bp/QMUL_Final_Project
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Begun summary table of segmentation algorithms

Browse Source
This commit is contained in:
Sam Perry
2017-08-02 15:34:10 +01:00
parent 3ed50ab0d2
commit 3b05e3503f
1 changed files with 127 additions and 42 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Project_Writeup.tex
+127 -42
View File
@@ -4,8 +4,8 @@
\DeclareLanguageMapping{british}{british-apa}
\usepackage{url}
\usepackage{float}
\usepackage[labelformat=empty]{caption}
\restylefloat{table}
\usepackage{caption}
%\restylefloat{table}
\usepackage{perpage}
\MakePerPage{footnote}
\usepackage{abstract}
@@ -14,8 +14,11 @@
% Create hyperlinks in bibliography
\usepackage{hyperref}
\usepackage{amsmath}
\usepackage{booktabs}
\usepackage{tabulary}
\usepackage[pass]{geometry}
\usepackage{pdflscape}
\usepackage{graphicx}
\usepackage[T1]{fontenc}
@@ -41,6 +44,13 @@
% {}{}
%\makeatother
\newcommand{\dtoprule}{\specialrule{1pt}{0pt}{1.4pt}%
\specialrule{1pt}{0pt}{\belowrulesep}%
}
\newcommand{\dbottomrule}{\specialrule{1pt}{0pt}{1.4pt}%
\specialrule{1pt}{0pt}{\belowrulesep}%
}
\DeclareCiteCommand{\citeyearpar}
{}
{\mkbibparens{\bibhyperref{\printdate}}}
@@ -69,7 +79,7 @@
tabsize=4,
showspaces=false,
showstringspaces=false}
\usepackage[shortcuts]{extdash}
\begin{document}
\newgeometry{lmargin=1.5cm}
@@ -78,24 +88,24 @@
\begingroup
\setlength{\tabcolsep}{1.5cm}
\begin{tabular}[c]{p{0.30\textwidth} | p{0.4\textwidth}}
{\vspace{1.2cm} \Large School of Electronic Engineering and Computer Science \par}
&
{\vspace{1.2cm} \Large School of Electronic Engineering and Computer Science \par}
&
{\vspace{1.2cm} \large Sound and Music Computing \newline Project Report \the\year \par}\\
& {\vspace{0.5cm} \Large \textbf{Extraction of Statistical Features from PCG Signals for the
Classification of Heart Abnormalities} \par}\\
\vspace{0.4\textheight}
\includegraphics[width=5cm]{qmul_logo}
&
{\vspace{1cm} \large \textbf{Samuel Perry}}\\
&
\multicolumn{1}{|r}{August \the\year}
\end{tabular}
\endgroup
@@ -120,36 +130,90 @@ I'd like to thanks anyone and everyone...
There are currently a wide variety of methods employed for the analysis and
classification of PCG signals. Current research can be divided into 3 areas,
each of which are combined to create full classification system. These areas
are: signal preprocessing and segmentation, feature extraction methods and
classification methods.
are: signal preprocessing, signal segmentation and feature extraction methods,
and classification methods.
The performance and evaluation of complete systems are also discussed in
section~\ref{performance}
\subsection{Signal Preprocessing and Segmentation}
Due to factors such as recording conditions and
Algorithms for the pre-processing and segmentation of PCG data
aim to extract the structure of the signal over time. This is a key
stage in the analysis of PCG signals as the structure and relationships between the
fundamental heart sounds (FHSs) form the basis for much of the further
analysis performed on PCG data. A number of methods exist for the
extraction of FHSs. Some rely on direct extraction of peaks in the time
domain to determine the structure of a signal. These methods perform
various transformation in order to accentuate the transient events with
the intention of isolating them~\parencite{Groch1992, Liang1997}.
However, these methods tend to suffer significantly from background
noise and so perform poorly in sub-optimal conditions.\\
Other methods rely on spectral representations to assist in the
splitting of the FHSs, in particular using wavelet
decomposition~\parencite{LiangHuiying1997, Vepa2008}. This allows for
the separation of components based on their frequency content in place
of, or in addition to their temporal characteristics.\\
In addition, Machine learning algorithms have been employed, such as
$k$-Nearest Neighbour~\parencite{Gupta2007} and Neural
Networks~\parencite{Oskiper2002} to improve segment classification.
More recently, particular success has been observed in Springer's use
of logistic regression and Hidden semi-Markov
models~\citeyearpar{Springer2016}.
\subsection{Signal Preprocessing}
There are a large number of factors that lead to variation in quality of PCG
recordings: stethescope type, make and model, it's microphone/sensors used for
recording of the data, the position used to record (i.e. lower left sternal
border, apex, pulmonic area, aortic area), built in filters/signal processing
used by the stethescope (i.e. noise filters, anti-tremor filters), medication that
a pacient may be taking, as well as many other factors that may influence the
recorded signal~\parencite[p.4]{Pavlopoulos2004}. This presents a significant
issue when attempting to analyse and compare a dataset of signals, as
variations in recordings and artefacts caused by factors other than heart
sounds will most likely interfere with analysis and comparison methods. To
account for this, pre-processing methods are widely used aiming to standardize
a dataset. This is also used as a way to accentuate features of the data that
are expected to be relevant during classification.\\
\subsection{Statistical Feature Extraction}
A common method employed is the use of decimation and a static filter to remove
unwanted spectral content that is most likely noise~\parencite{Liang1997a,
Homsi2016, Springer2016, Gupta2007}. This helps reduce higher frequency noise
such as speech, microphone movement and other interference caused externally.
Decimation tends to downsample to around 1--4KHz, with anti-aliasing filter
specifications varying across the literature. Generally, highpass chebychev or
butterworth filters are favoured with cutoff frequencies ranging from
400--750Hz.\\
In addition, many methods decompose the filtered signal using wavelet based
methods such as the discrete wavelet transform
(DWT)~\parencite{Liang1997a, Pavlopoulos2004}, continuous
wavelet transform (CWT)~\parencite{Langley2016} or wavelet
package decomposition (WPD)~\parencite{Liang1998}.
Wavelet transforms are popular as unlike Fourier transforms, they are well
localized in both the time and frequency domain. This allows for the analysis
of PCG signals across multiple frequency bands whilst maintaining transient
temporal events in the resulting decomposition~\parencite[p.93]{Ari2008}.
This may be used for analysis of transient events such as murmurs, that may
consist of higher frequency components than normal heart sounds.
\subsection{Signal Segmentation}
Algorithms for the segmentation of PCG data aim to extract the structure of
the signal over time. This is a key stage in the analysis of PCG signals as the
structure and relationships between the fundamental heart sounds (FHSs) form
the basis for much of the further analysis performed on PCG data. A number of
methods exist for the extraction of FHSs. Tradiational methods rely on direct
extraction of peaks in the time domain to determine the structure of a signal.
These methods perform various transformation in order to accentuate the
transient events with the intention of isolating them~\parencite{Liang1997b}.
However, these methods tend to suffer significantly from background noise and
so perform poorly in sub-optimal conditions.\\ More recent methods use spectral
representations to assist in the splitting of the FHSs, in particular using
wavelet decomposition~\parencite{Liang1997a, Vepa2008}. These methods tend to
perform more robustly on signals of varying conditions\\ In addition, Machine
learning algorithms have been employed, such as $k$-Nearest
Neighbour~\parencite{Gupta2007} and Neural Networks~\parencite{Oskiper2002} to
improve segment classification. Particular success has been observed in
Springer's use of logistic regression and Hidden semi-Markov models
(HSMM)~\citeyearpar{Springer2016}.
% TODO: Insert table of segmentation methods and results
\newgeometry{margin=1cm} % modify this if you need even more space
\begin{landscape}
\begin{table}[htbp]
\captionof{table}{Summary of Segmentation Algorithms} \label{SegmentationTable}
\small
%\centering
\begin{tabulary}{\linewidth}{LLLLL}
\dtoprule
Author & Method & Datasets & Reported Metrics and Results & Notes \\ \midrule
Springer, D. B., Tarassenko, L., \& Clifford, G. D. (2016) & HSMM/Logistic regression & 10,172s of recordings from 112 patients. 12 181 first and 11 627 second heart sounds. & F1 score of 95.630.85\% & Supervised algorithm. \\
Huiying, Sakari, \& Iiro, (1997b) & Normalised Average Shannon Energy Envelope/Peak Picking & 37 recordings, 14 pathological murmurs and 23 physiological murmurs. 515 cycles & 91.03\% correct, 5.83\% missing, 1.17\% incorrect & Unsupervised Algorithm. Dataset consists entirely of child recording. Optimized on entire dataset \\
Gupta, C. N., Palaniappan, R., Swaminathan, S., \& Krishnan, S. M. (2007) &
Homomorphic Filtering\slash K\=/means clustering & 41 recordings (340 cycles). Mix of normal (32\%), systolic (36\%) and diastolic murmurs (32\%) & 90.29\% Ac. & Unsupervised Algorithm. \\
\dbottomrule \\
\end{tabulary}
\end{table}
\end{landscape}
\restoregeometry
\subsection{Feature Extraction}
A wide variety of methods exist for the extraction of statistical
features from PCG data. These features are used for the creation of
robust, meaningful representations of the data.\\
@@ -192,7 +256,10 @@ detectable through time domain analysis~\citeyearpar{Yaghouby2009}.\\
Further in-depth analysis of statistical features for HRV can be found
in~\parencite{Electrophysiology1996}
\subsection{Signal Classification}
\subsection{Classification Models}
% TODO: Revise to include physionet entries
% TODO: Add section for parameter optimization/feature selection methods
Classification of signals for diagnostic purposes. The aim being to
distinguish healthy signals from those with certain heart
conditions/abnormality. This is most commonly achieved by extracting
@@ -240,22 +307,40 @@ may add considerable complexity to computations, and so care should be
taken, particularly when considering systems in a real-time
context~\citeyearpar{Orhan2013}.
\subsection{System Performance}\label{performance}
\subsubsection{Work prior to the Physionet Challenge}
\subsubsection{Physionet Challenge 2016 Entries}
% TODO: Insert table of previous research methods, datasets and results
\section{Dataset}
\section{Design}
The system aims to provide robust heart abnormality detection for PCG signals,
such that use of the system could reliably recommend further medical attention
when neccesary.
when neccesary.
\subsection{Signal Segmentation}
\subsection{Choice of features}
Augmentation of features using 2nd order polynomial features
- Dangers of overfitting with higher order features
\subsubsection{Wavelet Decomposition}
% TODO: Insert wavelet diagram here
\subsection{Feature selection method}
dimensionality reduction
\subsection{Classification Algorithm}
PCA/KPCA
Sequential forward feature selection
\subsection{Classification Model Selection/Optimization}
Particle Swarm Optimization
Individual model structures used in optimization
\section{Implementation}
\section{Evaluation}
Group cross-validation
Weighted specificity and weighted Accuracy measures
Computational cost was not considered, unlike other entries to the physionet
challenge
Comparison with T-Pot
\section{Conclusion}