Research

To research students, please download this researchPack. File names are ordered in the order you should read them in. At least read “0_researchAdvices” documents. It will guide you through other readings when needed, and if needed.

My current research areas:

  • Parallel & Distributed Computing
  • Big Data & high dimensional (Tensor) Computing
  • Optimisation Techniques & Combinatorial Algorithms
  • Legal Ontologies
  • Bioinformatics
  • Artificial Intelligence
  • Graphics, Audio, Image & Video or generally signal processing

Brunel Research Fellow Position From December 2016 to Present:
Technologies used: MS Visual Studio C#, SolidWorks APIs, Neo4j, Alchemy.js, D3.js, OpenGL, probably will need OpenCL soon.

  1. Design Image Semantic Descriptors using ontology represented in Graph Databases for Engineering Drawings
  2. 3D graphics to summarise patents database search results scored on similarity with current Design Document in SolidWorks.

UCL Research Associate Position from June 2016 to December 2016:
Technologies used: Database (MySQL stored procedures), Python Combinatorics code, Stata code, matlab code
Administrative Data Analytics such as:

  1. Extracting cohorts from HES datasets based on various diagnosis (ICD-10) and procedures codes (OPCS),
  2. analysing co-morbidity and mortality by age groups, location, and any required variable (hospital codes, praticitioner, … etc)
  3. Analyse Data Missing Values patterns and Hence selection of identifiers
  4. Finished this work on Down Syndrome cohort in HES, and is parametrizable and can be generalised to any cohort
  5. Attended training on various types of regression analysis, linkage methods from statistical point of view (very relevant to record linkage with fuzy rules in database management), Hospital Records as Hospital Episode Statistics (HES) – the English standards.

My MSc and PhD research work developed a library for high dimensional datasets (hypercube or tensors) data structure representation and indexing, using the conformal computing – Mathematics of Arrays (MoA) constructs. I developed a hyper-cube partitioning scheme for parallel processing, and applied it to the Multiple Sequence Alignment problem, using a reduced search space method.

My Distributed MSA: https://github.com/mhelal/mmDST

Research Resources:
Searching Research Journals
Funding:
1. Egyptian National S&T Information Network

  • GERSS-German-Egyptian Research Short term Scholarships
  • EGYPT-SOUTH AFRICA JOINT SCIENCE & TECHNOLOGY RESEARCH PROGRAMME
  • EU-Egypt Science and Innovation
  • Science and Technology Development Fund (STDF)

2. World-Wide Universities

  • Funded Research Projects
  • Fulbright Egypt
  • EURAXESS
  • FP7
  • Marie Curie

3. Industrial Research & Development

  • Companies such as Google, Microsoft, Oracle, and other specialised companies announce research funding and sponsorships

Leave a Reply

Your email address will not be published. Required fields are marked *