PGPatrick Gaydecki

Patrick Gaydecki

*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


Patrick Gaydecki has a total of 121 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Cough sound spectro-temporal analysis and automated detection using Vision Transformers

    by Keming Tan, Jacky Smith, Patrick Gaydecki

    Objective: Cough is a key symptom in respiratory diseases, yet its clinical assessment remains challenging, often relying on subjective questionnaires or inefficient manual cough counting. Existing automated cough detection algorithms have limited generalisability due to a lack of validation on large, diverse datasets. This study aimed to develop and evaluate a fully automated cough detection system using spectro-temporal analysis and a Vision Transformer (ViT) model. Methods: A total of 231 annotated 24-hour cough recordings across 9 diagnostic categories from the RaDAR database were analysed. Recordings were segmented and converted into spectrograms using different short-time Fourier transform settings. Data were split subject-wise into training, validation, and test sets to prevent data leakage. The ViT model was fine-tuned in two stages: a pilot stage using 10% of the data to determine optimal spectrogram parameters, followed by full-scale training on the remaining data. Results: Spectrogram configuration significantly affected performance, with 750 ms segment duration, 128-point frame size, and 32-point hop identified as optimal. With these parameters, the model achieved an F1 score of 85.02%, sensitivity of 83.64%, precision of 86.44%, and specificity of 99.67% on the test set. Diagnostic category-wise analysis showed high F1 scores in interstitial lung disease (90.83%), chronic obstructive pulmonary disease (89.60%), asthma (88.13%) and chronic cough (85.32%). Conclusions: Vision Transformers with optimised spectrogram preprocessing enable accurate, scalable cough detection across diverse populations, performing comparably to popular convolutional neural networks on a larger, more diverse, and more challenging dataset. These findings support the use of ViT-based systems for objective, automated cough monitoring in clinical practice.

  • Open Access

    A wearable gait-analysis device for idiopathic normal-pressure hydrocephalus (INPH) monitoring

    by Erdem Atbas, Patrick Gaydecki, Michael J Callaghan

    Idiopathic Normal Pressure Hydrocephalus (iNPH) is a progressive neurologic disorder (fluid build-up in the brain) that affects 0.2%–5% of the UK population aged over 65. Mobility problems, dementia and urinary incontinence are symptoms of iNPH but often these are not properly evaluated, and patients receive the wrong diagnosis. Here, we describe the development and testing of firmware embedded in a wearable device in conjunction with a user-based software system that records and analyses a patient’s gait. The movement patterns, expressed as quantitative data, allow clinicians to improve the non-invasive assessment of iNPH as well as monitor the management of patients undergoing treatment. The wearable sensor system comprises a miniature electronic unit that attaches to one ankle of the patient via a simple Velcro strap which was designed for this application. The unit monitors acceleration along three axes with a sample rate of 60 Hz and transmits the data via a Bluetooth communication link to a tablet or smart phone running the Android and the iOS operating systems. The software package extracts statistics based on stride length, stride height, distance walked and speed. Analysis confirmed that the system achieved an average accuracy of at least 98% for gait tests conducted over distances 9 m. This device has been developed to assist in the management and treatment of older adults diagnosed with iNPH.


  • An Inhaler Tracking System Based on Acoustic Analysis: Hardware and Software

    by Wenyang Xie, Patrick Gaydecki, Ann-Louise Caress

    In treating asthma and chronic obstructive pulmonary disorder (COPD), acquisition of authentic and effective feedback from patients on regimen adherence is difficult. Face-to-face and oral reporting methods do not satisfy current intelligent medication best practices. This paper presents a system to track and analyze daily inhaler usage. A portable electronic device that attaches to the inhaler uses an accelerometer and capacitive sensors to detect users' motion and an embedded digital microphone to capture sounds while the inhaler is in use. In terms of analysis, sound features are extracted, and breath phases are identified by employing a hidden Markov model with a Gaussian mixture model. A feature template is also constructed and used to search for and identify ``canister pressed'' events. The system provides objective feedback, quantifying asthma, and COPD patients' adherence to medication regimens. Although interest in asthma adherence to medication regimens is growing, there is still a relative paucity of research and, indeed, compliance devices in this area; the tracking system can help doctors better understand the patient's condition and choose an appropriated treatment plan. At the same time, patients can also improve their self-management by system feedback.


  • Automated moth flight analysis in the vicinity of artificial light

    by P. Gaydecki

    Instrumentation and software for the automated analysis of insect flight trajectories is described, intended for quantifying the behavioural dynamics of moths in the vicinity of artificial light. For its time, this moth imaging system was relatively advanced and revealed hitherto undocumented insights into moth flight behaviour. The illumination source comprised a 125 W mercury vapour light, operating in the visible and near ultraviolet wavelengths, mounted on top of a mobile telescopic mast at heights of 5 and 7.1 m, depending upon the experiment. Moths were imaged in early September, at night and in field conditions, using a ground level video camera with associated optics including a heated steering mirror, wide angle lens and an electronic image intensifier. Moth flight coordinates were recorded at a rate of 50 images per second (fields) and transferred to a computer using a light pen (the only non-automated operation in the processing sequence). Software extracted ground speed vectors and, by instantaneous subtraction of wind speed data supplied by fast-response anemometers, the airspeed vectors. Accumulated density profiles of the track data revealed that moths spend most of their time at a radius of between 40 and 50 cm from the source, and rarely fly directly above it, from close range. Furthermore, the proportion of insects caught by the trap as a proportion of the number influenced by the light (and within the field of view of the camera) was very low; of 1600 individual tracks recorded over five nights, a total of only 12 were caught. Although trap efficiency is strongly dependent on trap height, time of night, season, moonlight and weather, the data analysis confirmed that moths do not exhibit straightforward positive phototaxis. In general, trajectory patterns become more complex with reduced distance from the illumination, with higher recorded values of speeds and angular velocities. However, these characteristics are further qualified by the direction of travel of the insect; the highest accelerations tended to occur when the insect was at close range, but moving away from the source. Rather than manifesting a simple positive phototaxis, the trajectories were suggestive of disorientation. Based on the data and the complex behavioural response, mathematical models were developed that described ideal density distribution in calm air and light wind speed conditions. The models did not offer a physiological hypothesis regarding the behavioural changes, but rather were tools for quantification and prediction. Since the time that the system was developed, instrumentation, computers and software have advanced considerably, allowing much more to be achieved at a small fraction of the original cost. Nevertheless, the analytical tools remain useful for automated trajectory analysis of airborne insects.


  • Listener preference towards a real and emulated violin

    by T. Lloyd, P. Gaydecki, H. Johannsson, J. Ginsborg, C. Yates

    An account is provided of a double-blind experiment in which musically trained and untrained participants listened to recordings of a real and emulated (virtual) violin and rated them according to their personal preference. Post-experiment the participants were also asked to identify which one was the real instrument. The emulated violin was developed by convolving the impulse response from an acoustic instrument with the output from an electric violin. The real violin was preferred by both groups with no significant difference between scores for each group. However, more of the trained musicians correctly identified the real violin.

  • Violin mpulse esponse ength and erceptions of cceptability

    by T. Lloyd, P. Gaydecki, J. Ginsborg, C. Yates

    The design, execution and analysis of a double-blind listening study is described, in which participants gave preference ratings for nine versions of the same piece of music, obtained by convolving violin impulse responses, of different length, with the original piece played on an electric instrument and stored in digital form. The original impulse response, with a length of 91.4 ms was measured from a Stradivarius violin and progressively degraded by shortening its length. The participants, who were all trained musicians, were asked to record their preferences based on personal taste, not perceived measures of quality. Analysis of the data revealed a sigmoid relationship between preference and length of impulse response. Emulated music generated using short impulse responses was the least preferred, and this aversion was reasonably constant for responses shorter than 2.7 ms. However, and perhaps surprisingly, impulse responses of only 5.78 ms were deemed acceptable. Extending the length beyond this value had little effect on the attributed preferences. Between these two values, there was a steep increase in the assigned scores, suggesting a sigmoid relationship. The findings are of theoretical interest for psychoacoustics and can be applied to the development of electronic devices that emulate stringed instruments in real time.


  • Real-Time Emulation of the Acoustic Violin using Convolution and Arbitrary Equalization

    by Sheheera Ismail, Patrick Gaydecki, Thomas Lloyd, Hans Johannsson

    This report describes a real-time audio effects processor that modifies the signal generated by an electric violin to produce a sound that closely resembles the tonal qualities of an equivalent acoustic instrument. The processor convolves the incoming signal with an impulse response measured from an acoustic violin in the far field. This approach is justified if the violin body behaves as a linear system, which is typically a good approximation. Because the processor system can store sixteen...


  • Open Access

    Structural Health Monitoring Using Lamb Wave Reflections and Total Focusing Method for Image Reconstruction

    by Aurelia Muller, Bradley Robertson-Welsh, Patrick Gaydecki, Matthieu Gresil, Constantinos Soutis


  • Sensor array system for real time visualization of magnetostatic fields and imaging of ferrous objects

    by Diego S. Benitez, Sung Quek, Patrick Gaydecki

    This paper describes a solid-state based multi-element sensor array system capable of real-time imaging of DC magnetic fields and ferrous objects. The final laboratory system comprises 1155 solid state magneto-inductive sensors arrange in a 2D array matrix of 35 × 33 files an columns, equally spaced in order to cover an approximate area of 320 by 300 mm. The sensor array is located within a large current-carrying coil, which may be exited when required to produce a steady magnetic field. This coil is used when required to imaging metallic objects. Data from the sensors array are sampled sequentially by several DSP controlling units and finally streamed to a host computer using USB 2.0 interface; images are generated and displayed at a rate of about 30 frames per minute. Extensive numerical modeling of field distribution patterns using boundary element method has complemented the development of the system. Both numerical simulations and experimental results are shown and discussed. The system was originally intended for deployment in the non-destructive evaluation (NDE) of reinforced concrete, being capable of imaging embedded rebar of different diameters and configurations up to a depth of 100 mm. Nevertheless, the system is also capable of producing real-time images of the magnetic field distribution emitted by either permanent magnets or geometries carrying current.