Aignostics started in 2018 within the walls of the Berlin Institute of Health at Charité – Universitätsmedizin Berlin, before ultimately spinning-out in 2020. But did you know that Charité Berlin has a rich history dating all the way back to 1710? Let’s explore some of that history, along with the history of pathology’s digital transformation, to understand the origins and drivers of Aignostics’ founding.
History of Charité – Universitätsmedizin Berlin
Charité – Universitätsmedizin Berlin is one of the largest university hospitals in Europe. It was originally founded in 1710 as a quarantine hospital before being converted into a charity hospital for impoverished communities. By 1828, it had grown to include a medical school which was integrated into Humboldt University as a teaching hospital. Many influential scientists have been affiliated with the hospital in its 300+ year history, including Rudolf Virchow, known as “the father of modern pathology” and the namesake of our first foundation model, RudolfV.
Today, Charité Berlin is regarded as one of the premiere hospitals in Europe and sees over 900,000 patients annually. In 2013, the Berlin Institute of Health (BIH) was established at Charité Berlin with the goals of advancing research into major diseases and serving as a hub for translational medicine. This hub has grown to include a Digital Health Accelerator (DHA), whose mission is to help clinicians within the Charité Berlin and BIH system turn their research results into digital health solutions. Within the past eight years, DHA has spun-out more than 10 companies, including Aignostics.
Timeline of AI in Pathology
While AI originated in the 1950s, it wasn’t until the past few decades that AI has been transformatively applied to medicine due to improvements in computing power and availability of large, high-quality medical training datasets.
Histopathology has adopted AI more slowly than fields like radiology due to differences in digitization timelines. Radiology embraced digital workflows in the early 1990s with the development of the DICOM (Digital Imaging and Communications in Medicine) standard in 1993, a universal protocol that defines how medical images are stored, transmitted, and processed. The implementation of PACS (Picture Archiving and Communication Systems) soon followed, which established networks for storing, retrieving, and displaying digital medical images. These advances allowed for the transition of radiology imaging from film-based to fully digital workflows.
On the other hand, pathology did not begin its digital workflow transformation until around 1999 with the introduction of whole slide imaging technology. This later digitization, combined with significantly higher data complexity and size of pathology images relative to other radiology images (gigapixels vs megapixels), has delayed the timeline of AI integration into histopathology.
When AlexNet, a convolutional neural network (CNN) architecture, won the ImageNet Large Scale Visual Recognition Challenge in 2012, it demonstrated an unprecedented accuracy in image-based machine learning approaches to train downstream models, proving that deep learning could effectively analyze complex visual data at scale. This served as a catalyst to give pathology researchers confidence that a similar approach could be applied to histopathology images. Aignostics was founded within this evolving technological landscape, with the goal of addressing computational pathology's specialized needs—combining advanced machine learning with pathological expertise to extract meaningful insights from digital tissue samples.
Founding of Aignostics
Dr. Frederick Klauschen (at the time a Professor at Charité – Universitätsmedizin Berlin, now at Ludwig Maximilian University of Munich) and Dr. Klaus-Robert Müller (Professor at Technische Universität Berlin), two of Aignostics’ co-founders, have been at the forefront of innovation when it comes to AI in pathology. Recognizing the possibility of a global shortage of pathologists amid rising cancer cases and an aging population, Dr. Klauschen and Dr. Müller began collaborating to explore AI applications in pathology as early as 2011, when image-based machine learning was still a novel idea. They filed their first patent in 2012 for a method of automatically analyzing biological sample images to determine regions of biological significance, such as cancer.
Within a few years, their innovative research caught the attention of the BIH, which invited them to join the DHA program to support and grow the research they were pursuing. Aignostics was formally established within the accelerator program in 2018, with Viktor Matyas and Dr. Maximillian Alber completing the founding team while Dr. Klauschen and Dr. Müller continued as Lead Advisors. In 2020, Aignostics successfully spun-out of DHA and incorporated.
In the last five years, Aignostics has grown to a diverse team of over 120+ biologists, pathologists, machine learning engineers, and other experts, committed to transforming drug development and improving patient outcomes with AI. Most recently, Aignostics announced a strategic collaboration with Mayo Clinic to develop foundation models and clinical products for pathology. Looking forward, Aignostics plans to build new foundation models, launch “plug-and-play” research products, and expand its offerings in target and biomarker identification, translational research, and digital diagnostics.