Adaptive / Context Sensitive Data Provision in Mobile Applications
Mentored by Jochen Hampe, Karsten Wendt
at Chair of Software Technology, Andreas-Pfitzmann-Bau, TU Dresden and secondary at: Clinic 1 / EKFZ for Digital Health, UKD
From data perspective, current medical environments can be characterized as followed: Clinical data is digital, complex, heterogeneous, large and distributed, but forms the base of daily medical business. The information can be provided by printed paper or numerous and heterogeneous devices, e.g. screens, mobile handhelds, headphones or other wearables. On the other side, a lot of different possibilities for data capturing and control exit, such as PCs, wearables with touch screens, microphones, cameras, other sensors, or the scanning of handwritten notes. The constraints when and how to use these input and output potentials are defined by the actual user scenarios, e.g., by physical or functional availability, hygienic restrictions, hands-free actions or confidential or emergency settings.
These circumstances result in different obstacles during the working processes: Crucial data are elaborate to allocate, data input is time-intensive, there exist cumbersome analog-digital barriers and many different interactions points due to inappropriate technical integration and distributed systems, which result in many redundancies, repetions and thus, unnecessary mental load and time costs during data acquisition and input.
Hence, the vision is make data interaction smarter, i.e. adaptive, context-sensitive, preemptive, learning, mobile and lean. This should be based on new software layer between the actual data provision and end user devices, utilizing recent software and ML technologies to a) provide data automatically in the right form at the right time to the right place and b) capture user input, enhance it by context information and delegate it to the correct data sink.
The actual tasks will be:
- Design & develop general purpose models / adapt architectures to cover such smart data interaction based on Wearables and Machine Learning for different use cases and related research projects (see work environment)
- Create evaluable prototypes based on the models and architectures for different use cases
- Enforce synergy and transition effects, focus software quality
- Produce publications
- Related research questions:
- Analyze current state of the art (data provision wearables, ML for context and reinforcement learning, data mapping to and from heterogeneous user devices, data-driven software design)
- Retrieve & abstract requirements from associated / perspective research projects
- Derive patterns, models, methods and adapted architectures for such system classes
- Provide general purpose software framework as Open Source
Work environment
The work will be conducted in close collaboration with software and medical experts at the chair of software technology and the EKFZ for Digital Health. This constellation will provide an unique, interdisciplinary team of technical and medical researchers as well as practical experts in their daily work. The position will be related to the following and not exclusive ongoing research projects:
- MAINFRAME: Computer vision for hematology
- DigiPhenoMS: AI for Multiple Sclerosis patient treatment
- DECISIONS: Genome similarity detection
- VIPFLUID: Preemptive Monitoring and Anomaly Detection
Prerequisites
- Master’s Degree (or equivalent) in electrical engineering, computer science, applied mathematics or related fields of expertise
- Software designing skills
- Programming (e.g. PyTorch, Tensorflow, Keras, Python, Java)
- skills and practical experience in one or more of the following research areas is beneficial:
- Machine Learning
- Software Architectures
- Designing of data-driven Systems
- Wearables / IoT
- Basic understanding of medical requirements, procedures and restrictions
- Ability to collaborate well in an interdisciplinary environment
Topics All Topics
Adaptive / Context Sensitive Data Provision in Mobile Applications
AI Controlled Wireless Mesh Networks for Medical Applications in Crisis Situations (AWESOME)
Artificial Intelligence in the Context of Biomedical Signals
Explainable AI for reliable computer assisted surgery
External Systems Biology Knowledge Integration in Large Language Models
Features of Fitness Landscapes
From protein structure prediction to explainable AI in Oncology
GeoSOP: Geometry for surgical outcome prediction
Natural Neural networks inspire Artificial Neural Networks in Therapeutic Design
Reinforcement Learning for Mechanical Ventilation
StochasticAI – Integrating Stochastic Modeling with AI
SuperTouch - Data-driven tactile sensing for robot-assisted cancer surgery
Teaching artificial intelligence conformational dynamics of proteins
The Use of AI by the Civil Service in Light of the Principles of Democracy and the Rule of Law