Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Noushin Eftekhari is a Research Associate at the Alan Turing Institute developing machine learning for weather prediction, marine biodiversity monitoring and cancer research.
Posts
Adventures with plankton on the North Sea
Published:
Noushin Eftekhari on life aboard the RV Cefas Endeavour, and how plankton monitoring could revolutionise marine research.
portfolio
FastNet: Machine Learning Weather Prediction
Graph neural network models for global weather forecasting, developed with the Met Office
Real-Time Plankton Monitoring with RAPID
Developing AI-driven tools for real-time plankton analysis at sea
publications
A new dataset of computed-tomography angiography images for computer-aided detection of pulmonary embolism
Published in Scientific Data, 2018
Public dataset of CT angiography images for benchmarking computer-aided detection of pulmonary embolism.
Recommended citation: M. Masoudi, H. Pourreza, M. Saadatmand-Tarzjan, N. Eftekhari, F. S. Zargar, M. P. Rad, "A new dataset of computed-tomography angiography images for computer-aided detection of pulmonary embolism," Scientific Data, vol. 5, 180180, 2018.
Download Paper
Microaneurysm Detection in Fundus Images Using a Two-Step Convolutional Neural Network
Published in Biomedical Engineering Online, 2019
This study presents a novel approach for detecting microaneurysms in fundus images, enhancing early diagnosis in diabetic retinopathy.
Recommended citation: N. Eftekhari, H. Pourreza, M. Masoudi, K. Ghiasi-Shirazi, and E. Saeedi, "Microaneurysm detection in fundus images using a two-step convolutional neural network," Biomedical Engineering Online, 2019.
Download Paper
Computer Aided Detection for Pulmonary Embolism Challenge (CAD-PE)
Published in arXiv preprint arXiv:2003.13440, 2020
Benchmark challenge for automated pulmonary embolism detection in CTA images.
Recommended citation: G. González, et al. (incl. N. Eftekhari), "Computer Aided Detection for Pulmonary Embolism Challenge (CAD-PE)," arXiv:2003.13440, 2020.
Download Paper
Optimizing Plankton Image Classification with Metadata-Enhanced Representation Learning
Published in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024
This research explores the integration of metadata for improving the classification of plankton images, enhancing real-time environmental monitoring.
Recommended citation: M. Masoudi, S. Verma, N. Eftekhari, M. Massot-Campos, J. O. Irisson, B. Thornton, "Optimizing plankton image classification with metadata-enhanced representation learning," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024.
Download Paper
Cross-Attention Enables Deep Learning on Limited Omics-Imaging-Clinical Data of 130 Lung Cancer Patients
Published in Cell Reports Methods, 2024
This study introduces a cross-attention mechanism for multi-modal deep learning on limited data in lung cancer research.
Recommended citation: S. Verma, G. Magazzù, N. Eftekhari, A. Occhipinti, C. Angione, "Cross-attention enables deep learning on limited omics-imaging-clinical data of 130 lung cancer patients," Cell Reports Methods, 2024.
Download Paper
Improving In Situ Real-Time Classification of Long-Tail Marine Plankton Images for Ecosystem Studies
Published in ECCV, CV4E Workshop, 2024
This paper presents a model for enhancing in situ, real-time classification of long-tail marine plankton species, essential for biodiversity monitoring.
Recommended citation: N. Eftekhari, S. Pitois, M. Masoudi, R. E. Blackwell, J. Scott, S. L. C. Giering, "Improving in situ real-time classification of long-tail marine plankton images for ecosystem studies," ECCV, CV4E Workshop, 2024.
Download Paper
RAPID: real-time automated plankton identification dashboard using Edge AI at sea
Published in Frontiers in Marine Science, 2025
An Edge AI system that classifies plankton images in real time aboard a research vessel and streams results to a dashboard on land.
Recommended citation: S. G. Pitois, R. E. Blackwell, H. Close, N. Eftekhari, S. L. C. Giering, M. Masoudi, E. Payne, J. Ribeiro, J. Scott, "RAPID: real-time automated plankton identification dashboard using Edge AI at sea," Frontiers in Marine Science, vol. 11, 2025.
Download Paper
Technical overview and architecture of the FastNet Machine Learning weather prediction model, version 1.0
Published in arXiv preprint arXiv:2509.17658, 2025
Architecture and technical overview of FastNet, a graph neural network model for global weather forecasting developed by The Alan Turing Institute and the Met Office.
Recommended citation: E. G. Daub, T. Dunstan, T. Bennett, ..., N. Eftekhari, et al., "Technical overview and architecture of the FastNet Machine Learning weather prediction model, version 1.0," arXiv:2509.17658, 2025.
Download Paper
FastNet: Improving the physical consistency of machine-learning weather prediction models through loss function design
Published in Artificial Intelligence for the Earth Systems, 2026
Loss function design (spherical harmonic loss, horizontal gradients and wind representation) to make ML weather forecasts more physically realistic.
Recommended citation: T. Dunstan, O. Strickson, T. Bennett, ..., N. Eftekhari, et al., "FastNet: Improving the physical consistency of machine-learning weather prediction models through loss function design," Artificial Intelligence for the Earth Systems, vol. 5, no. 3, e250090, 2026.
Download Paper
Fusing imaging and metabolic modeling via multimodal deep learning in ovarian cancer
Published in Cell Systems, 2026
Patient-specific metabolic models built from transcriptomics are fused with CT imaging through multimodal deep learning to improve survival prediction in ovarian cancer.
Recommended citation: N. Eftekhari, S. Verma, A. Saha, G. Zampieri, S. Sawan, A. Occhipinti, C. Angione, "Fusing imaging and metabolic modeling via multimodal deep learning in ovarian cancer," Cell Systems, vol. 17, 101594, 2026.
Download Paper
talks
Real-time Particle Analysis
Published:
This talk focused on real-time particle analysis using advanced machine learning techniques. I presented methods for processing high-dimensional marine particle data, highlighting AI-driven innovations in ecosystem monitoring.
AI Biodiversity Monitoring: Developing Computer Vision Tools to Assess Ecosystem Health
Published:
This poster presentation highlighted the development of computer vision tools for biodiversity monitoring, showcasing how AI-driven methodologies can be used to assess ecosystem health effectively. It covered novel approaches in visual data analysis for environmental sustainability.
Plankton Classification Using Edge Artificial Intelligence
Published:
At this international symposium, I discussed the use of edge AI for in situ plankton classification, presenting how real-time image analysis tools can support marine ecosystem research. This approach enhances data availability directly in the field, supporting timely ecological assessments.
Improving In Situ Real-Time Classification of Long-Tail Marine Plankton Images for Ecosystem Studies
Published:
At the CV4E workshop during ECCV, I presented a poster on enhancing real-time classification of marine plankton images, particularly focusing on long-tail species. The discussion centered around AI advancements that support more accurate ecosystem studies through improved image classification in real-world conditions.
