Techn. Fakultät Willkommen am Institut für Informatik FAU-Logo

Dipl.-Phys. Samuel Reinfelder

Alumnus of the Pattern Recognition Lab of the Friedrich-Alexander-Universität Erlangen-Nürnberg

Research Interests

In the past physicians could rely for a diagnosis of patients only on few measureable parameters like the pulse. Nowaday medical engineering offers a wide range of technical devices for diagnosis applications. Recent advances in MEM-technology allow the use of cheap, small, wearable magnetic-inertial measuring units. These devices can be used to calculate the position and orientation of body segments. Engineers can use this data to unobtrusive measure movements of the patient without disturbing the natural motion sequence.

In the special case of gait analysis, parameters like step length, step height, step time and many more can be extracted. These features can be used to classify  movements and to distinguish between normal and abnormal movements. These devices can collect data over several hours or even days. Unsupervised data mining techniques can provide meaningful data for the physician which can base the diagnosis on more data or adapt the therapy based on this knowledge. In the scope of this research area I am especially interested in the following ideas:

    • Unsupervised monitoring of Parkinson's patients
    • Identification of gait phases of Parkinson's patients
    • Magnetic-inertial sensor orientation/ position estimation
    • Kinematic data analysis
    Telemedical Monitoring of Nutrition & Activity of Nutritional Disorder Patients
    • Analysis of clinical disorders using telemedical applications is a current hot topic in computer science and biomedical engineering. Our team of biomedical researchers at the Faculty of Engineering and nutrition experts at the Faculty of Medicine are working in a joint project.

      Our goal is to develop a novel telemedical monitoring concept for patients with nutritional disorder to allow an individualized, quantitative, and qualitative assessment of the disorder. In a first step, we are developing a smartphone application that helps treating physicians to better interface with patients, and that at the same time collects data about the ambulatory nutrition status of the patients. The developed Android app will be connected to additional sensors for activity recognition and assessment of the physical activity.

      In a parallel approach, we are investigating individualized intervention concepts that build on this telemedical application based on a Joomla web server.


      This work is funded through a research grant and is co-supervised by Prof. Dr. Bjoern Eskofier (Faculty of Engineering) and Prof. Dr. med. Yurdagül Zopf (Faculty of Medicine).

    Posturography sensor sole
    • Posturography is a technique for quantifying the postural sway during standing (Static Posturography) or walking (Dynamic Posturography). Patients can be inspected visually or measured with technical devices. Postural plates are used in clinical practise for measuring the center of pressure (COP) and COP based parameters. The disadvantage is that measurements need to done in a lab environment. Changes of the postural control during the day are only detectable with recurrent measurements during the day.

      Postural insoles are more portable and can be worn by a patient during the hole day. A novel postural insole was validated in this project against a postural plate. Special focus was on the preprocessing of the raw data.



      This work was funded by the Emerging Fields Initiative (EFI), Friedrich-Alexander University Erlangen-Nürnberg. It is part of the EFIMoves project.


      Articles in Conference Proceedings
      Reinfelder, Samuel; Durlak, Felix; Barth, Jens; Klucken, Jochen; Eskofier, Björn
      Wearable Static Posturography Solution Using a Novel Pressure Sensor Sole
      Engineering in Medicine and Biology Society (EMBC), 2014 36th Annual International Conference of the IEEE (36th Annual International Conference of the IEEE EMBS), Chicago, Illinois, USA, August 26-30, 2014, pp. 2973-2976, 2014 (BiBTeX, Who cited this?)
    Instrumented Timed Up and Go test
    • The Timed Up and Go (TUG) mobility test is a widely used performance assessment in the treatment of elderly and patients with movement disorders. It rates the functional ability to walk and is an important tool for assessing balance deficits, risk of falling and disease stages of movement disorders e.g. Parkinson's disease.

      The TUG test is performed in the following manner: The patient starts sitting on a chair, stands up, walks 3 m, walks back to the chair and sits down again. Although complex movements like standing up, turning or sitting down are required for the TUG test, the only parameter taken is the required time for the procedure.

      The aim of this project is to automatically segment the test in several subphases. Features describing these parts are used to classify the patient more precisely which may lead to a better understanding of differences or agreement of the course of motion between the patients. The timings of the subphases can also be used for extracting medical relevant information like turning strides or freezing events.

      A trial study of 16 Parkinson's disease patients was conducted, resulting in an overall sensitivity of 82% over all phases. This work is continued with further pattern recognition methods as well as applications in the analysis of previously, unlabeled TUG tests.

      This work is funded by the Emerging Fields Initiative (EFI), Friedrich-Alexander University Erlangen-Nürnberg. It is part of the EFIMoves project.


      Articles in Conference Proceedings
      Reinfelder, Samuel; Hauer, Roland; Barth, Jens; Klucken, Jochen; Eskofier, Björn
      Timed Up-and-Go phase segmentation in Parkinson's disease patients using unobtrusive inertial sensors
      Conference Proceedings (37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society), Milan, Italy, 25.08.2015, pp. 5171-5174, 2015 (BiBTeX, Who cited this?)
    Evaluation of Torticollis patients
    • Spasmodic Torticollis (Cervical Dystonia) is a neurological movement disorder causing the neck to involuntarily turn left, right, upwards, and/or downward. The most common rating scale is the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) which is based on a questionnaire and on visual assessment by a physician. The rating is subjective and only covers static movement parameters.

      Parameters based on motion capture data are obejctively calculated and can cover dynamic movements. The aim of this project is to preprocess the data and to extract meaningful parameters out of motion capture data.