Philipp Rostalski

Philipp Rostalski

Auto-generated profile, based on publicly available information

Philipp Rostalski

Fraunhofer Research Institution for Individualized Medical Technology and Engineering, IMTE, Luebeck, Germany

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 May 2022 in IEEE Access
Verified
Journal Article
Open access
EPEike Petersen
Eike Petersen
YPYannik Potdevin
Yannik Potdevin
EMEsfandiar Mohammadi
Esfandiar Mohammadi
SZStephan Zidowitz
Stephan Zidowitz
SBSabrina Breyer
Sabrina Breyer
CHChristian Herzog
Christian Herzog
Machine learning is expected to fuel significant improvements in medical care. To ensure that fundamental principles such as beneficence, respect for human autonomy, prevention of harm, justice, privacy, and transparency are respected, medical machine learning systems must be developed responsibly. Many high-level declarations of ethical principles have been put forth for this purpose, but there is a severe lack of technical guidelines explicating the practical consequences for medical machine learning. Similarly, there is currently considerable uncertainty regarding the exact regulatory requirements placed upon medical machine learning systems. This survey provides an overview of the technical and procedural challenges involved in creating medical machine learning systems responsibly and in conformity with existing regulations, as well as possible solutions to address these challenges. First, a brief review of existing regulations affecting medical machine learning is provided, showing that properties such as safety, robustness, reliability, privacy, security, transparency, explainability, and nondiscrimination are all demanded already by existing law and regulations—albeit, in many cases, to an uncertain degree. Next, the key technical obstacles to achieving these desirable properties are discussed, as well as important techniques to overcome these obstacles in the medical context. We notice that distribution shift, spurious correlations, model underspecification, uncertainty quantification, and data scarcity represent severe challenges in the medical context. Promising solution approaches include the use of large and representative datasets and federated learning as a means to that end, the careful exploitation of domain knowledge, the use of inherently transparent models, comprehensive out-of-distribution model testing and verification, as well as algorithmic impact assessments.

10 February 2020 in IEEE Access
Verified
Journal Article
Open access
EPEike Petersen
Eike Petersen
JSJulia Sauer
Julia Sauer
JGJan Graßhoff
Jan Graßhoff
PRPhilipp Rostalski
Philipp Rostalski
Electromyographic (EMG) measurements of the respiratory muscles provide a convenient and noninvasive way to assess respiratory muscle function and detect patient activity during assisted mechanical ventilation. However, surface EMG measurements of the diaphragm and intercostal muscles are substantially contaminated by cardiac activity due to the vicinity of the cardiac muscles. Many algorithmic solutions to this problem have been proposed, yet a conclusive performance comparison of the most promising candidates currently is missing. The objective of this work is to provide a quantitative performance comparison of six previously proposed cardiac artifact removal algorithms operating on singlechannel EMG measurements, and two newly proposed, improved versions of these algorithms. Algorithmic performance is evaluated quantitatively based on four different measures of separation success, using both synthetic validation signals and electromyographic measurements of the respiratory muscles in eight subjects. The compared algorithms are two versions of the empirical template subtraction algorithm, two model-based Bayesian filtering procedures, a wavelet denoising approach, an empirical mode decomposition (EMD) based approach, and classical high-pass filtering. Different algorithms perform well with respect to different performance measures. Template subtraction algorithms yield the best results if the characteristics of the raw signal are of interest, while filtering algorithms like simple high-pass filtering, wavelet denoising, and EMD-based denoising show superior performance for calculating a cleaned envelope signal. No algorithm completely removes the cardiac interference, but the characteristic errors introduced by the considered algorithms differ. Hence, the choice of the algorithm to use should be made depending on the target application. Finally, we also demonstrate that our empirical SNR measure, which can be calculated without knowledge of the true, undisturbed signals, correlates strongly with the exact reconstruction error. Thus, it represents a reliable indicator for algorithm performance on real measurement data.

7 July 2023 in Medical Image Analysis
Verified
Journal Article
Closed access
ABAlexander Bigalke
Alexander Bigalke
LHLasse Hansen
Lasse Hansen
JDJasper Diesel
Jasper Diesel
CHCarlotta Hennigs
Carlotta Hennigs
PRPhilipp Rostalski
Philipp Rostalski
MHMattias P. Heinrich
Mattias P. Heinrich

2 February 2024 in IEEE Sensors Journal
Verified
Journal Article
Open access
RKRoman Kusche
Roman Kusche
AOAndra Oltmann
Andra Oltmann
PRPhilipp Rostalski
Philipp Rostalski
The reliable detection of muscle contractions in real-time is important for many applications. Both for prosthesis control and in the field of human-computer interaction, the physiological commands of the user must be recognized. However, conventional methods such as the electromyography are susceptible to interferences. A particularly robust method is the electrical impedance myography. Especially the temporal changes of the bioimpedance phase response are of interest for muscle activity monitoring. However, available wearable measurement systems are not capable of detecting muscle contractions in real-time to control prostheses or human-computer interaction devices. This work presents the development and metrological characterization of a wearable real-time bioimpedance spectrometer for the detection of muscle contractions. It can record the frequency responses in the range of 20-230 kHz from two antagonistic muscles. The sampling rate of 25 impedance spectra per second and per channel provides sufficient temporal resolution for many applications. Phantom measurements show that the statistical errors are below 1 % for the magnitude and below 0.4° for the phases, which is sufficient for electrical impedance myography. This system is used to perform first subject measurements. For the first time these measurements demonstrate the temporal impedance behavior and frequency responses of two antagonistic muscles during contraction. In addition, the directional dependence of the electrical impedance myography during a muscle contraction is investigated for the first time. The presented measurement system and novel measurement approaches are promising for many electrical impedance myography applications, especially for reliable muscle contraction detection e.g. in prosthetics or human-computer interaction applications.

4 July 2022 in IEEE Transactions on Biomedical Engineering
Verified
Journal Article
Open access
JGJan Graßhoff
Jan Graßhoff
EPEike Petersen
Eike Petersen
SWStephan Walterspacher
Stephan Walterspacher
PRPhilipp Rostalski
Philipp Rostalski
Objective: The quantification of inspiratory patient effort in assisted mechanical ventilation is essential for the adjustment of ventilatory assistance and for assessing patient-ventilator interaction. The inspiratory effort is usually measured via the respiratory muscle pressure ( P $_{\mathsf{mus}}$ ) derived from esophageal pressure ( P $_{\mathsf{es}}$ ) measurements. As yet, no reliable non-invasive and unobtrusive alternatives exist to continuously quantify P $_{\mathsf{mus}}$ . Methods: We propose a model-based approach to estimate P $_{\mathsf{mus}}$ non-invasively during assisted ventilation using surface electromyographic (sEMG) measurements. The method combines the sEMG and ventilator signals to determine the lung elastance and resistance as well as the neuromechanical coupling of the respiratory muscles via a novel regression technique. Using the equation of motion, an estimate for P $_{\mathsf{mus}}$ can then be calculated directly from the lung mechanical parameters and the pneumatic ventilator signals. Results: The method was applied to data recorded from a total of 43 ventilated patients and validated against P $_{\mathsf{es}}$ -derived P $_{\mathsf{mus}}$ . Patient effort was quantified via the P $_{\mathsf{mus}}$ pressure-time-product (PTP). The sEMG-derived PTP estimated using the proposed method was highly correlated to P $_{\mathsf{es}}$ -derived PTP ( $\mathsf{r}=\text{0.95}\pm \text{0.04}$ ), and the breath-wise deviation between the two quantities was $-\text{0.83}\pm \text{1.73}\,\text{cmH}_\text{2}\text{O}\text{s}$ . Conclusion: The estimated, sEMG-derived P $_{\mathsf{mus}}$ is closely related to the P $_{\mathsf{es}}$ -based reference and allows to reliably quantify inspiratory effort. Significance: The proposed technique provides a valuable tool for physicians to assess patients undergoing assisted mechanical ventilation and, thus, may support clinical decision making.

15 December 2023 in IEEE Access
Verified
Journal Article
Open access
MSMarlin Siebert
Marlin Siebert
JGJan Graßhoff
Jan Graßhoff
PRPhilipp Rostalski
Philipp Rostalski
Diabetic retinopathy (DR) is a leading cause of vision loss. Therefore, screening for early signs and assessment of the DR severity is crucial and extensively studied. To support clinicians, screening could be automated by algorithms that can, for example, refer difficult decisions to specialists for further investigation. However, frequently used neural networks (NNs) typically do not know when they do not know and approximate Bayesian NNs often equally do not suffice to provide well-calibrated uncertainty estimates. Thus, in this work, we investigate whether and how to use deep kernel learning (DKL) which we designed as a hybrid combination of the state-of-the-art EfficientNet-B0 and a Gaussian process (GP) layer to improve the quality of uncertainty estimates in referral-based DR screening. To this end, we first analyze the necessity for recently proposed extensions to the DKL framework to resolve miscalibrated uncertainties, despite the theoretical superiority of GPs to uncertainty quantification. Our subsequent comprehensive comparison of the curated DKL’s performance to that of the most common approximate Bayesian NNs shows our DKL models to particularly improve the detection of near out-of-distribution (OOD) samples containing other eye diseases through epistemic uncertainty information, but also enhance the calibration of aleatoric in-distribution uncertainty and diagnostic performance. Hence, it can provide a substantial benefit for safety-critical medical applications, like automated DR screening, particularly by potentially reducing the risk of missing diseases other than DR due to the improved near-OOD detection performance.