AD/PD™ 2026 · Copenhagen · 17–21 March 2026 · Two ePosters

Every speech reading traced to features a clinician can check

Name: Cephalgo ePosters at AD/PD™ 2026, Copenhagen
Start: 2026-03-17
End: 2026-03-21
Location: Copenhagen

Cephalgo’s two AD/PD™ 2026 ePosters, Copenhagen, March: speech recordings distinguished people with Alzheimer’s dementia or MCI, and people with Parkinson’s, from healthy controls in existing public datasets. Each reading traces to named features.

Request both ePosters Talk to our team

The posters

Validation of fully explainable speech biomarkers for AD/MCI screening

Validation of fully explainable speech biomarkers for PD/HC screening

Request the AD/MCI ePoster Request the PD/HC ePoster

Conference: AD/PD™ 2026, the International Conference on Alzheimer’s and Parkinson’s Diseases and Related Neurological Disorders
Dates: 17–21 March 2026
Venue: Copenhagen, Denmark
Data: Recordings from two independent public datasets, analysed after the fact

What it shows

Speech alone, with every reading explained

Both posters test one method on recordings that already existed: two branches read the same audio, and each reading can be traced back to named speech features.

Alzheimer’s or MCI versus healthy adults

Speech recordings distinguished people with Alzheimer’s dementia or MCI from healthy older adults of a similar age, in an independent public cohort.

Parkinson’s versus healthy controls

In the second ePoster, speech recordings distinguished people with Parkinson’s disease from healthy controls, in a second independent public dataset.

Each reading can be traced

A feature branch uses named acoustic, timing and, for Alzheimer’s, language measures, so each reading goes back to features a clinician can inspect.

Two branches, combined by agreement

A deep‑learning branch on the raw audio picks up subtle patterns the named measures do not fully capture. The two are combined through agreement‑based decision logic, which supports stable classification. Repeat recordings of the same people were used to test how well readings hold up from one test to the next.

What it means for your trial

Start with audio you already hold

The method works on recordings that already exist, so a sponsor can test it before planning any new study.

Audio from a trial you already ran

A re‑analysis pilot reads recordings you already hold, on your trial’s own scale, with the named features behind each reading. Re‑analysis pilot ›

Alzheimer’s and MCI programs

What Cephalgo reads in Alzheimer’s and MCI trials, including pre‑screening. Alzheimer’s › Pre‑screening ›

Parkinson’s programs

What Cephalgo reads in de novo and early Parkinson’s trials. Parkinson’s ›

How each reading is tested

Each reading is evaluated stage by stage, from the recording to the clinical scale. Validation approach ›

Our 2026 posters

From Copenhagen to Boston

Five posters in 2026, from explainable screening in Copenhagen to phone and software changes in Basel, Philadelphia and Boston.

Posters and white papers