Global ID, Idiap partner to scale finger vein biometrics with machine learning

Swiss startup Global ID has announced that it has launched a new project with the Idiap Research Institute that aims to improve finger-vein authentication (FVA) with machine learning.

The initiative is named KIDOLEH and will focus on industrializing the Global ID’s F2 finger-vein imaging device by implementing a new data processing system that meets the speed and security needs of high-security environments, the firm announced last week.

The goal is to enable finger vein biometrics for use in high-stakes environments, such as hospitals, public administration, and secure infrastructure. Healthcare is a particular target, as the technology could help match patients to medical records, strengthen access control for medical staff and prevent fraud.​

“KIDOLEH will allow us to deliver ultra-secure biometric solutions that meet the demands of modern healthcare and other sensitive domains,” says Global ID’s CEO Lambert Sonna.​

The new data processing pipeline will be evaluated offline using data captured from the Global ID’s F2 device. The aim is to achieve high accuracy and fast enough template comparison for practical de-duplication, the company explains in a release.​

Global ID and Idiap have a long history of collaboration on finger vein biometrics.​

Sonna helped develop the technology with a multidisciplinary team from Idiap, EPFL (École Polytechnique Fédérale de Lausanne), the Swiss Centre for Electronics and Microtechnology (CSEM) and HES-SO Valais-Wallis. The team created a biometric authentication technology based on multi-view vein scanning called BioID and BioLocker.​

In 2016, he founded Global ID in 2016 as a spin-off from EPFL. The startup has developed a device that captures biometric data by taking a multi-view image of finger veins. The scans are converted into a cryptographic representation, while the original images are deleted to ensure data security.

In 2020, Idiap and Global ID launched the CANDY project (Contactless finger vein recognition and presentation attack detection on-the-fly), which aimed to deliver a mobile contactless scanner that reads veins as a biometric identifier. The scanner was created to identify people in medical settings without having to clean scanners of potential COVID contamination.

The project, funded by the Swiss innovation agency Innosuisse, delivered a contactless biometric sensor platform that scans both palm veins and palm prints. The sensor platform collected a new dataset for research into vein biometrics accuracy and fraud attacks.

Global ID teamed up with humanitarian organization Project HOPE to introduce finger vein authentication for health and emergency response interventions in 2024. The technology was introduced in Namibia as part of a PEPFAR and USAID-funded project aimed at reducing HIV infections and improving health care.​

Earlier this year, Global ID was awarded a new U.S. patent for its BioID security protocol, following another U.S. patent awarded to the firm in March 2025.

“Unlike fingerprints or facial recognition, vein biometrics relies on analyzing the internal blood vessel pattern of the hand or finger. Invisible to the naked eye, unique to each individual, and extremely difficult to replicate, this biological marker is considered one of the most reliable to date,” the company said during the announcement.

Article Topics

biometric authentication  |  biometrics  |  finger vein  |  Global ID  |  Idiap  |  patient identification  |  Swiss Center for Biometrics Research and Testing  |  vein biometrics