TSTheodosia Stratoudaki

Theodosia Stratoudaki

*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


Theodosia Stratoudaki has a total of 111 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    On the delay-and-sum processing of shear wave signals from a scanning thermo-elastic laser ultrasound system

    by Matthew W. Riding, Theodosia Stratoudaki

    Scanning laser ultrasound (LU) systems capture sets of signals from different laser positions that may be combined using delay-and-sum (DAS) processing. The most common model used to calculate DAS delays for point-source systems assumes that ultrasound propagates in circular wavefronts. Vertically polarised shear wave (SV) signals from point-like, thermo-elastic LU sources are expected to violate this assumption due to phase rotation effects and the impact of SV head waves. This work explores the consequences of LU SV-SV wavefront non-circularity on DAS processing. Model SV-SV signal sets from point- and plane-reflectors are shown to interfere destructively during DAS when the circular wavefront model is applied, with a loss of 40% of the available signal predicted for an example plane reflector imaging scenario. Signal sets from experiments reproducing the modelled scenarios validate these predictions, with 51% signal loss observed in the corresponding plane reflector scenario. Phase compensation factors derived from the analytical models are applied to the experimental signals but are not found to improve phase coherence, indicating disparities between the models and experiments. An alternative delay model considering a shuttlecock-shaped pre-reflection SV wavefront is shown to yield signals up to 1.8 times those obtained using the circular wavefront model.

  • Open Access

    Adaptive laser-induced phased arrays for imaging of defects on arbitrary non-flat surfaces

    by Rafael Fuentes-Domínguez, Martin Todd, Theodosia Stratoudaki, Richard J. Smith, Matt Clark

  • Open Access

    Neural networks for faster laser ultrasound tomography in tissue phantoms

    by Ahmed Al Fuwaires, Peter Lukacs, Don Pieris, Geo Davis, Helen Mulvana, Katherine Tant, Theodosia Stratoudaki


  • Two-Dimensional Laser-Induced Phased Arrays for Remote Volumetric Ultrasonic Imaging

    by Peter Lukacs, Don Pieris, Geo Davis, Matthew W. Riding, Theodosia Stratoudaki

    Three dimensional (3D) modality is critical for ultrasonic imaging to provide a representative volumetric view within objects, for better evaluation of shape, size and orientation of internal features. Conventional ultrasonic transducers are still used for the majority of phased array ultrasonic measurements. However, transducer arrays have certain drawbacks and limitations: it is a contact technique, requiring couplant. They also have a considerable size/weight/footprint preventing use in places with restricted access and/or extreme environments. Laser generated and detected ultrasound presents a non-contact, remote, ultrasonic imaging method. Furthermore, laser induced phased arrays (LIPAs) have been developed for ultrasonic imaging. Data captured with LIPAs have already been used in conjunction with the delay-and-sum imaging algorithm, the Total Focusing Method (TFM), for high-quality, two-dimensional (2D) ultrasonic imaging. However, there have been very limited instances of 3D laser ultrasonic imaging in the literature and no realisation of TFM yet, even though the TFM is considered the gold standard for ultrasonic imaging. This paper presents a laser ultrasonic system capable of synthesising 2D LIPAs which acquire all-optical data for 3D TFM imaging. The potential and advantages of 2D LIPAs for volumetric imaging are demonstrated first by comparing the use of 1D and 2D arrays to image a crack-like defect. Following this, 3D TFM imaging using 2D LIPAs is evaluated by imaging a sample with 4 bottom-drilled holes. The results are compared to those using three other 3D laser ultrasonic techniques presented in the literature: Monostatic SAFT, Bistatic SAFT and a method called Fixed Detector.


  • Open Access

    Online evolution of a phased array for ultrasonic imaging by a novel adaptive data acquisition method

    by Peter Lukacs, Theodosia Stratoudaki, Geo Davis, Anthony Gachagan

    Ultrasonic imaging, using ultrasonic phased arrays, has an enormous impact in science, medicine and society and is a widely used modality in many application fields. The maximum amount of information which can be captured by an array is provided by the data acquisition method capturing the complete data set of signals from all possible combinations of ultrasonic generation and detection elements of a dense array. However, capturing this complete data set requires long data acquisition time, large number of array elements and transmit channels and produces a large volume of data. All these reasons make such data acquisition unfeasible due to the existing phased array technology or non-applicable to cases requiring fast measurement time. This paper introduces the concept of an adaptive data acquisition process, the Selective Matrix Capture (SMC), which can adapt, dynamically, to specific imaging requirements for efficient ultrasonic imaging. SMC is realised experimentally using Laser Induced Phased Arrays (LIPAs), that use lasers to generate and detect ultrasound. The flexibility and reconfigurability of LIPAs enable the evolution of the array configuration, on-the-fly. The SMC methodology consists of two stages: a stage for detecting and localising regions of interest, by means of iteratively synthesising a sparse array, and a second stage for array optimisation to the region of interest. The delay-and-sum is used as the imaging algorithm and the experimental results are compared to images produced using the complete generation-detection data set. It is shown that SMC, without a priori knowledge of the test sample, is able to achieve comparable results, while preforming $$\sim$$ ∼ 10 times faster data acquisition and achieving $$\sim$$ ∼ 10 times reduction in data size.


  • Open Access

    Online evolution of a phased array for ultrasonic imaging by a novel adaptive data acquisition method

    by Peter Lukacs, Theodosia Stratoudaki, Geo Davis, Anthony Gachagan

    Ultrasonic imaging using ultrasonic phased arrays has an enormous impact in science, medicine and society and is a widely used modality in many application fields. The maximum amount of information which can be captured by an array is provided by the data acquisition method capturing the complete data set of signals from all possible combinations of ultrasonic generation and detection elements of a dense array. However, capturing this complete data set requires long data acquisition time, large number of array elements and transmit channels and produces a large volume of data. All these reasons make such data acquisition unfeasible due to the existing phased array technology or non-applicable to cases requiring fast measurement time. This paper introduces the concept of an adaptive data acquisition process, the Selective Matrix Capture (SMC), which can adapt, dynamically, to specific imaging requirements for efficient ultrasonic imaging. SMC is realised experimentally using Laser Induced Phased Arrays (LIPAs), that use lasers to generate and detect ultrasound. The flexibility and reconfigurability of LIPAs enable the evolution of the array configuration, on-the-fly. The SMC methodology consists of two stages: a stage for detecting and localising regions of interest, by means of iteratively synthesising a sparse array, and a second stage for array optimisation to the region of interest. The delay-and-sum is used as the imaging algorithm and the experimental results are compared to images produced using the complete generation-detection data set. It is shown that SMC, without a priori knowledge of the test sample, is able to achieve comparable results, while preforming ∼10 times faster data acquisition and achieving ∼ 10 times reduction in data size.

  • Automated defect detection in ultrasonic images from a reconfigurable Laser Induced Phased Array

    by University of Strathclyde Glasgow, Don Pieris, Panagiotis Kamintzis, Geo Davis, Theodosia Stratoudaki

    Laser Induced Phased Arrays (LIPAs) are based on the principles of laser ultrasonics where ultrasound is generated and detected by lasers. The laser beams are scanned on the inspected component, collecting data from various array element positions. In this way, an array is synthesized in post processing by applying a variety of ultrasonic imaging algorithms. Unlike transducer based phased arrays, which have fixed geometry of instrumentation, fixed number of elements and pitch as well as fixed operational ultrasonic frequency, LIPAs present a flexibility in array design and introduce the element of reconfigurability, tailored to the requirements of the specific inspection scenario. This paper proposes a method for automated defect detection that is immune to the inherent flexibility of LIPAs and does not depend on fixed array characteristics or large volumes of training data as is the case with methods such as amplitude threshold or machine learning methods. The proposed method is based on the analysis of the noise distribution of the ultrasonic images using TFM images as a case study. This noise distribution is assumed to exhibit Rayleigh distribution for the type of materials presented and the defect detection is based on the Rayleigh Distribution Shift (RDS). A predetermined threshold is set in the noise distribution. TFM images of LIPAs with progressively increasing number of array elements have been experimentally acquired from an aluminum sample in areas with and without defects (1 mm side drilled hole). The shift in the noise distribution peak is compared to the defined threshold. If the peak of the noise distribution crosses the threshold this indicates a detected local Region of Interest (RoI) with the corresponding image pixels signaling its location. Otherwise, the image is considered defect-free. The paper evaluates the performance of this method in defect detection and explores how the threshold of the noise distribution is set, presenting the average precision as a performance metric.

  • Optimising Laser Induced Phased-Arrays for Continuous Inspection of Large Components

    by Sergio Cantero-Chinchilla, Anthony Croxford, Paul Wilcox, University of Bristol, Don Pieris, Peter Lukacs, Theodosia Stratoudaki, University of Strathclyde

    The recently proposed selective matrix capture approach used in inspections with laser induced phased-arrays has made this technology faster and more accurate. The process consists of two stages, one for rapid defect detection and another one for high fidelity defect characterisation. The phased-array configuration is optimally adapted to the current inspection scenario for each stage. This paper proposes a rigorous optimisation process for the defect detection stage for the case of a large component where the array is scanned along the length of the component. This requires an array configuration that achieves the required detection performance with the minimum number of generation and detection points per unit distance. An experiment is set-up to demonstrate the optimality of the resulting array in an aluminium specimen.

  • Grating Lobe Suppression Through Novel, Sparse Laser Induced Phased Array Design

    by Peter Lukacs, Geo Davis, Don Pieris, Anthony Gachagan, Theodosia Stratoudaki

    Laser Induced Phased Array (LIPA) is a non-contact, couplant-free inspection technique providing high quality ultrasonic imaging. LIPAs use lasers for ultrasound generation and detection: ultrasonic beam forming is performed in post processing, using delay-and-sum algorithms, while the data acquisition is done by scanning the two lasers, offering a unique flexibility in array designs that are currently impossible to achieve with conventional transducer based phased arrays. When synthesizing a 2D LIPA, 3D imaging can be achieved. However, due to the mechanical scanning and the large number of elements required for 2D arrays, LIPAs acquisition is time consuming, limiting the potential of the technique. One approach to overcome this limitation is to reduce the number of array elements, however this can lead to the production of grating lobes, resulting in imaging artefacts. Grating lobes can be suppressed through careful LIPA design and this is explored in this study. Rapid array prototyping is demonstrated for synthesizing 3 arrays designs using the same experimental setup.


  • Open Access

    Laser Induced Phased Arrays (LIPA) to detect nested features in additively manufactured components

    by Don Pieris, Theodosia Stratoudaki, Yashar Javadi, Peter Lukacs, Sam Catchpole-Smith, Paul D. Wilcox, Adam Clare, Matt Clark