The Vital Signs team, AI developer Eetu Vilkki (left), engineer Aki Laakso and...
The Vital Signs team, AI developer Eetu Vilkki (left), engineer Aki Laakso and team leader Alexis Kouros presented the device at Web Summit in Lisbon in November 2021

Image source: Aalto University

News • Diagnostic advance

New AI assisted device could soon replace traditional stethoscope

A stethoscope is one of a doctor's most important instruments, but there haven't been any fundamental improvements in the device itself since the 1960s.

Now, researchers at Aalto University have developed a device that analyses a broad range of bodily functions and offers the doctor a probable diagnosis, as well as suggestions for appropriate further examinations. The researchers believe that the new device could eventually replace the stethoscope and enable quicker and more precise diagnoses. A startup called Vital Signs is taking the device to the market. The researchers are currently testing the device in a clinical pilot trial. Their intention is to launch the product in the most significant European markets by the end of 2023. ‘We have a well-functioning prototype, and the development path is clear,’ says Alexis Kouros, the doctor leading the research team at Aalto.

Photo
The device analyses bodily functions and offers the doctor probable diagnoses and suggestions for further examinations.

Image source: Aalto University

Vital Signs will be at Slush on Dec 1-2 and will take part in the Slush 100 pitching competition. The startup has reached the semifinals, a shortlist of the 20 best startups from the original pool of 100 candidates. ‘I’m proud to present our work, as we can really impact people’s lives with it. If financial success comes alongside this, good,’ says Kouros, describing his mood before the big event.

The new device records various bodily sounds and examines a broad range of characteristics, from oxygen saturation to pulse, ECG and temperature. It uses artificial intelligence to analyse this data together with information from a range of databases about patient records, laboratory results, medication and more. Following this analysis, the system presents the most probable diagnosis to the doctor in the form of a color-coded word cloud. It also offers a list of further examinations that could prove useful. No additional work is needed by a doctor to make use of the device. 

The research behind this began with the Biodesign-project, a concept developed at Stanford University in which a multidisciplinary team explores a hospital environment for weeks to identify and study areas that need improvement and to develop innovations to address them. The research team behind Vital Signs took part in the project in 2019. At the time, it also involved researchers from the University of Helsinki and the Hospital District of Helsinki and Uusimaa. After this initial phase, the project continued only at Aalto University, where it received funding from Business Finland. The project at the university will end in a few months, after which the work will continue in the startup.


Source: Aalto University

30.11.2021

More on the subject:
Read all latest stories

Related articles

Photo

Article • Lesion detection and mammographic texture analysis

Dual AI model to optimize breast cancer risk assessment

A team of Danish and Dutch researchers has combined an AI diagnostic tool with a mammographic texture model to improve the assessment of short- and long-term breast cancer risk. This new approach…

Photo

News • Voxel-wise classification

Deep learning tool uses MRI to enhance brain tumor diagnosis

A novel AI-based, non-invasive diagnostic tool enables accurate brain tumor diagnosis, outperforming current classification methods. The tool leverages MRI information to aid clinical decision making.

Photo

News • Interpretable machine learning system

AI to detect colorectal cancer from pathology slides

Researchers work on the first prototype that applies AI to colorectal diagnosis. The prototype achieved a diagnostic acuity of 93.44% and a sensitivity of 99.7% in the detection of high-risk lesions.

Related products

Subscribe to Newsletter