Inès Krissaane

Ines Krissaane

PhD Student in Bayesian Statistics

University of Nottingham

I am interested in machine learning, with a particular focus on computational statistics, deep learning, and uncertainty quantification in the field of biology, genetics and medicine.

My doctoral research aim to understand and develop robust statistical inferential methods for handling model misspecification particularly for cardiac electrophysiology modelling.

Currently, I am pursuing my PhD at the University of Nottingham affiliated with Microsoft Research Cambridge. My PhD supervisors are Richard Wilkinson , Ted Meeds , Jeremy Oakley and Gary Mirams .

Interests

  • Artificial Intelligence
  • Deep Learning
  • Computational Statistics
  • Health Data
  • Genetics

Education

  • Current PhD in Statistics, 2019-2023

    University of Nottingham

  • MSc in Statistics, 2018 (Highest honors)

    Sorbonne Université

  • BSc in Mathematics, 2016

    Pierre and Marie Curie University

Publications

(2020). Scalability and cost-effectiveness analysis of whole genome-wide association studies on Google Cloud Platform and Amazon Web Services.. In Journal of the American Medical Informatics Association.

DOI Interview

(2019). Anomaly Detection Semi-Supervised Framework for Sepsis Treatment.. In IEEE.

DOI

(2019). A generalized statistical framework to assess mixing ability from incomplete mixing designs using binary or higher order variety mixtures and application to wheat.. In Field Crops Research.

PDF DOI

Experience

 
 
 
 
 

Research Associate

Harvard University, Biomedical Informatics Department

Sep 2018 – Feb 2019 Boston, USA
Scalability and cost-effectiveness analysis of whole genome wide association studies on Google Cloud Platform and Amazon Web Services.
 
 
 
 
 

Visiting Research Scholar in Data Science

Harvard University, Biomedical Informatics Department

Apr 2018 – Aug 2018 Boston, USA
Supervisor : Paul Avillach
 
 
 
 
 

Statistics Internship

AgroParisTech - INRA

Apr 2017 – Aug 2017 Paris, France
Supervisors : Stephane Robin and Christophe Ambroise

Projects

*

Data Science Africa

Image Processing for parasites detection in Uganda.

Hodgkin Huxley Model

The Hodgkin–Huxley model, or conductance-based model, is a mathematical model that describes how action potentials in neurons are initiated and propagated. It is a set of nonlinear differential equations that approximates the electrical characteristics of excitable cells such as neurons and cardiac myocytes.

PhysioNet/Computing in Cardiology Challenge 2019

Early detection of sepsis using physiological data

Challenge Data Societé Generale

Predict the Crude Oil production trend - Rank 12th position

Markov chains for detection of genetic motifs

Simulate and analyze genetic sequences using hidden Markov chains, implement the Expectation Maximization algorithm. (Supervisor Catherine Matias).

American Election Analysis (2016)

Analyze the vote in favor of the Republican or Democratic party during the American elections in Python (Supervisor Tabea Rebafka).

Spread of COVID-19

Predict the Global Spread of COVID-19