[Comp-neuro] postdoc opportunity at NIMH (machine learning approaches in anxiety and depression)

Francisco Pereira francisco.pereira at gmail.com
Sun Nov 17 23:25:25 CET 2019


Dear list,

We have a postdoc opportunity in the Section on Development and
Affective Neuroscience at NIMH, working with Daniel Pine, David
Jangraw, and the NIMH Machine Learning Team. Please see below for full
details.

Francisco Pereira

## Who we are:

Drs. Daniel Pine
(https://www.nimh.nih.gov/research/research-conducted-at-nimh/research-areas/clinics-and-labs/edb/sdan/index.shtml)
and David Jangraw (https://davidjangraw.wordpress.com) are seeking
applicants for a Post-Doctoral Fellowship position that would develop
new machine learning-based approaches for the diagnosis and treatment
of mental illnesses. Our aims are to better understand the causes and
mechanisms of certain psychiatric disorders, improve their definition
and classification, and ensure the best treatment can be offered to
psychiatric patients. We do this by identifying neuroscience-based
phenotypes that map onto clinical phenotypes, and which can be
targeted with novel treatments. The position is co-advised by the NIMH
Machine Learning Team (https://cmn.nimh.nih.gov/mlt), led by Dr.
Francisco Pereira, whose mission is to solve the scientific problems
of NIMH researchers by developing new machine learning methods, or
developing new analysis approaches relying on existing methods.

## What you will do:

One line of research will focus on refining and testing latent
constructs derived from multiple brain imaging and behavioral markers
of inhibitory control. Inhibitory control deficits are common to many
psychiatric disorders, and if the latent construct shows good
reliability, it may serve as a target for novel treatments. Another
line of research will hinge on extracting characteristics of
psychotherapy session data, such as therapist speech and facial
expressions, that can predict patient clinical outcomes. While the
work will concentrate on anxiety and mood disorders in young people,
the approaches will be applicable to any mental disorder.

The fellow will work with multi-modal imaging datasets that include
MRI (functional, structural), EEG, MEG, and associated behavioral and
clinical data, all of which may be used as input to machine learning
methods. Data sets are already available, and the fellow will devote
most of their time to analyzing these data or developing new methods.
The fellow will also have access to excellent computational resources
within NIH (e.g., top-100 supercomputer with hundreds of thousands of
CPUs and hundreds of GPUs).

## Who you are:

Candidates with a strong computational background (e.g., PhD in
Engineering, Physics, Computer Science, Mathematics, Statistics,
Computational Neuroscience, and related areas) who are interested in
brain development and psychopathology are particularly encouraged to
apply. Requirements for this position include:
- Programming experience in data-intensive computation tasks, in
Python (preferably) or in R/Matlab
- Practical machine learning experience (e.g., training of
classification/regression models, statistical testing of results,
interpretation and visualization of models), especially using open
source machine learning platforms such as scikit-learn
- Excellent interpersonal and written (English) communication skills

Experience in psychiatry or knowledge of software for processing
neuroimaging data are not required. However, the candidate will be
expected to learn some of these topics as part of their role in our
research group.

## How to apply:

The successful candidate will work jointly with Drs. Daniel Pine,
David Jangraw, and Francisco Pereira. Please write to them
(pined at mail.nih.gov, david.jangraw at nih.gov, francisco.pereira at nih.gov)
with your CV, using the email as the cover letter. Other enquiries are
also very welcome. The National Institutes of Health is an equal
opportunity employer.


More information about the Comp-neuro mailing list