Speech recognition can detect warning signals for schizophrenia or Alzheimer's dementia

Ingelheim, Germany, July 08, 2019  – Can an app for speech recognition calculate a person’s individual risk of schizophrenia or Alzheimer’s Dementia? An international team of neuroscientists and IT specialists at Boehringer Ingelheim wants to find out by using artificial intelligence to grasp speech. This approach thus goes beyond the development of drugs with this research approach. The application analyzes language using specialized software with the intention to help detect mental disorders at an earlier stage to enable a treatment that is more effective. The underlying algorithms identify and classify speech patterns typical for these diseases to support a faster, more precise diagnosis than was possible in the past.

“From previous studies, we know how spoken language changes in patients with schizophrenia or Alzheimer’s Dementia, for instance,” explains Christoph Eschenfelder, Product Owner for the digital research project ‘Pathologic Speech Processing’ at Boehringer Ingelheim. “Particular when the disease is advanced many conspicuous characteristics occur. Using specialized algorithms, we want to identify these changes more quickly and thus support the early detection of such diseases.”

Patients benefit from earlier diagnosis
The example schizophrenia: In most patients (75%), schizophrenia develops between the ages of 16 and 25 and worsens as the disorder progresses. If detected and treated at an early stage, this process can be slowed significantly or even halted entirely. “We hope that the intelligent software will enable us to conduct a reliable risk assessment in minutes, with the level of precision that is expected for reliable early detection”, explains Eschenfelder. No unequivocal biomarkers exist for diagnosing schizophrenia, and the initial symptoms are nonspecific, making early detection of the disease difficult. For example, sufferers tend to withdraw socially, seem emotionless, or their sleep patterns change. Frequently these symptoms are misinterpreted as being related to puberty. Seeing a psychiatrist is widely stigmatized, but an app could overcome such inhibitions if it could predict an elevated risk of developing the disease based on answers to just a few questions.

Analyzing speech patterns faster and more precisely
Words, syllables, sentences, tones, and rhythm can be combined in an infinite number of ways. “Digital speech recognition software can read between the lines and use algorithms to identify patterns and logical relationships,” reports Eschenfelder. “We’ve now launched clinical studies to explore the use of artificial intelligence for speech analysis in the diagnosis of disorders like Alzheimer’s dementia and schizophrenia.” Previous studies with schizophrenia patients have already shown that altered intonation and reduced complexity of speech, for example, can be early signs of the disease. Alzheimer’s patients display changes in their speech, too.
Recordings of interviews conducted by psychologists with trial participants are based on a specific system of questions and provide the raw data for digital speech analysis. Complex search algorithms analyze the digitalized speech. Subsequently, the speech signals are converted for graphical representation so that sentence structure, meaning, intonation, and rhythm can be shown using colorful visuals. By means of artificial intelligence and machine learning, the system learns to recognize and classify patterns along the way.

Early detection during what is known as the “prodromal phase” of a mental or neurodegenerative disease (i.e., when no characteristic symptoms of schizophrenia or Alzheimer’s dementia have yet attracted notice) would be ideal. Boehringer Ingelheim is therefore targeting a combination of different methods for early detection, and to this end, the company is advancing the search for appropriate biomarkers, which could support early detection.

Boehringer Ingelheim
Improving the health of humans and animals is the goal of the research-driven pharmaceutical company Boehringer Ingelheim. The focus in doing so is on diseases for which no satisfactory treatment option exists to date. The company therefore concentrates on developing innovative therapies that can extend patients’ lives. In animal health, Boehringer Ingelheim stands for advanced prevention. 
 
Family-owned since it was established in 1885, Boehringer Ingelheim is one of the pharmaceutical industry’s top 20 companies. Some 50,000 employees create value through innovation daily for the three business areas human pharmaceuticals, animal health and biopharmaceuticals. In 2018, Boehringer Ingelheim achieved net sales of around 17.5 billion euros. R&D expenditure of almost 
3.2 billion euros, corresponded to 18.1 per cent of net sales. 
 
As a family-owned company, Boehringer Ingelheim plans in generations and focuses on long-term success. The company therefore aims at organic growth from its own resources with simultaneous openness to partnerships and strategic alliances in research. In everything it does, Boehringer Ingelheim naturally adopts responsibility towards mankind and the environment. 
 
More information about Boehringer Ingelheim can be found on www.boehringer-ingelheim.com or in our annual report: http://annualreport.boehringer-ingelheim.com.