Healthcare

We are developing an intelligent decision-making aid for discharge planning in case management and are supporting our clients with integrating data-based recommendations for action within their patient discharge process. Within this framework, we are implementing statistical planning algorithms and self-learning models that allow lengths of stay and discharge data to be estimated with precision based on relevant treatment parameters. These are then embedded in the underlying processes of bed scheduling, capacity planning and the planning of ward utilisation. Besides technical development work on the models and creating technical concepts and checking their plausibility together with the client, this also includes integrating our models into the clinic’s system landscape and IT infrastructure.

A manufacturer of medical inspection machines commissions d-fine to design an AI solution for the optical inspection of medicine containers. Using a highly efficient training pipeline, performance benchmarks for various deep learning methods are systematically examined and compared with classical methods from machine learning.
Based on the results for recognition quality and data throughput, a real-time capable solution is created for deployment on the company's machines. Further components of the system are a module for visualising and explaining the AI results to the operator and a software framework based on open source for training own neural networks.

An international healthcare organization, which delivers large medical portfolios for infectious diseases, appoints d-fine to optimize the supply chain networks regarding storage and delivery costs, quality of service, robustness, and future demands. Via data-driven insights and qualitative factors, d-fine analyses the capacity utilization of warehouses on all tiers of the network. This includes the design of best options for long-haul delivery and the costly last-mile distribution, supporting procurement to 3rd party logistics providers via tailored pricing models and business intelligence. We tailor our methodology to the supply chain’s maturity, ranging from variable stockholdings for fast- and slow-moving products to incoming and outgoing volatility analyses, in order to reach the best possible balance between cost and service-level. d-fine enables maximum effectiveness of the client’s healthcare distribution system by analytical methods, to meet medical, social and business needs in the fight against infectious diseases.

The regulatory playing field in the use of new technologies such as Artificial Intelligence and Machine Learning is evolving – including overarching activities like the European AI act, and also sector specific guidance in terms of the 2nd revision of the GAMP 5. Heading the GAMP D-A-CH Special Interest Group on GxP Compliant Use of AI, we pool knowledge and expertise to establish best practices in pharma with a focus on trustworthiness and quality in a risk-based manner. Our three core publications in the ISPE Pharmaceutical Engineering summarize our results – shaping the approach to AI projects and governance across the pharma industry globally:

• AI maturity model
• AI Governance & QA Framework
• The road to excplainable AI

For a European research institute, d-fine is building a genome analysis pipeline for artificial intelligence technologies on a scalable cloud infrastructure. In this public-private partnership, we are developing, training and evaluating machine learning algorithms that model an organism's specific responses to a vaccine's active ingredient and to infection from one of the leading viral pathogens in zoology. Our goal is to gain a deeper understanding of the mechanisms and cellular pathways and to provide innovative bioinformatics tools that will enable the identification of crucial gene patterns and new targets. For our customers, this is an important step towards targeted therapies and data-driven drug design.

To this end, we are building a state-of-the-artMLOps platform for processing data from scRNA sequencing, including explainable AI methods that follow operational best practices. The d-fine team members in this project combine deep scientific knowledge in the life sciences with extensive technological expertise.

A research institute initiated the project touch.connect.care, to transition the use of individual biography based care of the elderly to the next, digital level. It does so by sourcing d-fine’s technology expertise, partnering with a senior residence operator. 

Applying co-creation as the product design process together with an agile project management model for software and cloud engineering, we have realized the product vision with the appropriate, cloud-native technology stack. This product will foster valuable social interactions and sustain the quality of life in senior residences.

Building upon our responsible cloud strategy, our solution both complies with data protection of sensitive personal biography data and is available for a wide range of parties in the healthcare sector – small and large alike.

MedTech companies and contract research organizations (CROs) appoint d-fine to develop information intelligence platforms, which automate manual workflows and provide business-relevant insights from large data sources. These platforms integrate, analyze, and visualize post-market data and clinical trial data from various sources, such as database systems from medical regulatory bodies (BfArM, FDA, Health Canada, etc.), the primary WHO registries for clinical trials, PubMed medical journal articles, MAUDE or FDA.

The platforms offer smart user interfaces, which are customized to business needs: Interactive searches, relevance rating of search results, and search algorithm learning by user feedback. This continuously enhances the user experience and improves business advantages. The incorporated explainable AI methods (xAI) make search results comprehensible. Moreover, clients have full control over their platform systems including access to metadata and ownership of the source code. 

The information intelligence platforms enable d-fine’s clients to strengthen their product and services portfolio in the market by smart transformation of information into clinical evaluation and decision-making.

Sustainability does not stop at the pharmaceutical & healthcare sector. Reporting according to CSRD as well as subsequent sustainability strategies are not only a central topic for the pharmaceutical industry, but also for medical technology, wholesale and even many service providers.

Together with our clients from the pharmaceutical industry and the area of reportable health care facilities, we develop concepts for the implementation of a sustainability strategy taking into account requirements according to the CSRD. In numerous projects on this topic, it has proven successful for d-fine to identify fields of action (topic clusters) that are relevant in view of the EU taxonomy and the CSRD as well as in consideration of other reporting standards (e.g. Global Reporting Initiative) in close exchange with the client's experts. This enables the identification of "gaps" and a subsequent prioritisation and grouping based on various relevant criteria (e.g. regulatory and customer-strategic relevance). In a systematic roadmap, we consolidate measures and define first steps to achieve reporting capability and develop a sound project plan that ensures the customer’s closure of the "gaps" based on the previous prioritisation.

We support billing service providers in the German healthcare sector in modernising their heterogeneous application landscape in order to be able to efficiently process billing and factoring services.

For this purpose, we help to create a future-proof platform operated in the private cloud, which provides internal and external applications in modular services: From customer portal and contract management,  accounting and receivables management to credit risk management and regulatory reporting. Flexible interfaces and highly automated processes leverage synergies between departments while addressing the specific needs of billing and factoring in different healthcare sectors; this also provides the opportunity to bring new products to market in record time.

The integrated d-fine approach of design, implementation and project management from a single source enables us to meet the complex business requirements, while maintaining the high quality and reliability requirements for applications in the financial environment as well as the demands for data protection, with regards to patient and treatment data, whilst keeping on target from budget and schedule perspectives.

The processes and methods within the institutions of the public health system in Germany are currently experiencing a digital "push" through national funding initiatives. Our proven concept for a contemporary digitisation step, for example of a health department, builds on our cross-sector experience in digital change management. Central elements in our structured process are the analysis of the status quo, the development of a target image together with the client and then the careful realisation of the change. It is essential to involve all those affected, including a training and sensitisation concept. Only in this way can all those involved come together constructively from the start, and the change to a safer, more effective and more digital working world succeed – for the benefit of those directly affected and all participants in the health system: all citizens.