Voice as a new diagnostic tool for ALS and FTD

Language changes
gradually in people with neurodegenerative diseases.
With the help of artificial intelligence, even the slightest change can become the basis
for diagnosis.

News from the German Research Center for Neurodegenerative Diseases on October 10

With the help of language tests, a preliminary diagnosis of severe neurodegenerative diseases should be possible
in the future.

Researchers at the German Research Center for Neurodegenerative Diseases (DZNE) are developing so-called digital biomarkers for this, which are criteria
that AI can use to detect whether a patient's speech patterns change due to disease.

The technology manages to detect the slightest change in speech that
the human ear cannot hear.
The DZNE project PROSA "A High-Frequency PROgnostic Digital Speech Biomarker with Low Stress" is currently supported by the Alzheimer's Drug Discovery Foundation and the ALS Targeting Program in the United States ALS Initiative) with a $200,000 grant
.

Researchers have long believed that language is a possible indicator
of neurodegenerative diseases.
"In early methods, scientists have assessed textual factors: how complex is the grammar used by the subjects, how large their vocabulary is, how do they string words together?" Professor Anja Schneider, head of the DZNE working group and head of the Department of Neurodegenerative Diseases and Geriatric Psychiatry at Bonn University Hospital, explains
.
However, according to the detailed record of what was said, it was a tedious and slow exercise
.

Modern technology now offers new possibilities: "AI can perform such analyses much faster and can also take into account the melodic aspects of speech
.
As part of the now funded project, Anja Schneider is leading research into language change in FTD patients, while her DZNE colleague, Professor Andreas Hermann, focuses on ALS patients
.
Two other private companies are also involved in the project
.

The procedure for the language test is very simple for patients: they are asked three open-ended questions, for example about their leisure activities, or about their profession
.
It is also possible to freely describe
the presented image.

The decisive factor is the spontaneous speech
of the test subject.
The AI then analyzes the complexity of speech, which also takes into account pauses between words, the rate of speech, and other melodic aspects of
the language.
In ALS patients, their breathing is often limited by disease progression, and AI can also detect abnormalities
at a very early stage.
"Dialect and other personal characteristics of speech had no effect on the accuracy of the results," says
Anja Schneider.
Her observation was: "AI detects nuances of speech changes that the average listener simply cannot recognize
without the help of technology.
" ”

Researchers in Bonn are incorporating data from two DZNE studies into the development of the method: in Describe-FTD and Describe-ALS, patients are followed up for longer periods of time through complex clinical
examinations.
For the PROSA program, some subjects also undergo various language tests
.
Their results are combined
with cognitive examinations.

The researchers hope that, thanks to these multi-layered insights, language indicators can be further developed: If brain records and genetic tests can show how far a disease has progressed, scientists can use language records at the same stage of disease development to determine how language changes
。 "We also wanted to find answers to other questions," says Anja Schneider, "for example, does the precision of the language test relate to the patient's form of daily activities?" Does language deteriorate linearly over the course of the disease so that we can use linguistic markers not only to make a diagnosis, but even to draw conclusions about the speed of progression of the disease in the corresponding patient? ”

100 patients from the ALS cohort and 100 patients from the FTD cohort will be included in the study, and additional subjects
will be added to the control group.

Until now, the diagnosis of patients with FTD and ALS has been very complicated
.
It can usually only be carried out
in highly specialized specialist centers.
The necessary examinations take at least ten hours, and the waiting time for appointments is up to half a year
.
With the help of a speech test, a potential patient can be examined in advance, preferably with diagnostic support
to the doctor over the phone.

： ，
。   ， ，