Skip to content

Commit d84bfaf

Browse files
committed
add biblio to intro
1 parent 513313e commit d84bfaf

File tree

5 files changed

+1124
-19
lines changed

5 files changed

+1124
-19
lines changed
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,7 @@
1+
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
2+
<launchConfiguration type="org.eclipse.ant.AntBuilderLaunchConfigurationType">
3+
<booleanAttribute key="org.eclipse.ui.externaltools.ATTR_BUILDER_ENABLED" value="false"/>
4+
<stringAttribute key="org.eclipse.ui.externaltools.ATTR_DISABLED_BUILDER" value="net.sourceforge.texlipse.builder.TexlipseBuilder"/>
5+
<mapAttribute key="org.eclipse.ui.externaltools.ATTR_TOOL_ARGUMENTS"/>
6+
<booleanAttribute key="org.eclipse.ui.externaltools.ATTR_TRIGGERS_CONFIGURED" value="true"/>
7+
</launchConfiguration>

report/.texlipse

+14
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
#TeXlipse project settings
2+
#Sun Sep 07 15:30:52 BST 2014
3+
markTmpDer=true
4+
builderNum=2
5+
outputDir=
6+
makeIndSty=
7+
bibrefDir=
8+
outputFormat=pdf
9+
tempDir=tmp
10+
mainTexFile=report.tex
11+
outputFile=report.pdf
12+
langSpell=en
13+
markDer=true
14+
srcDir=

report/intro.tex

+37-17
Original file line numberDiff line numberDiff line change
@@ -7,16 +7,19 @@ \section{Introduction} \label{intro}
77
\TODO{wake vs awake}
88
\TODO{conclusion should do more justice}
99

10-
Sleep is considered to be a ubiquitous and necessary behaviour amongst animals.
10+
Sleep is considered to be a ubiquitous and necessary behaviour amongst
11+
animals\cite{siegel_all_2008,cirelli_is_2008}.
1112
However, its real physiological functions remain debated.
1213
In vertebrate, electrophysiological recordings, in particular, \gls{eeg};
1314
the recording of the global electrical activity in the brain,
1415
but also \gls{emg}, which records muscular activity,
15-
have extensively used to study the structure of sleep during the last century.
16+
have extensively used to study the structure of sleep during the last
17+
century\cite{loomis_distribution_1938,aserinsky_regularly_1953}.
1618
They have the advantage of being non-invasive an relatively high throughput.
1719
Today, \gls{eeg} remains one of the main asset in the study sleep physiology.
1820

19-
Rodents, in particular mice and rats, have proved very successful model for understanding of the mechanisms of sleep in mammals.
21+
Rodents, in particular mice and rats, have proved very successful model for
22+
understanding of the mechanisms of sleep in mammals\cite{toth_animal_2013}.
2023
Classically, three main distinct types of sleep related behaviours: wakefulness, \gls{nrem} sleep and \gls{rem}
2124
sleep are referred as \emph{vigilance states}.
2225
Vigilance states are usually defined on the basis of \gls{eeg} and \gls{emg} (fig.~\ref{fig:sleep_description}).
@@ -42,41 +45,58 @@ \section{Introduction} \label{intro}
4245
This severely limits data throughput, and human subjectivity is likely to introduce systematic bias.
4346
Indeed, it is expected that scoring will be performed differently by each expert, making result difficult to reproduce independently.
4447
Often, two experts score the same data, in order to ensure satisfying agreement.
45-
Although, manual scorers are generally reported as being very consensual with each other\citationneeded{},
46-
it can be argued that experts most likely work in the same laboratory and trained one another, or were trained by the same third person.
48+
Although, manual scorers are generally reported as being very consensual with
49+
each other\cite{costa-miserachs_automated_2003,sen_comparative_2014}, it can be
50+
argued that experts most likely work in the same laboratory and trained one another, or were trained by the same third person.
4751
In this context, agreement between experts does not account for the variability between communities of researchers, and cannot be used to assess reproducibility.
4852

49-
In order to overcome both speed and subjectivity limitations, efforts have been directed towards automation of sleep scoring.
53+
In order to overcome both speed and subjectivity limitations, efforts have long
54+
been directed towards automation of sleep
55+
scoring\cite{chouvet_automatic_1980, haustein_automatic_1986}.
5056
However, little adoption has occurred and very few available implementations, in the form of software that biologists could use, have been developed.
5157
Typically, two different approaches to classification have been followed: unsupervised or supervised learning.
5258

53-
Unsupervised learning has the advantage of making no assumption about the nature of the different vigilance states, and how they should be defined.
59+
Unsupervised learning \cite{l&xe4_sleep_2012,sunagawa_faster:_2013} has the
60+
advantage of making no assumption about the nature of the different vigilance states, and how they should be defined.
5461
Therefore, this approach can lead to the discovery of truly new states.
55-
One issue is that the choice of the variables used for clustering is very critical.
56-
Often, variables such as frequency domain variables are in fact chosen in order to generate clusters that will match human defined clusters.
57-
In addition, unsupervised methods may lack robustness in so far as the cannot easily include covariates explaining, for instance, variability between recording equipments.
62+
One issue is that the choice of the variables used for clustering is very
63+
critical.
64+
Often, variables such as frequency domain variables are in fact chosen in order
65+
to generate clusters that will match human defined clusters.
66+
In addition, unsupervised methods may lack robustness\cite{sunagawa_faster:_2013} in so far as the
67+
cannot easily include covariates explaining, for instance, variability between recording equipments.
5868

5969
Another approach is to assume human annotations are, although imperfect, biologically relevant and generally consistent,
60-
and therefore to use supervised learning techniques.
61-
Of course, if human decisions were biased, such a method may suffer from the same bias.
70+
and therefore to use supervised learning
71+
techniques\cite{crisler_sleep-stage_2008,ventouras_performance_2012,doroshenkov_classification_2007,pan_transition-constrained_2012,sen_comparative_2014}.
72+
Of course, if human decisions were biased, such a method may suffer from the same bias.
6273
However, a vast corpus of experimental work has provided hypothesis about function of these states which tends to validate the actual `existence' of these discrete vigilance states.
6374
Building a classifier that would produce a consensual prediction of vigilance states could be seen as an attempt to formalised and rationalise the definition of such states.
6475
This would improve future research without denying decades of sleep neurobiology.
6576

66-
Many supervised learning techniques such as from \glspl{svm}, \glspl{ann}, to \glspl{hmm} have been investigated.
77+
Many supervised learning techniques such as from
78+
\glspl{svm}\cite{crisler_sleep-stage_2008},
79+
\glspl{ann}\cite{ventouras_performance_2012},
80+
to
81+
\glspl{hmm}\cite{doroshenkov_classification_2007,pan_transition-constrained_2012} have been investigated.
6782
In general, the first step is to compute features on consecutive segments of annotated electrophysiological signals know as epochs.
6883
Then, the relation between the response variable(annotation) and the independent variables (features) can be modelled.
6984
Either epochs are considered to be independent from one another or time-dependent structures are explicitly modelled (\eg{} using \glspl{hmm}).
7085
Time aware modelling has the advantage of accounting for the interdependence of consecutive epochs (see fig.~\ref{fig:sleep_description}B).
7186
However, it generally does not model non-linear relationships between large numbers of predictors as well as classical classifiers.
7287

7388
Recently, promising results were obtained for automatic scoring of human sleep stages by performing an exhaustive
74-
feature extraction, including variables resulting from discrete wavelet decomposition.
75-
Then, the authors compared several classifiers and found that random forest were, overall, the most accurate predictors.
89+
feature extraction, including variables resulting from discrete wavelet
90+
decomposition\cite{sen_comparative_2014}.
91+
Then, the authors compared several classifiers and found that random
92+
forest\cite{breiman_random_2001} were, overall, the most accurate predictors.
7693

7794
The study herein bases itself on these promising results by computing an even larger number of features.
78-
An important addition was the computation of time-aware features which significantly improved accuracy.
79-
Furthermore, rigorous stratified cross-validation procedure and comparisons of sleep structure were performed.
95+
An important addition was the computation of time-aware
96+
features\cite{dietterich_machine_2002,deng_time_2013} which significantly improved
97+
accuracy.
98+
Furthermore, rigorous stratified cross-validation\cite{ding_querying_2008} procedure and
99+
comparisons of sleep structure were performed.
80100
These improvement altogether contributed to achieve a very satisfying overall accuracy of 92\%.
81101
In order to pave the way to an implementation of an ubiquitous sleep scoring software.
82102
\pr, a new \py{} package was also build to facilitate efficient feature extraction.

0 commit comments

Comments
 (0)