vs. 83% human experts)*

95%

Normal vs abnormal (distinguishing)*

*Tveit J, Aurlien H, Plis S, et al. Automated Interpretation of Clinical Electroencephalograms Using Artificial Intelligence.
JAMA Neurol. Published online June 20, 2023. doi:10.1001/jamaneurol.2023.1645

Enhance EEG workflow efficiency

Designed with the user in mind, the autoSCORE application has been seamlessly integrated into industry-leading Natus EEG software, providing an intuitive experience. autoSCORE aims to improve efficiency and time to diagnosis, optimizing resources and giving your team more time to spend on patient care.

autoSCORE helps increase productivity when reviewing high volumes of EEGs, streamlines workflow planning by providing a daily overview, and may help reduce interscorer variability to improve overall quality.

Step 1

Start study with autoSCORE enabled

Step 2

Use study-level visual indicators to support workday planning

Step 3

Quickly identify traces or segments of clinical significance

Step 4

Facilitate quality assessments, faster review time, and more consistent reporting

Built into the software you know for results you can trust

autoSCORE is available exclusively on Natus EEG software, the most used and trusted platforms in the world

Confidence-weighted results, from study to trace level

autoSCORE classifies abnormalities into four clinically relevant categories

Focal epileptiform

Generalized epileptiform

Focal non-epileptiform

Diffuse non-epileptiform

Studies less than 4 hours - Routine EEG

Evaluated as a whole

Routine EEGs are evaluated as a whole. A probability score and color indication summarize the overall result at a glance, with the probability for each abnormality shown.

Studies over 4 hours - LTM

Evaluated in segments

Long-term EEGs are evaluated in segments. Probabilities show how many 2-hour segments contain each abnormality type, so the distribution across time gives better insight into review complexity.

Detailed analysis with trace-level assessments

View sort and filter the full event list then click an event to navigate directly to the abnormal segment. The distribution of events per abnormality type is easily visualized across the recording.