SRStefania Russo

Stefania Russo

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Stefania Russo has a total of 27 co-authors. Here are the top co-authors by number of shared publications:

  • Alessandro Tognetti

    Department of Information Engineering, University of Pisa, Pisa, Italy

    5 shared publications
  • Nicola Carbonaro

    Department of Information Engineering, University of Pisa, Pisa, Italy

    3 shared publications
  • Blake Matthews

    Eawag, Swiss Federal Institute of Aquatic Science and Technology, 8600 Dübendorf, Switzerland; Eawag, Department of Fish Ecology & Evolution, Centre for Ecology Evolution and Biogeochemistry, 79 Seestrasse, 6047, Luzern

    2 shared publications
  • Kris Villez

    Eawag, Swiss Federal Institute of Aquatic Science and Technology, 8600 Dübendorf, Switzerland; Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA

    2 shared publications
  • Moritz Lürig

    Eawag, Swiss Federal Institute of Aquatic Science and Technology, 8600 Dübendorf, Switzerland; Eawag, Department of Fish Ecology & Evolution, Centre for Ecology Evolution and Biogeochemistry, 79 Seestrasse, 6047, Luzern; Department of Biology, Lund University, 22362 Lund, Sweden

    2 shared publications
  • Nicola Carbonaro

    Information Engineering Department, Enrico Piaggio Research Center, University of Pisa, Pisa, Italy

    2 shared publications

Latest Publications

  • Open Access

    Prospects and Pitfalls of Machine Learning in Nutritional Epidemiology

    by Stefania Russo, Stefano Bonassi

    Nutritional epidemiology employs observational data to discover associations between diet and disease risk. However, existing analytic methods of dietary data are often sub-optimal, with limited incorporation and analysis of the correlations between the studied variables and nonlinear behaviours in the data. Machine learning (ML) is an area of artificial intelligence that has the potential to improve modelling of nonlinear associations and confounding which are found in nutritional data. These opportunities notwithstanding, the applications of ML in nutritional epidemiology must be approached cautiously to safeguard the scientific quality of the results and provide accurate interpretations. Given the complex scenario around ML, judicious application of such tools is necessary to offer nutritional epidemiology a novel analytical resource for dietary measurement and assessment and a tool to model the complexity of dietary intake and its relation to health. This work describes the applications of ML in nutritional epidemiology and provides guidelines to avoid common pitfalls encountered in applying predictive statistical models to nutritional data. Furthermore, it helps unfamiliar readers better assess the significance of their results and provides new possible future directions in the field of ML in nutritional epidemiology.


  • Open Access

    The value of human data annotation for machine learning based anomaly detection in environmental systems

    by Stefania Russo, Michael D. Besmer, Frank Blumensaat, Damien Bouffard, Andy Disch, Frederik Hammes, Angelika Hess, Moritz Lürig, Blake Matthews, Camille Minaudo, et al.


  • Open Access

    Active learning for anomaly detection in environmental data

    by Stefania Russo, Moritz Lürig, Wenjin Hao, Blake Matthews, Kris Villez


  • Touch Position Detection in Electrical Tomography Tactile Sensors Through Quadratic Classifier

    by Stefania Russo, Roy Assaf, Nicola Carbonaro, Alessandro Tognetti

    Traditional electrical tomography tactile sensors consider the usage of the system's finite element model. This approach brings disadvantages that jeopardize their applicability aspect and wide use. To address this limitation, the main thrust of this paper is to present a method for touch position identification for an electrical tomography flexible tactile sensor. This is done by using a supervised machine learning algorithm for performing classification, namely quadratic discriminant analysis. This approach provides accurate contact location identification, increasing the detection speed and the sensor versatility when compared with traditional electrical tomography approaches. Results obtained show classification accuracy rates of up to 91.6% on unseen test data and an average Euclidean error ranging from 1 to 10 mm depending on the contact location over the sensor. The sensor is then applied in real case scenarios to show its efficiency. These outcomes are encouraging since they promote the future practical usage of flexible and low-cost sensors.


  • Open Access

    A Quantitative Evaluation of Drive Pattern Selection for Optimizing EIT-Based Stretchable Sensors

    by Stefania Russo, Samia Nefti-Meziani, Nicola Carbonaro, Alessandro Tognetti

    Electrical Impedance Tomography (EIT) is a medical imaging technique that has been recently used to realize stretchable pressure sensors. In this method, voltage measurements are taken at electrodes placed at the boundary of the sensor and are used to reconstruct an image of the applied touch pressure points. The drawback with EIT-based sensors, however, is their low spatial resolution due to the ill-posed nature of the EIT reconstruction. In this paper, we show our performance evaluation of different EIT drive patterns, specifically strategies for electrode selection when performing current injection and voltage measurements. We compare voltage data with Signal-to-Noise Ratio (SNR) and Boundary Voltage Changes (BVC), and study image quality with Size Error (SE), Position Error (PE) and Ringing (RNG) parameters, in the case of one-point and two-point simultaneous contact locations. The study shows that, in order to improve the performance of EIT based sensors, the electrode selection strategies should dynamically change correspondingly to the location of the input stimuli. In fact, the selection of one drive pattern over another can improve the target size detection and position accuracy up to 4.7% and 18%, respectively.


  • Open Access

    Development of a High-Speed Current Injection and Voltage Measurement System for Electrical Impedance Tomography-Based Stretchable Sensors

    by Stefania Russo, Samia Nefti-Meziani, Nicola Carbonaro, Alessandro Tognetti

    Electrical impedance tomography (EIT) is an imaging method that can be applied over stretchable conductive-fabric materials to realize soft and wearable pressure sensors through current injections and voltage measurements at electrodes placed at the boundary of a conductive medium. In common EIT systems, the voltage data are serially measured by means of multiplexers, and are hence collected at slightly different times, which affects the real-time performance of the system. They also tend to have complicated hardware, which increases power consumption. In this paper, we present our design of a 16-electrode high-speed EIT system that simultaneously implements constant current injection and differential potential measurements. This leads to a faster, simpler-to-implement and less-noisy technique, when compared with traditional EIT approaches. Our system consists of a Howland current pump with two multiplexers for a constant DC current supply, and a data acquisition card. It guarantees a data collection rate of 78 frames/s. The results from our conductive stretchable fabric sensor show that the system successfully performs voltage data collection with a mean signal-to-noise ratio (SNR) of 55 dB, and a mean absolute deviation (MAD) of 0.5 mV. The power consumption can be brought down to 3 mW; therefore, it is suitable for battery-powered applications. Finally, pressure contacts over the sensor are properly reconstructed, thereby validating the efficiency of our EIT system for soft and stretchable sensor applications.


  • A Quantitative Evaluation of Drive Patterns in Electrical Impedance Tomography

    by Stefania Russo, Nicola Carbonaro, Alessandro Tognetti, Samia Nefti-Meziani


  • Towards the Development of an EIT-based Stretchable Sensor for Multi-Touch Industrial Human-Computer Interaction Systems

    by Stefania Russo, Samia-Nefti Meziani, Tauseef Gulrez, Nicola Carbonaro, Alessandro Tognetti


  • Open Access

    Soft and Stretchable Sensor Using Biocompatible Electrodes and Liquid for Medical Applications

    by Stefania Russo, Tommaso Ranzani, Hongbin Liu, Samia Nefti-Meziani, Kaspar Althoefer, Arianna Menciassi

    This article introduces a soft and stretchable sensor composed of silicone rubber integrating a conductive liquid-filled channel with a biocompatible sodium chloride (NaCl) solution and novel stretchable gold sputtered electrodes to facilitate the biocompatibility of the sensor. By stretching the sensor, the cross section of the channel deforms, thus leading to a change in electrical resistance. The functionalities of the sensor have been validated experimentally: changes in electrical resistance are measured as a function of the applied strain. The experimentally measured values match theoretical predictions, showing relatively low hysteresis. A preliminary assessment on the proposed sensor prototype shows good results with a maximum tested strain of 64%. The design optimization of the saline solution, the electrodes, and the algebraic approximations derived for integrating the sensors in a flexible manipulator for surgery has been discussed. The contribution of this article is the introduction of the biocompatible and stretchable gold sputtered electrodes integrated with the NaCl-filled channel rubber as a fully biocompatible solution for measuring deformations in soft and stretchable medical instruments.