Alzheimer's disease
Open the annotated sources →Clinical seizures, memory and awareness interruptions, sleep-related discharges, and the strongest body of dementia-specific detection research.
Read the full guide →Evidence and methods
See what supports each claim
Every important finding should answer four questions: Who was studied? What did researchers measure? What did they find? How far can that finding be applied?
Sources for all five guides
Find the evidence for each dementia type.
Each guide places citations and study limitations beside the finding they support. Its source list then explains what each paper studied and how directly it applies.
Dementia with Lewy bodies
Open the annotated sources →Documented seizures alongside changing alertness, sleep events, movement problems, and fainting. The guide separates general EEG slowing from seizure-related activity.
Read the full guide →Frontotemporal dementia
Open the annotated sources →Recent evidence on epilepsy before and after diagnosis, focal events, and the recognition challenges created by behavioral and language-led forms of FTD.
Read the full guide →Vascular dementia
Open the annotated sources →Published clinical frequencies, why the timing and location of vascular injury matter, and what stroke research can and cannot tell us about dementia.
Read the full guide →Mixed dementia
Open the annotated sources →Seizures when more than one dementia-related condition is present, why the exact combination matters, and why subtype percentages cannot simply be averaged.
Read the full guide →Evidence strength
Judge each claim by the study behind it.
A dementia type may have strong evidence for diagnosed seizures and very little direct research on seizures without outward signs. One overall grade would hide that difference, so the atlas explains the strength and limits of each major finding.
Did the study examine the right people and finding?
A stroke study can help explain a vascular dementia question, but it cannot provide a seizure percentage for people with vascular dementia.
Did the method fit the question?
Researchers may review events in medical records, confirm them clinically, or measure electrical activity with EEG. The group, comparison, recording method, and definition of the outcome all affect the answer.
How certain is the result?
Independent studies provide more confidence when they support the same point. The number of relevant people and events also matters. A large study can still have a very small and uncertain subgroup.
Different studies answer different questions
- Clinical groups and health records
- Show how many diagnosed seizures or cases of epilepsy were recorded in a particular setting and period. Looking back through records can miss unrecognized events and may provide few details.
- Studies of brain tissue after death
- Confirm which brain diseases were present and show how they relate to seizure history. They can find associations, but they usually cannot prove which condition caused a seizure.
- EEG recording studies
- Show what a particular method detects under particular conditions. Time, sleep, electrode coverage, interpretation, and findings in controls all matter.
- Reports about one person or a small group
- Can show that an event or detection limit exists. They cannot tell us how common it is or reliably compare treatments.
- Treatment and technology studies
- A carefully designed treatment trial can test whether an intervention changes a defined outcome. Showing that a device works or detects activity is an earlier step and does not establish better thinking, safety, or quality of life.
No formal clinical grading is implied. This is a focused educational synthesis, not a systematic review, a validated risk model, or a clinical practice guideline.
Reading published numbers
Ask who was counted, what was counted, and for how long.
How many people had the condition
A percentage may describe how many people had epilepsy at one point or across a defined period. Epilepsy at the time of diagnosis differs from all seizures recorded across the full course of dementia.
How often new diagnoses occurred
A rate per 1,000 person-years combines new diagnoses with the amount of time people were followed. It is not a percentage. An estimated chance over time also differs from simply dividing the affected group by the starting group.
How many people had an EEG finding
The percentage of people with brief discharges describes what one study found under its recording conditions. It does not show how common clinical seizures are or how reliably the test detects all activity.
How much uncertainty surrounds the number
A wide confidence interval means the true value could be spread across a broad range. A difference that is not statistically clear does not prove that the groups are identical.
Where the evidence comes from
Readers should be able to trace every important claim.
Current evidence first
Publications from the preceding ten years are prioritized. Older studies are retained only for a specific foundational contribution or important citation context, with the exception explained. The vascular guide's 2013 population study is one such exception.
Primary sources, with access limits visible
The core dementia-seizure collection is supplemented by targeted primary research and professional guidance. Links lead to publication records, DOI pages, or lawful full text. When only an abstract was reviewed, that limit is stated.
Study findings and Seagull's perspective
The guide separates what researchers reported from Seagull's care implications and research priorities. A relationship between two findings is not presented as proof that one caused the other. An early result from a small subgroup is not presented as an established treatment benefit.
Shared data and commercial involvement
Two papers using the same database do not count as two fully independent confirmations. The guide notes relevant manufacturer involvement in technology studies. Including a technology does not endorse the product or mean it is available for dementia care.
Who is responsible for the atlas
Seagull Health selects the sources, explains the findings, and makes editorial corrections. The five subtype guides and shared explanations were aligned on . Each subtype page records its evidence-review date and explains the role of its sources.
The atlas is a focused educational review and has not undergone independent clinical peer review. It supports informed understanding and discussion. It cannot diagnose an event or select tests or treatment for one person. Read who stands behind the atlas and what editorial review means.
Corrections and suggested research: send a page-specific comment or publication link to Seagull Health. Do not include identifying health information or medical records.
Previous editorial update: 4 September 2026
: expanded the homepage's explanation of why recognition and detection matter in everyday dementia care, with links to existing studies and guide sections. Added publisher responsibility, founder background, review boundaries, and a corrections route.
Added Research insights, with study-specific findings and Seagull commentary on home recording, computer-assisted detection, and subscalp monitoring. The EEG explainer links to these analyses rather than duplicating them.
Clarified Seagull's contribution to care discussions and detection research, and disclosed AI-assisted editorial preparation separately from the AI technologies discussed in the literature.
This is an editorial update with targeted source checks, not a new comprehensive literature search. The evidence-review dates on the five subtype guides remain unchanged.
Search scope and selection
Make the review process as clear as the findings.
The atlas draws on the same collection of studies as Seagull's Dementia-Seizure Evidence Map, along with targeted primary studies and professional guidance. The map shows every paper in that collection. The atlas is a selected educational synthesis and does not claim exhaustive coverage of the literature.
- What the dates mean
- Evidence-review dates identify when study details were reviewed for a guide. Editorial dates identify changes to wording, navigation, or methods. Neither is automatically the date of a comprehensive database search.
- What is documented today
- The guides identify their reviewed sources, roles, access limits, and evidence-review dates. A complete historical search log with exact database queries, result counts, and screening decisions is not available for this edition. Those details have not been reconstructed or inferred.
- What future source updates will record
- Each update will identify the question and subtype, sources searched, search date, exact query and filters where applicable, records considered, selection decisions, full-text or abstract access, and the claims affected. Suggested papers and citation-following will be recorded separately from database searches.
- How a paper earns a place
- It must inform an identified atlas question. Direct subtype evidence is distinguished from broader dementia research, related epilepsy or stroke research, and professional guidance. The review considers the population, outcome definition, study design, uncertainty, overlapping datasets, access limits, and relevant commercial relationships. Recent studies receive priority; an older study needs an explained foundational contribution.
A new publication may strengthen a finding, weaken it, or leave the answer uncertain. Selection must allow for each possibility.
Maintenance and corrections
Keep the evidence open to revision.
Seagull Health is responsible for editorial decisions. The maintenance plan calls for a quarterly review of new evidence and unresolved questions, with earlier assessment of potentially important corrections. This is a planned review cycle, not continuous literature monitoring.
Record the question.
A submitted paper, reported error, retraction, corrected source, changed professional guidance, or relevant new study starts a review record. Record the affected claim and whether it could change interpretation.
Check the source and its implications.
Compare the original claim with the source. Preserve its population, clinical or EEG outcome, uncertainty, and access limits. Seek relevant independent expertise for disputed clinical interpretations.
Explain the change.
A substantive correction records the page, previous interpretation, corrected interpretation, supporting source, and date. A clarification is identified separately from a change in evidence. Personal information from feedback is not published.
If an important claim cannot be supported, it should be corrected, qualified, or withdrawn. The change record should explain the decision. Publication dates and study-review dates remain separate.
Suggest a correction or a study →Version and citation
Identify the edition you used.
Editorial version 2026.09.05 identifies this revision of the atlas. The study-review dates remain on the individual guides. A version change does not mean every source was searched or reviewed again.
Editorial and correction history
- 2026.09.05 · Editorial revision
- Added the search-record boundary, maintenance plan, version and citation guidance, independent-review invitation, international-use context, and focused collaboration questions. Corrected the expansion of DSEF on the About page to Dementia Seizure Evaluation Framework. This revision adds no new clinical evidence and does not change the subtype study-review dates.
- 4 September 2026 · Earlier update
- Added research commentary, editorial accountability, AI-assistance disclosure, and a corrections route. Read the retained update record. No version identifier was recorded for that update.
This history begins with the documented updates above. It is not a reconstruction of every earlier edit. The public URL may change over time; a citation identifies the version used but is not an immutable archive link.
Authorship and accountability
Who stands behind this atlas?
Seagull Health publishes the atlas and is responsible for selecting sources, explaining the evidence, and correcting the public pages. Study findings and Seagull's perspective are identified separately.
- Founder and research background
- Russ Barker, MHA, DHSc(c), founder of Seagull Health, brings long-term care leadership and doctoral research focused on recognizing seizures in dementia. He holds a Master of Health Administration and is a Doctor of Health Science candidate studying Clinical Nutrition and Global Health. Read the founder's background.
- Editorial work and clinical review
- Seagull reviews sources and prepares the public explanations. The atlas has not received independent clinical peer review, and citation does not mean the study's authors or institutions endorse it. This is an educational resource rather than a clinical practice guideline.
- AI-assisted preparation
- AI tools assisted with research summaries, source checks, drafting, and website development. Seagull Health remains responsible for every source, published claim, and correction. This assistance is separate from the AI detection methods discussed in the research.
- Review scope and dates
- The guides draw on a focused research collection, selected additional studies, and professional guidance. Each subtype guide states when its evidence was reviewed. A wording or navigation change does not mean the full literature was searched again. Read the scope and latest editorial update.
- Commercial context
- Seagull Health also develops the separate public and organizational DSEF resources. Research discussed in this atlas is not an endorsement of those resources or of a recording device. Relevant manufacturer involvement is noted in the technology discussions.
Suggest a correction or a paper
Send editorial feedback to Seagull Health’s contact page. The “Atlas: correction or suggested study” topic is selected for you. Include:
- The page title or link.
- The statement you think needs correction, or the question a new paper addresses.
- A publication link or DOI and a brief explanation of why it matters.
Do not send medical records, patient or resident names, or other identifying health information. This is an editorial contact, not a route for individual medical advice or emergency care. Read how corrections are handled and view the change history.
Independent review
Help scrutinize the clinical and evidence interpretation.
Independent clinical peer review has not been completed. Seagull welcomes expressions of interest from clinicians with epilepsy and dementia expertise, and researchers experienced in evidence synthesis.
A defined review task
Review one subtype guide or a specific clinical-event, EEG, or treatment-evidence section. Check claim-to-source accuracy, competing explanations, applicability, and whether the plain-language account preserves uncertainty.
Accountability readers can inspect
Any completed independent review should identify the reviewer with their consent, relevant expertise, section and version reviewed, review date, and conflicts of interest. Unresolved disagreements should remain visible. Reviewing one section does not imply endorsement of the whole atlas or DSEF.
Send your area of expertise, the section you could review, and relevant professional or commercial relationships. An expression of interest does not make someone a named reviewer or adviser.
Express interest in independent review →Research collaboration
Bring a specific question the evidence has not settled.
Seagull invites focused discussion with clinicians, researchers, care teams, families, and technology developers. The starting point is a defined gap in understanding, with the outcome and population made explicit.
Recognition across care settings
Which observations are lost between an episode and its clinical review? Propose a population and setting, and how completeness of the event account and competing explanations could be assessed.
Detection in underrepresented groups
Which subtype, disease stage, or care need has been insufficiently studied? Define the electrical finding, reference method, missed activity, false detections, recording quality, and assistance required.
Outcomes that matter to people
Does additional information change care or improve daily life? Propose how function, comfort, safety, care-partner workload, access, and unintended effects would be measured.
What to include in an initial message
Send the atlas page or research gap, a short question, the dementia population and setting, relevant publication links, your team's contribution, and the specific help you are seeking. Disclose relevant commercial relationships. A short, nonconfidential outline is enough.
Do not send patient records, identifying health information, unpublished datasets, or confidential study materials. Data access, research approvals, funding, and responsibilities would need to be agreed separately before any study work.
International use
Keep the local context beside the evidence.
This edition is available in English. It does not establish that its explanations, observation prompts, or research findings have been validated across languages or healthcare systems.
Ask who and where a study represents.
Country, care setting, access to testing, participant selection, disease stage, and care-partner support affect how directly a finding may apply. Evidence from a specialist service may leave important questions for community or residential care.
Adapt language and pathways with local expertise.
Any future translation needs clinical and plain-language review, a named language version, and preservation of the source limitations. Emergency contacts, referral pathways, and professional responsibilities depend on the local setting.