Periodic Reporting for period 1 - SocketSense (Advanced sensor-based design and development of wearable prosthetic socket for amputees) Reporting period: 2019-01-01 to 2020-12-31 Summary of the context and overall objectives of the project The H2020 project SocketSense aims at developing an innovative prosthetic socket system with integration of advanced sensing, data analysis, AI methods, embedded edge and cloud computing. It is motivated by the fact that limb amputations often cause serious physical disabilities and thereby compromise the quality of life of many people around the globe. The World Health Organization estimates that there are ~40 million amputees in the world. One common reason for limb loss is diabetes. One undesired current situation is that many amputees do not wear their prostheses because of discomfort or residual limb pain. Consequently, their overall life quality is often negatively affected. For example, disabled people are at a more severe risk of unemployment as they begin from precarious positions and usually meet more barriers in finding new jobs. The lack of employment significantly affects the perceived social exclusion index. The levels of social exclusion are the highest among the long-term unemployed and those unable to work for reasons of poor health or disability.The key cutting-edge is given by the support for effective monitoring of dynamic operational conditions within prosthetic sockets that are traditionally not directly observable. The measurement values are evaluated against biomechanical models based on the residual limb tissue properties of individual patients. The result of that comparison may then indicate that the socket fit is non-optimal and that fabricating a better fitting socket may be necessary. See the figure below for the overall system architecture of SocketSense. With SocketSense, the prosthetists will be able to achieve an optimised socket within the same day when the patient needs a renewal, and the technique will apply to all lower limb amputees (above knee and below knee). See the figure below for the overall architecture. Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far The work that has been conducted so far focusing on the problem definition, investigation of candidate methods and technologies, system architecture conceptualization, sensor design and optimization, the establishment of an embedded system platform for the data processing and communication, and the preparation of clinic procedures for data collection and system validation.The results achieved so far include the finalization of component design and the accomplishment of the first prototype system integrating the sensors, sensor electronics, firmware and application software. Major progress has been made in the following respects:- On the senor system development: The final choice for optimal sensel design has been made with the prototypes manufactured and evaluated. To support the decision, extensive lab tests have been carried out to collect data for performance evaluation. Meanwhile, the electronic system for sensor data acquisition, processing and communication has been updated and implemented. On the software side, streaming data via Internet communication to the remote influxDB database was implemented and tuned.- On the data analysis: The work on data analysis emphasizes a systematic characterization of operational features and identification of models and algorithms for data treatment and operational condition inference. The work is still ongoing. The initial design for smart sensor data analysis includes MLP (Multi-Layer Perceptron) and HMM (Hidden Markov Model) for accounting for the sensor and biomechanical uncertainties. To allow (re)generating the intra-socket pressure conditions without direct clinical trials, the OpenSim tool has been used to capture the overall operational conditions of amputee behaviour, and then the AnSys tool has been employed to elicit the pressure map inside socket using Finite Element Analysis (FEA). To support the data analysis, dynamic testing and sensor deployment, a Steward Manipulator has been built up. The design is currently being refined for a hybrid force-motion control. - On the comfort assessment and socket optimization: A protocol for soft tissue measurement, including a simulation-based method in Abaqus tool for tissue sensitivity analysis, has been established. See the figure below. The work on virtual socket generation has been finished. Currently, we are also continuing with the design of a Decision Support System (DSS) based on a fuzzy-logic inference engine (IE) to support prosthetists in applying socket rectification actions.- On the overall system integration: We have also developed a reference scheme for sensor deployment inside socket, allowing the mapping of measurement data across 2D/3D representations. The design of a new ICT platform for prosthetic clinics has also been refined. The key content includes a backend system for data storage (influxDB), data management and computation, and a frontend system for the 3D visualization and data presentation.Moreover, regarding the clinical trial design, a Protocol and application form for ethics (in both Spain and the UK), as well as other trial documents such as the participant information sheet, have been completed. Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far) The approach taken by SocketSense is highly multidisciplinary, characterized by the integration of a wide range of topics beyond state-of-the-art, relating to- Novel flexible sensors for real-time monitoring of operational conditions.- AI based estimation of intra-socket operational conditions that are inherently stochastic and only partially observable.- Advanced biomechanical models for limb diagnostics and prognostics, and socket design optimization.- Integrated edge - and cloud-based software services for effective data processing and communication.- Clinic data collection and IoT system validation.