AI system
- System
-
Nelli audio-video seizure monitoring system (computer vision and AI-based event detection)
- Model
-
Not assessed
- Release
-
Not assessed
- Application developer
-
Neuro Event Labs, Tampere, Finland
- Role
-
Not assessed
- Configuration
-
Not assessed
- Inputs and outputs
-
Not assessed
- Autonomy
-
Not assessed
- Workflow
-
A camera and microphone unit streaming to a server where computer-vision and AI algorithms flag motor events. Night-time use for people with suspected nocturnal seizures who could not tolerate EEG, with family or carer consent. Clinician review of flagged clips (the Cornwall paper says 8–9% of the recording is sampled by an epilepsy expert).
- Human oversight
-
Not assessed
- Relation to AI
-
The system flagged motor events that night staff and carers had not noticed (247 of 707 in Cornwall). In the Danish validation it detected every tonic-clonic seizure in real time. Clinicians still reviewed sampled clips to confirm events. Sensitivity for minor motor seizures was 8.3%. One Cornwall participant's unit was removed to prevent injury and equipment damage. The validation paper is co-authored by two vendor employees. The Cornwall evaluation is independent of the maker.
- AI contribution
-
The system flagged motor events that night staff and carers had not noticed (247 of 707 in Cornwall). In the Danish validation it detected every tonic-clonic seizure in real time. Clinicians still reviewed sampled clips to confirm events. Sensitivity for minor motor seizures was 8.3%. One Cornwall participant's unit was removed to prevent injury and equipment damage. The validation paper is co-authored by two vendor employees. The Cornwall evaluation is independent of the maker.
Benefits
- Outcome identifiers
-
Not assessed
- Benefit
-
Nelli flagged 707 possible seizure events, 247 unheard or unrecognised by carers
- Category
-
Improved Physical Health
- Status
-
Not assessed
- Beneficiary groups
-
People. Organizations.
- Beneficiaries
-
Not assessed
- Reach
-
Not assessed
- Distribution and access
-
Not assessed
- Duration
-
Not assessed
- Expected?
-
Yes
- Costs and harms
-
Not assessed
Measurement
- Studies
-
Not assessed
- Study design
-
Not assessed
- Measurement
-
Not assessed
- Setting and sample
-
Not assessed
- Measures
-
Nelli flagged 707 possible seizure events, 247 unheard or unrecognised by carers
- Comparison
-
carers' hearing or recognition of night-time events in the same people
- Results
-
Reference: 247 were not heard or recognised by carers. Result: Nelli recorded 707 possible seizure events across the study cohort.
- Additional value from AI
-
Not assessed
- Limitations
-
Not assessed
- Denominators
-
Not assessed
- Materials for replication
-
Not assessed
Money
- Kind of value
-
Not assessed
- Amount
-
Not assessed
- Currency and prices
-
Not assessed
- Period and unit
-
Not assessed
- Realization
-
Not assessed
- Recipients and payers
-
Not assessed
- Gross or net
-
Not assessed
- Included costs
-
Not assessed
- Comparison
-
Not assessed
- Method
-
Not assessed
- Return and payback
-
Not assessed
People and places
- Participants
-
Developer: Neuro Event Labs, Tampere, Finland. Operator: Cornwall Partnership NHS Foundation Trust (community evaluation); Danish Epilepsy Center, Dianalund, and Aarhus University Hospital (validation).
- Jurisdictions
-
Benefit: Cornwall,
United Kingdom; Dianalund and Aarhus, 
Denmark. Developer: 
Finland. Operator: United Kingdom. - Ownership and control
-
Not assessed
- Dependencies
-
Not assessed
- Cross-border connections
-
Not assessed
- Geopolitical implications
-
Not assessed
Follow-up
- Conditions for replication
-
Device deficiency kept low (10.5% of monitored time in the Danish validation). Staff available at night to act on alarms, the setting the validators say the system is suited to. The target is major motor seizures; minor motor seizures were detected at 8.3% sensitivity.
- Replication attempts
-
1 follow-up entries. Benefit recurred: 1.
- Later findings
-
9 October 2025. A later single-center feasibility study in Dravet and Lennox-Gastaut syndromes, with vendor co-authors, reported that Nelli detected 368 seizures at 97.8% accuracy over four weeks of home monitoring while diaries documented only 26.1% of seizures.
- Availability
-
Not assessed
- Institutional decision
-
Not assessed
- Grounds
-
Not assessed
- Action and result
-
Not assessed
- Success criterion
-
Not assessed
Evidence
- Source numbers
-
AIBD-0297-source-0001–0005
- Citations
-
Numbered references below.
- Language
-
Not assessed
- Source relationships
-
Operator, Developer
- Original evidence
-
Not assessed
- Access and versions
-
Not assessed
- Supported findings
-
Not assessed
- Evidence links
-
Not assessed
- Evidence strength
-
Not assessed
- Missing evidence
-
Not assessed
Grades
- Assessed benefits
-
People. Organizations.
- Depth
-
Not assessed
- Reasons
-
People: Carers reported peace of mind (13 of 14). Clinicians gained event counts for people who cannot report seizures. No cited source documents a treatment change or injury avoided for a named individual. Organizations: The Cornwall authors describe improved care planning and carer training. No measured change in staffing, response times or incidents is reported.
- Observed or intended
-
Not assessed
- Rubric
-
Not assessed
- Assessor and date
-
Not assessed
- Alternative assessments
-
Not assessed
- Fill
-
The deployer's account
Entry details
- Kind
-
Not assessed
- Related entries
-
Not assessed
- Field
-
Healthcare (epilepsy; nocturnal seizure monitoring). Residential and institutional seizure monitoring for people with intellectual disability.
- Purpose
-
Built to detect nocturnal major motor seizures from audio-video so that staff can intervene and clinicians can see events patients cannot report. The unrecognized events it surfaced are that purpose.
- Dates
-
Date basis unassessed: 2023 (precision unassessed). Updated: 21 October 2026.
- Stage
-
Not assessed
- Version
-
SHA-256: 916bef6a0a2a…
References
- Other identifiers
-
Not assessed
- Category mappings
-
Not assessed
- Brief
-
Not assessed
- Revisions
-
5 revisions. Latest: 21 October 2026.
Sources
-
Peer-reviewed study Deployer
Lennard S, Newman R, McLean B, Jory C, Cox D, Young C, Corson E, Shankar R (2023). Improving nocturnal event monitoring in people with intellectual disability in community using an artificial intelligence camera. Epilepsy & Behavior Reports 22:100603 (PMC10160340).
Nelli recorded 707 possible seizure events across the 14 PwID of which 247 were either not heard or not recognised by carers.
Nelli allowed for improved communication and care planning in a hitherto difficult to investigate population.
-
Peer-reviewed study Developer
Armand Larsen S, Terney D, Østerkjerhuus T, Vinding Merinder T, Annala K, Knight A, Beniczky S (2022). Automated detection of nocturnal motor seizures using an audio-video system. Brain and Behavior 12(9):e2737 (PMC9480955). Annala and Knight are Neuro Event Labs employees and shareholders.
The system detected all 11 TCS and four out of five (80%) hypermotor seizures.
we assessed the performance of automated detection of nocturnal motor seizures using audio‐video streaming, computer vision and an artificial intelligence‐based algorithm (Nelli)
-
Peer-reviewed study Developer
Wilms LK, Lossius MI, Annala K, Abdel-Khalik J, Fanter L, Elomaa K, Peltola J (2025). Seizure classification using a multimodal seizure monitoring system (Nelli) in Dravet and Lennox-Gastaut syndromes: A non-randomized, single-center feasibility study. Epilepsia, online 9 October 2025; doi:10.1111/epi.18640 (PMID 41066202; PMC12893256). Neuro Event Labs co-authors.