The BigMedilytics Blueprint, is an interactive website, which aims at sharing the main findings of our project. The Blueprint describes the key enablers needed to drive the uptake of Big Data healthcare solutions, namely (a) business modeling, (b) performance modeling, (c) technical aspects, (d) legal, privacy and ethical aspects and (e) factors related to deployment at scale of Big Data technologies. It shows these aspects from five different perspectives which are represented by the five different bubbles below. With a bottom-up approach, starting from a fine-grained view of each single pilot (Pilot View), to a transversal overview about the different technologies and solutions across pilots (Transversal Aspects), to a high level and general learnings of the project (General Learnings). Next, those learnings are then mapped to important stakeholders of the health continuum (Stakeholder Perspective), and finally we show how the developed technologies could be applied to a fictitious patient journey (Patient Journey). All content is based on fine-grained information collected from the different pilots, but also by interactions between the different partners. More information can be found within the orange ABOUT bubbles.
Beside the Blueprint website, information can be also directly accessed by the different documents. These are particularly the Blueprint documents of each pilot (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12) which include a detailed overview about the different implementations. In addition to that the core of the Blueprint is described by the General Lessons Learned document and the CRISP DM for Healthcare document which should provide guidance to execute an AI and Big Data project in healthcare.
Navigate through the different perspectives, by clicking on one of the big bubbles. By clicking on the bubbles you can iteratively zoom until you eventually reach the final level in which you can see the assigned articles items within an overview.
The BigMedilytics Blueprint is intended to be for all stakeholders of the health continuum, particularly those who are involved or planning to setup a Big Data and AI project in healthcare. We suggest to spot the relevant pilots first, in order to read through the development, the challenges and learnings. To do so, have a look for the Key Information in the Transversal View.
In the BigMedilytics Blueprint you will find information on different levels, related to the 12 different pilots of our project. For each pilot you will find detailed information on their problems and their solutions, along with challenges, lessons learned and other outcomes, such as demos or publications. In addition to that you will also find high level learnings related to technical, legal, ethical and business aspects. Finally, the Blueprint presents options on how the developed solutions could be applied to a new scenario and aligns certain learnings to particular stakeholders.
The BigMedilytics Blueprint is a bottom-up approach, which analyses the project results from five different perspectives:
1. Pilot View. The Pilot View presents a fine-grained overview about the different BigMedilytics pilots including their implementations. Each pilot provides information about their task, how they solved difficulties, but also about barriers, lessons learned. Particularly the document contains information about the main technical building blocks, namely: Architecture, Data Processing, AI Components, Security and Privacy of Data Access and Processing, Trustworthy AI, as well as, System Interaction. Overall we cover 12 different pilots which are assigned to three different main themes, namely (1) Population Health and Chronic Disease Management, (2) Oncology and (3) Industrializing Healthcare Services. The different pilots with their name and assignment are shown below:
2. Transversal Aspects. The Transversal Aspects analyses the implementations, the outcomes and learning of the different pilots horizontally. According to different topics, contributions of each pilot to the Blueprint can be explored.
3. General Learnings. General Learnings is content which has been generated in close collaboration with different pilots. Different from the content in “Pilot View” and “Transversal Aspects”, the content presented here is brought to a more abstract and general level, in order to make it applicable to new scenarios. Cornerstones of General Learnings are the Lessons Learned, according to different aspects, such as for instance “General”, “Business” or “Ethical”. Moreover, General Learnings provides a general blueprint, how to setup a Big Data/Ai project in healthcare, applied to CRISP DM, the most widely-used open standard process model that describes common approaches.
4. Stakeholder Perspective. Within the Stakeholder Perspective we identify relevant players within the health continuum. Using the outcome of our project, we map the implementations and learnings to the different stakeholders, according to their possible interests. Therefore each stakeholder view can be used to explore corresponding information. For instance, from the perspective of a Data Privacy Officer, besides privacy related learnings, also topics such as Data Protection, Privacy Measures and Trustworthy AI might be of interest, of BigMedilytics. Overall, we identified the following stakeholders which are relevant to setup a Big Data/AI project in healthcare (alphabetical order): Clinical Staff, Data Scientist, Health Insurance, Hospital Decision Maker, Hospital IT/Equipment, Patient, Policy Maker, Privacy Officer
5. Patient Journey. The Patient Journey aims at showing how the developed technologies and learning of our project can be applied to new scenarios in healthcare. Particularly, how could the developments of BigMedilytics can support a possible new Big Data or AI project in healthcare? Or more specifically, considering it from the perspective of different stages of a patient journey, which parts of our project are relevant and how would it impact it? To do so, we describe a fictitious patient journey with a patient suffering a particular health condition undergoing different stages (e.g. general practitioner). Along with this exemplary patient journey we present some solutions of our project which help to improve the overall treatment.