Peter U. Diehl

Freie Universität Berlin

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Matthew Cook

Australian National University

Johannes Schauwecker

München Klinik

Bruno U. Pedroni

University of California, San Diego

Guido Zarrella

Mitre (United States)

Wolfram Mittelmeier

University of Rostock

Emre Neftci

Forschungszentrum Jülich

Elias Sprengel

Paul Meyer-Rachner

Technische Universität Berlin

Hans Gollwitzer

Praxisklinik für Orthopädie, Unfall- und Neurochirurgie

Veit M. Hofmann

Humboldt-Universität zu Berlin

Cooperative Institutions

Technical University of Munich

Klinikum rechts der Isar

ETH Zurich

Ludwig-Maximilians-Universität München

SIB Swiss Institute of Bioinformatics

Klinik und Poliklinik für Orthopädie, Physikalische Medizin und Rehabilitation

München Klinik

University of Zurich

Charité - Universitätsmedizin Berlin

Heidelberg University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Non-intrusive deep learning-based computational speech metrics with high-accuracy across a wide range of acoustic scenes

PLoS ONE (2022)

Peter U. Diehl Leifur Thorbergsson Yosef Singer Vladislav Skripniuk Annett Pudszuhn

Speech with high sound quality and little noise is central to many of our communication tools, including calls, video conferencing and hearing aids. While human ratings provide the best measure of sound quality, they are costly and time-intensive to gather, thus computational metrics are typically used instead. Here we present a non-intrusive, deep learning-based metric that takes only a sound sample as an input and returns ratings in three categories: overall quality, noise, and sound quality. This metric is available via a web API and is composed of a deep neural network ensemble with 5 networks that use either ResNet-26 architectures with STFT inputs or fully-connected networks with wav2vec features as inputs. The networks are trained and tested on over 1 million crowd-sourced human sound ratings across the three categories. Correlations of our metric with human ratings exceed or match other state-of-the-art metrics on 51 out of 56 benchmark scenes, while not requiring clean speech reference samples as opposed to metrics that are performing well on the other 5 scenes. The benchmark scenes represent a wide variety of acoustic environments and a large selection of post-processing methods that include classical methods (e.g. Wiener-filtering) and newer deep-learning methods.

Benchmark (surveying)

10.1371/journal.pone.0278170

Cite

Citations (2)

High hydrostatic pressure-induced cell death in human chondrocytes and chondrosarcoma cells.

PubMed (2005)

Florian-Dominique Naal Karin Mengele Johannes Schauwecker Hans Gollwitzer L. Gerdesmeyer

In orthopedic surgery, sterilization of bone used for reconstruction of osteoarticular defects caused by malignant tumors is carried out in different ways. At present, to devitalize tumor-bearing osteochondral segments, extracorporal irradiation or autoclaving is mainly used, although both methods have substantial disadvantages, e.g. loss of biomechanical and/or biological integrity of the bone and destabilization of the articular surface. In this regard, high hydrostatic pressure (HHP) treatment of bone is a new, advancing technology, now being used in preclinical testing to inactivate tumor cells. To find out if this technique is also suited for extracorporal inactivation of chondrocytes and chondral tumor cells, the effect of HHP on cell viability and morphology of human chondrocytes / chondrosarcoma cells was investigated in the present study.SW1353 chondrosarcoma cells and chondrocytes were subjected to HHP in the range of 50 to 350 MPa (10 min, 37 degrees C) and, subsequently, cell viability and cell morphology assessed.After exposure at 350 MPa, all HHP-treated chondral cells showed explicit morphological changes, evident by membrane ruffling and bleb formation; chondrosarcoma cells treated this way were irreversibly damaged and not alive.We anticipate that, in orthopedic surgery, HHP eventually can serve as a novel, promising technical approach for cell inactivation (including tumor cells) and allow subsequent reimplantation of the osteoarticular autograft.

Hydrostatic pressure

Viability assay

Source

Cite

Citations (19)

Conversion of artificial recurrent neural networks to spiking neural networks for low-power neuromorphic hardware

Peter U. Diehl Guido Zarrella Andrew Cassidy Bruno U. Pedroni Emre Neftci

In recent years the field of neuromorphic low-power systems gained significant momentum, spurring brain-inspired hardware systems which operate on principles that are fundamentally different from standard digital computers and thereby consume orders of magnitude less power. However, their wider use is still hindered by the lack of algorithms that can harness the strengths of such architectures. While neuromorphic adaptations of representation learning algorithms are now emerging, the efficient processing of temporal sequences or variable length-inputs remains difficult, partly due to challenges in representing and configuring the dynamics of spiking neural networks. Recurrent neural networks (RNN) are widely used in machine learning to solve a variety of sequence learning tasks. In this work we present a train-and-constrain methodology that enables the mapping of machine learned (Elman) RNNs on a substrate of spiking neurons, while being compatible with the capabilities of current and near-future neuromorphic systems. This "train-and-constrain" method consists of first training RNNs using backpropagation through time, then discretizing the weights and finally converting them to spiking RNNs by matching the responses of artificial neurons with those of the spiking neurons. We demonstrate our approach by mapping a natural language processing task (question classification), where we demonstrate the entire mapping process of the recurrent layer of the network on IBM's Neurosynaptic System TrueNorth, a spike-based digital neuromorphic hardware architecture. TrueNorth imposes specific constraints on connectivity, neural and synaptic parameters. To satisfy these constraints, it was necessary to discretize the synaptic weights to 16 levels, discretize the neural activities to 16 levels, and to limit fan-in to 64 inputs. Surprisingly, we find that short synaptic delays are sufficient to implement the dynamic (temporal) aspect of the RNN in the question classification task. Furthermore we observed that the discretization of the neural activities is beneficial to our train-and-constrain approach. The hardware-constrained model achieved 74% accuracy in question classification while using less than 0.025% of the cores on one TrueNorth chip, resulting in an estimated power consumption of ≈ 17μW.

Neuromorphic engineering

Backpropagation

10.1109/icrc.2016.7738691

Cite

Citations (177)

Biomechanische Eigenschaften von Gelenkknorpel nach hydrostatischer Hochdruckbehandlung Biomechanical properties of articular cartilage after high hydrostatic pressure treatment

Biomedical Engineering / Biomedizinische Technik (2006)

Peter U. Diehl Florian-Dominique Naal Johannes Schauwecker E. Steinhäuser Stefan Milz

Ziel: Zu den aktuellen Rekonstruktionsverfahren großer tumorbefallener Knochendefekte gehören die intraoperative extrakorporale Bestrahlung oder Autoklavierung mit anschließender Reimplantation. Dies führt jedoch zum Teil zu einer erheblichen Veränderung der Transplantatqualität. Mitl der hydrostatischen Hochdrucktechnik (HHD) kommt eine neue Alternative zur experimentellen Anwendung. Maligne Tumorzellen sollen unter Erhalt der biomechanischen Eigenschaften abgetötet werden. Ziel der vorliegenden Arbeit ist es, die biomechanischen und immunhistologischen Auswirkungen einer HHD-Behandlung auf Knorpelgewebe zu untersuchen.

Hydrostatic pressure

10.1515/bmt.2006.003

Cite

Citations (14)

Deep learning-based denoising streamed from mobile phones improves speech-in-noise understanding for hearing aid users

Frontiers in Medical Engineering (2023)

Peter U. Diehl Hannes Zilly Felix Sattler Yosef Singer Kevin Kepp

The hearing loss of almost half a billion people is commonly treated with hearing aids. However, current hearing aids often do not work well in real-world noisy environments. We present a deep learning based denoising system that runs in real time on iPhone 7 and Samsung Galaxy S10 (25 ms algorithmic latency). The denoised audio is streamed to the hearing aid, resulting in a total delay of around 65–75 ms, depending on the phone. In tests with hearing aid users having moderate to severe hearing loss, our denoising system improves audio across three tests: 1) listening for subjective audio ratings, 2) listening for objective speech intelligibility, and 3) live conversations in a noisy environment for subjective ratings. Subjective ratings increase by more than 40%, for both the listening test and the live conversation compared to a fitted hearing aid as a baseline. Speech reception thresholds, measuring speech understanding in noise, improve by 1.6 dB SRT. Ours is the first denoising system that is implemented on a mobile device, streamed directly to users’ hearing aids using only a single channel as audio input while improving user satisfaction on all tested aspects, including speech intelligibility. This includes overall preference of the denoised and streamed signal over the hearing aid, thereby accepting the higher latency for the significant improvement in speech understanding.

Hearing aid

Intelligibility (philosophy)

10.3389/fmede.2023.1281904

Cite

Citations (1)

Growth promoting in vitro effect of synthetic cyclic RGD-peptides on human osteoblast-like cells attached to cancellous bone

Journal of Biomechanics (2006)

Ulla Magdolen Jörg Auernheimer Claudia Dahmen R. Gradinger Harvey P. Kessler

3T3 cells

10.1016/s0021-9290(06)83788-7

Cite

Citations (6)

TrueHappiness: Neuromorphic emotion recognition on TrueNorth

2022 International Joint Conference on Neural Networks (IJCNN) (2016)

Peter U. Diehl Bruno U. Pedroni Andrew Cassidy Paul Merolla Emre Neftci

We present an approach to constructing a neuromorphic device that responds to language input by producing neuron spikes in proportion to the strength of the appropriate positive or negative emotional response. Specifically, we perform a fine-grained sentiment analysis task with implementations on two different systems: one using conventional spiking neural network (SNN) simulators and the other one using IBM's Neurosynaptic System TrueNorth. Input words are projected into a high-dimensional semantic space and processed through a fully-connected neural network (FCNN) containing rectified linear units (ReLU) trained via backpropagation. After training, this FCNN is converted to a SNN by substituting the ReLUs with integrate-and-fire neurons. We show that there is practically no performance loss due to conversion to a spiking network on a sentiment analysis test set, i.e. correlations with human annotations differ by less than 0.02 between the original DNN and its spiking equivalent. Additionally, we show that the SNN generated with this technique can be mapped to existing neuromorphic hardware - in our case, the TrueNorth chip. Mapping to the chip involves 4-bit synaptic weight discretization and adjustment of the neuron thresholds. The resulting end-to-end system can take a user input, i.e. a word in a vocabulary of over 300,000 words, and estimate its sentiment on TrueNorth with a power consumption of approximately 50 μW.

Neuromorphic engineering

Backpropagation

IBM

10.1109/ijcnn.2016.7727758

Cite

Citations (66)

Unsupervised learning of digit recognition using spike-timing-dependent plasticity

Frontiers in Computational Neuroscience (2015)

Peter U. Diehl Matthew Cook

In order to understand how the mammalian neocortex is performing computations, two things are necessary; we need to have a good understanding of the available neuronal processing units and mechanisms, and we need to gain a better understanding of how those mechanisms are combined to build functioning systems. Therefore, in recent years there is an increasing interest in how spiking neural networks (SNN) can be used to perform complex computations or solve pattern recognition tasks. However, it remains a challenging task to design SNNs which use biologically plausible mechanisms (especially for learning new patterns), since most such SNN architectures rely on training in a rate-based network and subsequent conversion to a SNN. We present a SNN for digit recognition which is based on mechanisms with increased biological plausibility, i.e., conductance-based instead of current-based synapses, spike-timing-dependent plasticity with time-dependent weight change, lateral inhibition, and an adaptive spiking threshold. Unlike most other systems, we do not use a teaching signal and do not present any class labels to the network. Using this unsupervised learning scheme, our architecture achieves 95% accuracy on the MNIST benchmark, which is better than previous SNN implementations without supervision. The fact that we used no domain-specific knowledge points toward the general applicability of our network design. Also, the performance of our network scales well with the number of neurons used and shows similar performance for four different learning rules, indicating robustness of the full combination of mechanisms, which suggests applicability in heterogeneous biological neural networks.

Spike-timing-dependent plasticity

MNIST database

Robustness

Benchmark (surveying)

10.3389/fncom.2015.00099

Cite

Citations (1,223)

Clinical outcome and success rates of ACI for cartilage defects of the patella: a subgroup analysis from a controlled randomized clinical phase II trial (CODIS study)

Archives of Orthopaedic and Trauma Surgery (2019)

Philipp Niemeyer Volker Laute Wolfgang Zinser Christoph Becher Peter U. Diehl

Subgroup analysis

10.1007/s00402-019-03264-x

Cite

Citations (13)

Conversion of Artificial Recurrent Neural Networks to Spiking Neural Networks for Low-power Neuromorphic Hardware

arXiv (Cornell University) (2016)

Peter U. Diehl Guido Zarrella Andrew Cassidy Bruno U. Pedroni Emre Neftci

In recent years the field of neuromorphic low-power systems that consume orders of magnitude less power gained significant momentum. However, their wider use is still hindered by the lack of algorithms that can harness the strengths of such architectures. While neuromorphic adaptations of representation learning algorithms are now emerging, efficient processing of temporal sequences or variable length-inputs remain difficult. Recurrent neural networks (RNN) are widely used in machine learning to solve a variety of sequence learning tasks. In this work we present a train-and-constrain methodology that enables the mapping of machine learned (Elman) RNNs on a substrate of spiking neurons, while being compatible with the capabilities of current and near-future neuromorphic systems. This "train-and-constrain" method consists of first training RNNs using backpropagation through time, then discretizing the weights and finally converting them to spiking RNNs by matching the responses of artificial neurons with those of the spiking neurons. We demonstrate our approach by mapping a natural language processing task (question classification), where we demonstrate the entire mapping process of the recurrent layer of the network on IBM's Neurosynaptic System "TrueNorth", a spike-based digital neuromorphic hardware architecture. TrueNorth imposes specific constraints on connectivity, neural and synaptic parameters. To satisfy these constraints, it was necessary to discretize the synaptic weights and neural activities to 16 levels, and to limit fan-in to 64 inputs. We find that short synaptic delays are sufficient to implement the dynamical (temporal) aspect of the RNN in the question classification task. The hardware-constrained model achieved 74% accuracy in question classification while using less than 0.025% of the cores on one TrueNorth chip, resulting in an estimated power consumption of ~17 uW.

Neuromorphic engineering

Reservoir computing

10.48550/arxiv.1601.04187

Cite

Citations (0)