News & Events

Functional requirements definition process

1 - March

During the first year of the project (2021), the five clinical partners worked hard to produce the Functional Requirements specifications. These aim to describe the core consortium clinical partners’ practical challenges and recognize opportunities, which form the basis of the Smart Hospital Care Pathway Engine AI solution development. The Functional Requirements document defines justified target states for information technology (IT), artificial intelligence (AI) and robotics partners. Technical partners can then define their AI solution technical requirements to match the concrete need of the clinical partners.

The current state analysis

The Functional Requirements work began with performing a current state analysis for each pilot. The current state analysis was carried out from February to April 2021. The process was supported by service designers from a subcontracted consultancy company Bitfactor. As a result of the current state analysis, the pilot partners created current state maps to provide an understanding of the current process phases, relevant actors in each phase and hospital IT systems relevant to each phase.

Mapping and identification of data sources were carried out during the current state modelling when the hospital IT systems of each process phase were identified. The data sources were mapped mostly from the clinicians’ perspective, which led to iterative discussions with each hospital’s IT department and AICCELERATE technical partners according to each clinical partners’ needs.

The user’s requirements

The user’s needs and requirements were described by clinicians in several ways. As a result, the user needs were presented in general and pilot specific solution charts, user personas, blueprints and some preliminary sketches of user interfaces. The user requirements were documented as a list that consists of epics and user stories. The epics and the user stories summarize the clinician’s ideas and expectations of an AI-enhanced system and describe what kind of visualizations and reports or functionalities they would like to have in the user interface of the system. The user stories are brief descriptions written from the clinician’s point of view without technical details. Furthermore, the high-level use cases were written to describe the interaction between actors (clinicians) and the system (SHCP engine).

Open exchange of ideas

The clinical partners have given their input to the functional specification documentation. To demonstrate the scalability of the SHCP Engine and AI model, three pilot cases are presented. Even though clinical partners have provided the main effort and content, there has been continuous dialogue between the consortium partners. The open exchange of ideas has ensured that the expected system functionalities specified in user needs and user stories can be translated into software solutions in the next stage of the project.

blurred figures of doctors and nurses in a hospital corridor

Based on the functional specification documentation, the technical partners of the consortium will define detailed technical requirements to create solutions to meet the needs of all three pilots. This work will be supported by the clinical partners. The next task is to start preparing for the SHCP Engine prototype testing. Clinical partners need to establish a required IT infrastructure in each hospital for the AICCELERATE data processing. Another point of focus is to enable data extraction from the clinical source systems. Once solution prototypes are available, those will be tested first in laboratory environments with retrospective data and later in real hospital environments with live data.


Authors from the OYS team

Timo Alalääkkölä · Janne Liisanantti · Riikka Lahtela
Merja Ahonen · Pasi Ohtonen · Minna Mäkiniemi

Similar Posts

Flecha hacia abajo EU model contractual AI clauses to pilot in procurements of...
Supporting public buyers in procuring AI-enabled solutions that are trustworthy, fair and secure is one of the aims of these...
See more
Flecha hacia abajo Artificial Intelligence’s potential to revolutionize healthcare
AI in healthcare is promising a lot, and in fact, we all are witnessing the exponential growth of AI applications...
See more
Arrow up