From dfa0fa0e1313885ed974f0548d2525fe9ce32565 Mon Sep 17 00:00:00 2001 From: arian-farokh <159565340+arian-farokh@users.noreply.github.com> Date: Wed, 7 May 2025 17:26:02 +0100 Subject: [PATCH 1/2] Add files via upload --- Alport.html | 160 ++++++++++++++++++++++++++++++++++++++++++++++++ Home_OCT.html | 164 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 324 insertions(+) create mode 100644 Alport.html create mode 100644 Home_OCT.html diff --git a/Alport.html b/Alport.html new file mode 100644 index 0000000..78df996 --- /dev/null +++ b/Alport.html @@ -0,0 +1,160 @@ + + + + + + + Alport Syndrome + + + + + + + + + + + + + + + +
+
+
+
+

Investigation of Ocular Implications in Alport Syndrome (kidney disorder) through Deep Learning

+

University of West London – School of Computing and Engineering

+
+
+
+
+ +
+
+
+
+
+
+

Supervisory Team

+
Dr Nasim Dadashi
+
+
+ +
+
+
+
+
+
+

Project Description:

+

In this collaborative project focusing on Alport syndrome (AS), we focus on the interplay between this rare genetic disorder, kidney function, and the eyes. AS poses significant risks to vision, necessitating a comprehensive understanding of its impact on the retina [1]. By designing state-of-the-art automatic techniques applied to in vivo 3D retinal imaging within one of the largest AS cohorts to date, our goal is to unveil the complete spectrum of retinal alterations associated with this condition, aided by AI models. This insight holds significant promise for early detection, disease monitoring, and treatment strategies, ultimately enhancing patient care and outcomes.

+ +

As a member of our team, you will play an important role in this clinically significant collaboration, contributing to advancements in medical data analysis, machine learning methodologies, and image processing techniques.

+ +

Retinal optical coherence tomography (OCT) imaging provides a non-invasive and high-resolution method for visualizing the various layers of the retina. This imaging modality enables clinicians and researchers to examine the structural integrity of the retina in detail, facilitating the early detection and monitoring of various ocular diseases.

+ +

The retina consists of several distinct layers, each serving specialized functions in the visual process. Segmentation of retinal layers in OCT images is essential for quantifying structural changes and analysing disease progression accurately. However, despite advancements in automated segmentation algorithms, challenges remain, particularly in cases with pathologies or abnormalities.

+ +

We will utilize previously developed retinal layer segmentation techniques in OCT imaging (e.g. NDD-SEG) to delineate the various retinal layers. However, recognising the potential limitations of existing segmentation models, our initial focus will be on designing an interactive tool to facilitate the correction of any inaccuracies in retinal layer segmentation. This tool will enable clinicians and the researcher to manually adjust segmentations as needed, ensuring the accuracy of subsequent analyses.

+ +

Subsequently, we will design deep learning methodologies to develop models that discover the relationships between retinal structural changes, genetic predispositions, and systemic manifestations of Alport syndrome (AS). By integrating multi-modal data and incorporating insights from both ocular and renal perspectives, we aim to advance our understanding of AS pathogenesis and improve clinical management strategies for affected individuals.

+ +

Ideal candidates should possess the following:

+
    +
  • A strong background in machine learning and computer vision, with experience in frameworks such as PyTorch or TensorFlow.
    • +
  • Familiarity with medical imaging analysis.
    • +
  • Interest in digital health innovation and personalized medicine, with an emphasis on real-world clinical impact.
    • +
+ + +
+
+
+
+
+
+
+ +
+
+
+
+ © Copyright IntSaV. All Rights Reserved +
+
+ +
+
+ +
+ + + + + + + + + + + + + diff --git a/Home_OCT.html b/Home_OCT.html new file mode 100644 index 0000000..329d64f --- /dev/null +++ b/Home_OCT.html @@ -0,0 +1,164 @@ + + + + + + + Home Based OCT + + + + + + + + + + + + + + + +
+
+
+
+

AI-Enhanced home-based Optical Coherence Tomography for Eye Care

+

University of West London – School of Computing and Engineering

+
+
+
+
+ +
+
+
+
+
+
+

Supervisory Team

+
Dr Nasim Dadashi
+
+
+ +
+
+
+
+
+
+

Project Description:

+

Age-related Macular Degeneration (AMD) is a leading cause of vision loss in individuals over 60 and accounts for more than 50% of blindness registrations in the UK. Globally, 196 million people were affected by AMD in 2020, with numbers expected to rise to 288 million by 2040 due to aging populations. Regular monitoring and treatment, such as intraocular injections, are essential to slow disease progression. However, the current model, frequent in-person hospital visits and manual interpretation of Optical Coherence Tomography (OCT) scans, is resource-intensive and places a considerable burden on elderly patients and healthcare systems.

+ +

This project proposes the development of an AI-enhanced home-based OCT solution. This research aims to create a portable, wearable device capable of acquiring retinal images at home. These images will be analysed using advanced AI models to detect disease progression, enabling timely clinical intervention while reducing the need for frequent hospital visits.

+ +

The system will also incorporate explainable AI (XAI) techniques to enhance transparency and clinician trust, ensuring its utility in real-world clinical workflows. This approach aims to increase accessibility to care, personalise treatment schedules, and empower patients to manage their eye health more independently.

+ +

Research Goals:

+
    +
  • Develop robust AI models capable of detecting subtle changes in retinal images captured from home-based OCT devices.
    • +
  • Enhance image quality and diagnostic reliability through data standardisation, noise reduction, and deep learning-based image enhancement.
    • +
  • Improve clinical trust by integrating explainable AI (XAI) methods that provide interpretable insights into AI-driven decisions.
    • +
  • Validate the AI framework using real-world datasets and clinical collaboration to assess accuracy, usability, and regulatory readiness
    • +
+ +

Candidates Profile:

+

Ideal candidates should possess the following:

+
    +
  • Experience with computer vision and deep learning frameworks such as Python, PyTorch, or TensorFlow.
    • +
  • Knowledge of image processing techniques, particularly in medical imaging and low-quality data enhancement.
    • +
  • Familiarity with clinical data, biomedical signal analysis, and interest in ophthalmology or healthcare AI.
    • +
  • An interest in explainable AI and ethical aspects of medical technology, particularly regarding patient safety and clinical adoption.
    • +
+ + +
+
+
+
+
+
+
+ +
+
+
+
+ © Copyright IntSaV. All Rights Reserved +
+
+ +
+
+ +
+ + + + + + + + + + + + + From a7debd5687c383ea32d5eafdad3d6417b7c95d7c Mon Sep 17 00:00:00 2001 From: arian-farokh <159565340+arian-farokh@users.noreply.github.com> Date: Wed, 7 May 2025 17:28:46 +0100 Subject: [PATCH 2/2] Update vacancies.html --- vacancies.html | 2 ++ 1 file changed, 2 insertions(+) diff --git a/vacancies.html b/vacancies.html index e7b6f7e..40d7665 100644 --- a/vacancies.html +++ b/vacancies.html @@ -51,6 +51,8 @@
The specific topics of research change according to our research strategy, b
For each vacancy, please contact the relevant project supervisor(s) for informal discussion and support with your application.
We are currently recruiting for the following specific projects: