1. General

Category

Category

Foundations, Philanthropists & Non-Profits

2. Project Details

Company or Institution

HIPPO AI

Project

Viktoria 1.0 @Hippo AI Foundation

General description of the AI solution

While most startups focus on AI models, we focus on solving the biggest challenge in AI; the availability of high-quality biodiverse medical data. As data is becoming increasingly valuable, the accessibility to high-quality datasets is decreasing.

We have founded the Hippo AI Foundation, an open-source foundation, that runs campaigns with patients, offers frameworks to build data and AI commons, and a new license model. This all with the focus to create the largest and most diverse data commons focusing on specific diseases.

To create access to good health care for all citizens, the Hippo AI Foundation puts patients and citizens in the focus of our actions. With your help and the help of patients who are suffering from specific diseases, we can democratize the development of Artificial Intelligence in healthcare, and allow local entrepreneurs across this globe to build independent AI systems.

The Hippo AI Foundation expects that compared to classic data projects, costs and time required can be drastically reduced by applying an open and community approach. This will benefit our organization, the medical community, and of course, patients.

Website

https://www.hippoai.org

Organisation

HIPPO AI gGmbH

3. Aspects

Excellence and Scientific Quality: Please detail the improvements made by the nominee or the nominees’ team or yourself if your applying for the award, and why they have been a success.

Type of AI: Open AI – our license model, is a license agreement intended to allow users to freely share, modify, and use this Database while maintaining this same freedom for others. The license explicitly states that it does not grant any patents of contributors, or derived AI models. All derivative works need to be published under the same license. AI models that are trained on the datasets of the Hippo AI Foundation need to be published (source code) under the same license in AI registries.

Quality: With Viktoria 1.0, an international top-class community led by the Hippo AI foundation is set to establish the first actionable database of breast cancer pathology data. With an interdisciplinary team of partners, external clinical experts, and Hippo AI community members the data aggregation, anonymization, and labeling are conducted. Artificial Intelligence is only as accurate—and therefore as effective—as the data it is trained with. To increase the availability of inclusive and comprehensive high-quality data commons, we use our funding to team up with leading trusted clinical researchers. We also use collective intelligence and performance incentives to achieve unprecedented levels of accuracy and effectiveness.

Describe the status of technology: We ran our successful first campaign at www.viktoriaonezero.org, which has attracted large data contributors of 7 different countries on 3 continents. This will lead to the most biodiverse open dataset in the area of breast cancer pathology. After we successfully close Viktoriaonezero.org we will publish all frameworks and methods under a CC-BY-SA license.

Scaling of impact to SDGs: Please detail how many citizens/communities and/or researchers/businesses this has had or can have a positive impact on, including particular groups where applicable and to what extent.

– The approach of the Hippo AI Foundation opposes the data colonialism that is handled by larger global corporates. Data can stay within the country, protecting its sovereignty. Similar to open-source software it allows local entrepreneurs to build local solutions, together with a global open-source community. SDG8, SDG9, SDG10, SDG11
– As current datasets in Medical AI systems are disproportionately built with data from just a few regions. Healthcare data lacks global diversity and is extremely white, which has a real-world impact. Our approach leads to global cooperation and the demonetization of the data leads to fewer information asymmetries between larger corporates and individuals. Protecting their fundamental rights and increasing access to health. SDG1, SDG3 & SDG5
– By creating access to high-quality datasets, we allow data scientists across the globe to train their skills and publish their results in competitive research papers. This supports SDG4 (good education).
– In the case of www.viktoriaonezero.org we aim to create higher access to early detection improve the specificity and sensitivity of breast cancer diagnosis using digital pathology diagnostics (focus on HER2 Gene Type Variant which is the diagnosis of Viktoria who is the breast cancer patient running this campaign)

Scaling of AI solution: Please detail what proof of concept or implementations can you show now in terms of its efficacy and how the solution can be scaled to provide a global impact ad how realistic that scaling is.

– Evidence of impact: www.viktoriaoenzero.org created a global community of physicians in Asia, Africa, Europe, and the USA that are collaborating within the framework given.
– The Hippo Foundation will publish all work under open source licenses to increase access and the potential replication of the initial project.
– The end-users are all ppl working in the field of medical AI. The Hippo AI Foundation does not build medical products but focuses on accelerating global collaboration, giving ppl access to global data and AI commons.
– The initial project – viktoriaonezero.org has attracted top clinical institutes that are contributing. The project also onboarded a top 20 pharmaceutical company that invests in data commons and supports the creation of AI commons.

Ethical aspect: Please detail the way the solution addresses any of the main ethical aspects, including trustworthiness, bias, gender issues, etc.

– The Hippo AI Foundation believes in a flourishing economy should be based on openness and collaboration, rather than on exclusive medical knowledge and exploitation, which leads to inequality. We foster open innovation for the medical industry and work against the monopolization and privatization of life-saving knowledge.
– The Hippo AI Foundation beliefs in the fundamental principles of reproducibility, therefore enforce the community to publish the course code and all derivates under the same license model. This is the only way to achieve a high level of replicability.
– The Hippo AI Foundation believes in the FAIR principles, Findability, Accessibility, Interoperability, and Reuse of the digital assets produced. These principles emphasize machine-actionability (i.e., the capacity of computational systems to find, access, interoperate, and reuse data with none or minimal human intervention) because humans increasingly rely on computational support to deal with data as a result of the increase in volume, complexity, and creation speed of data.
– The Hippo AI Foundation beliefs that citizens produce more than 95% of the data that make up our digital health universe, this data in healthcare can not become a tradable commodity. To maximize the benefits of medical Artificial Intelligence in healthcare, medical AI should create social value and become a common good.