RDRobin Dale

Robin Dale

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Robin Dale has a total of 14 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Towards real-time diffuse optical tomography with a handheld scanning probe

    by Robin Dale, Nicholas Ross, Scott Howard, Thomas D. O’sUllivan, Hamid Dehghani

    Diffuse optical tomography (DOT) performed using deep-learning allows high-speed reconstruction of tissue optical properties and could thereby enable image-guided scanning, e.g., to enhance clinical breast imaging. Previously published models are geometry-specific and, therefore, require extensive data generation and training for each use case, restricting the scanning protocol at the point of use. A transformer-based architecture is proposed to overcome these obstacles that encode spatially unstructured DOT measurements, enabling a single trained model to handle arbitrary scanning pathways and measurement density. The model is demonstrated with breast tissue-emulating simulated and phantom data, yielding - for 24 mm-deep absorptions (μ a ) and reduced scattering (μ s ′) images, respectively - average RMSEs of 0.0095±0.0023 cm−1 and 1.95±0.78 cm−1, Sørensen-Dice coefficients of 0.55±0.12 and 0.67±0.1, and anomaly contrast of 79±10% and 93.3±4.6% of the ground-truth contrast, with an effective imaging speed of 14 Hz. The average absolute μ a and μ s ′ values of homogeneous simulated examples were within 10% of the true values.


  • Open Access

    Towards real-time diffuse optical tomography with a handheld scanning probe

    by Robin Dale, Nick Ross, Scott Howard, Thomas O'Sullivan, Hamid Dehghani

    Diffuse optical tomography (DOT) performed using deep-learning allows high-speed reconstruction of tissue optical properties, and could thereby enable image-guided scanning, e.g. to enhance clinical breast imaging. Previously published models are geometry specific, and therefore require extensive data generation and training for each use case, and restrict the scanning protocol at the point of use. To overcome these obstacles, a novel transformer-based architecture is proposed to encode spatially unstructured DOT measurements, enabling a single trained model to handle arbitrary scanning pathways and measurement density. The model is demonstrated with breast tissue-emulating simulated and phantom data, yielding - for 24 mm-deep absorption (μa) and reduced scattering (μs') images respectively - average RMSEs of 0.0095±0.0023 cm¯¹ and 1.95±0.78 cm¯¹, Sørensen-Dice coefficients of 0.55±0.12 and 0.67±0.1, and anomaly contrast of 79±10\% and 93.3±4.6% of the ground-truth contrast, with an effective imaging speed of 14 Hz. The average absolute μa and μs' values of homogeneous simulated examples were within 10% of the true values.

  • Open Access

    Towards real-time diffuse optical tomography with a handheld scanning probe

    by Robin Dale, Nick Ross, Scott Howard, Thomas O'Sullivan, Hamid Dehghani

    Diffuse optical tomography (DOT) performed using deep-learning allows high-speed reconstruction of tissue optical properties, and could thereby enable image-guided scanning, e.g. to enhance clinical breast imaging. Previously published models are geometry specific, and therefore require extensive data generation and training for each use case, and restrict the scanning protocol at the point of use. To overcome these obstacles, a novel transformer-based architecture is proposed to encode spatially unstructured DOT measurements, enabling a single trained model to handle arbitrary scanning pathways and measurement density. The model is demonstrated with breast tissue-emulating simulated and phantom data, yielding - for 24 mm-deep absorption (μa) and reduced scattering (μs') images respectively - average RMSEs of 0.0095±0.0023 cm¯¹ and 1.95±0.78 cm¯¹, Sørensen-Dice coefficients of 0.55±0.12 and 0.67±0.1, and anomaly contrast of 79±10\% and 93.3±4.6% of the ground-truth contrast, with an effective imaging speed of 14 Hz. The average absolute μa and μs' values of homogeneous simulated examples were within 10% of the true values.

  • Transformer-Encoder for Real-Time DOT Scanning

    by Robin Dale, Nicholas Ross, Scott Howard, Thomas D. O’sUllivan, Hamid Dehghani

    Deep-Learning based Diffuse Optical Tomography enables real-time clinical breast imaging, but current models are scanning-pathway-specific. Here, a transformer architecture is proposed to encode arbitrary scanning pathways, enabling flexible real-time imaging with a single trained model.

  • Open Access

    Deep learning-enabled high-speed, multi-parameter diffuse optical tomography

    by Robin Dale, Biao Zheng, Felipe Orihuela-Espina, Nicholas Ross, Thomas D. O’sUllivan, Scott Howard, Hamid Dehghani

    Diffuse optical tomography (DOT) is a safe and non-invasive medical imaging technique in which near-infrared (NIR) light (600 to 1000 nm) is transmitted between an array of sources and detectors positioned on the surface of the body and the properties of the measured light are used to reconstruct images of the optical properties of the intervening tissue. DOT has been successfully applied in numerous research areas requiring characterization of tissue composition and metabolism, including neuroimaging1,2 and musculoskeletal3 imaging, and has been proposed as an alternative or complementary technique to X-ray mammography for clinical breast imaging.4 Specifically, as the spectral attenuation of NIR light is strongly dependent on tissue physiology, DOT-measured biomarkers can be used to detect tumors,5 distinguish malignant tumors from benign tumors,6,7 and predict individual responses to neoadjuvant chemotherapy,8 without exposing the patient to ionizing radiation.


  • Experimental validation of multi-parameter diffuse optical tomography using deep-learning

    by Robin Dale, Biao Zheng, Thomas O’sUllivan, Hamid Dehghani

    This study presents experimental validation of a deep-learning (DL) model for simultaneous 3D reconstruction of absolute absorption and reduced scattering. The model improved structural similarity and contrast, but increased cross-talk compared to an analytical algorithm.

  • FREEnet: a dynamic deep-learning model for freehand diffuse optical tomography

    by Robin Dale, Thomas O’sUllivan, Hamid Dehghani

    A deep-learning (DL) model for handheld diffuse optical tomography is presented. The fully convolutional network can reconstruct 3D absorption and scattering from arbitrarily undersampled scan data at a rate of 18.5Hz, enabling real-time imaging.


  • Open Access

    System Derived Spatial-Temporal CNN for High-Density fNIRS BCI

    by Robin Dale, Thomas D. O'SUllivan, Scott Howard, Felipe Orihuela-Espina, Hamid Dehghani

    An intuitive and generalisable approach to spatial-temporal feature extraction for high-density (HD) functional Near-Infrared Spectroscopy (fNIRS) brain-computer interface (BCI) is proposed, demonstrated here using Frequency-Domain (FD) fNIRS for motor-task classification. Enabled by the HD probe design, layered topographical maps of Oxy/deOxy Haemoglobin changes are used to train a 3D convolutional neural network (CNN), enabling simultaneous extraction of spatial and temporal features. The proposed spatial-temporal CNN is shown to effectively exploit the spatial relationships in HD fNIRS measurements to improve the classification of the functional haemodynamic response, achieving an average F1 score of 0.69 across seven subjects in a mixed subjects training scheme, and improving subject-independent classification as compared to a standard temporal CNN.