dataminingincancerresearch ieee
DESCRIPTION
mbTRANSCRIPT
![Page 1: DataMiningInCancerResearch IEEE](https://reader036.vdocuments.mx/reader036/viewer/2022081908/55cf9bf9550346d033a81340/html5/thumbnails/1.jpg)
14 IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE | FEBRUARY 2010 1556-603X/10/$26.00©2010IEEE
Paulo J.G. Lisboa, Liverpool John Moores University, UK Alfredo Vellido, Technical University of Catalonia, SPAIN Roberto Tagliaferri and Francesco Napolitano, University of Salerno, ITALY Michele Ceccarelli, University of Sannio, ITALY José D. Martín-Guerrero, University of Valencia, SPAINElia Biganzoli, University of Milano, ITALY
Application Notes
I. Introductiondvances in cancer medicine have
traditionally come from detailed
understanding of biological pro-
cesses, later translated into therapeutic
interventions, whose effectiveness is
established by rigorous analysis of
clinical trials. Over the last
two decades the increasing
throughput of data from
microarray screening, spec-
tral imaging and longitudinal
studies are turning the under-
standing of cancer pathology
into as much a data-based as
a biologically and clinically
driven science, with potential to impact
more strongly on evidence-based deci-
sion support moving towards personal-
ized medicine [1].
This article is not intended as a
comprehensive survey of data mining
applications in cancer. Rather, it provides
starting points for further, more targeted,
literature searches, by embarking on a
guided tour of computational intelli-
gence applications in cancer medicine,
structured in increasing order of the
physical scales of biological processes, as
outlined in Table 1.
II. Robust Clustering and Data VisualizationOver the last decade, cancer has become
a data-intensive area of research, with
accelerating developments in data-
acquisition technologies and specific
methodologies, for example, next-generation
sequencing for monitoring genetic
changes in tumor cells as they progress
from normal to invasive [2].
The first step in unrav-
elling these processes is
the study of co-occur-
rences, leading to the
identification of disease
sub-types. This already has
profound implications for the
clinical management of patients,
impacting on one of the most
funda mental precepts of cancer med-
icine – the histological characterisation
of malignancies. Currently this is over-
whelmingly done by measuring the dif-
ferentiation between cancerous cells and
the original normal cells from which
they have evolved, marking them as well
to poorly differentiated, meaning that the
changes are linked with good to poor
prognosis. Yet it is becoming apparent that
the metabolism of cells is a key factor in
uncontrolled cell proliferation, with a
potentially greater influence on disease
progression than cell differentiation alone.
Clustering is the first line analysis
methodology to mine data bases for use-
ful research hypotheses to take onto fur-
ther, more directed, study [3], [4]. From
a mathematical perspective the objects of
study, be they genes, patients, mode of
action drugs or disease sub-types, are
points in a multi-dimensional space
where similarity is defined, typically
trough a measure of the distance between
object pairs. The main classes of cluster-
ing approaches used in data mining for
cancer are partitional, creating a simple
partition of the data, or hierarchical, which
create a tree-structured hierarchy of the
data. Partitional approaches form a very
Digital Object Identifier 10.1109/MCI.2009.935311
Data Mining in Cancer Research
TABLE 1 Overview of healthcare interventions across physiological scales.
BIOLOGICAL PROCESSES
PHYSIOLOGICAL SCALE
INDUSTRIAL SECTOR
APPLICATION DOMAIN
MOLECULAR BIOLOGYGENOME AND PHYSIOME
BIOINFORMATICSPHARMACEUTICS
SUSCEPTIBILITY TO DISEASE
METABOLIC PATHWAYSTARGETS FOR INTERVENTION
CYTOLOGY AND HISTOLOGY
CELL AND TISSUE LEVELS
SYSTEMS BIOLOGYLABORATORY TESTS
DISEASE DIAGNOSIS
MEDICAL IMAGING
ORGAN LEVEL
NEUROINFORMATICS
MEDICAL INFORMATICS
MEDICAL EQUIPMENT & INSTRUMENTATION
PHARMACEUTICS
ANATOMICAL AND PHYSIOLOGICAL MONITORING
ELECTROPHYSIOLOGICAL MEASUREMENT
CLINICAL SIGNSSYSTEM LEVEL DIAGNOSIS, PROGNOSIS
AND SCREENING
OUT-OF-HOSPITAL CARE
LEVEL OF THE INDIVIDUAL
AMBULATORY MONITORING FOR EARLY DIAGNOSIS
LIFESTYLE FACTORSPOPULATION LEVEL PUBLIC HEALTH
INFORMATICSDISEASE PROTECTION AND PREVENTION
© EYEWIRE
A
Authorized licensed use limited to: LIVERPOOL JOHN MOORES UNIVERSITY. Downloaded on February 12, 2010 at 05:38 from IEEE Xplore. Restrictions apply.
![Page 2: DataMiningInCancerResearch IEEE](https://reader036.vdocuments.mx/reader036/viewer/2022081908/55cf9bf9550346d033a81340/html5/thumbnails/2.jpg)
FEBRUARY 2010 | IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE 15
heterogeneous class, including techniques
based on Global Optimization, Vector
Quantization, Kernel functions, Fuzzy
sets, and Graph theory [5]. Hierarchical
approaches can be agglomerative or divisive
and are often used as for first-line clus-
tering of high-throughput data. How-
ever, high- dimensional data spaces are
by nature ultra-metric, each data point
appearing to be at the centre of a hollow
sphere, in contrast with the usual intui-
tive picture in low dimensions [6]. This
effect needs to be taken into account
especially with agglomerative methods
to avoid early stage errors that can arise,
resulting in sub-optimal solutions com-
pared with partition algorithms. A third
main class of algorithms is biclustering
that is to say clustering of both objects
and features within a single model [7].
The range of clustering techniques illus-
trates the complexity of the problem.
The fundamental concept underlying
the definition of clusters, “similarity”, is
complex in itself because of its ambigui-
ty. This difficulty is compounded when
the most relevant features are not known
a priori, yet the choice of different fea-
tures usually produces different solutions.
An open research question is feature selec-
tion operating with unsupervised train-
ing. This will normally involve
parametric modeling and the application
of Bayesian methods to derive the dis-
criminative features which best separate
between the clusters [8].
In addition to such intrinsic complex-
ity, purely algorithmic issues may give
rise to different solutions for the same
data set, even for the same algorithm.
This raises the possibility that cluster sta-
bilization methods including global stabil-
ity assessments [10], [11] and consensus
clustering [12], [13] methods may miss the
real complexity of the problem.
An alternative approach to clustering
assessment shifts the focus from the
search of the best solution to the search
of a set of equally plausible, though
different, solutions [14]. Recently, meta-
clustering [15] (see fig 1) has been pro-
posed as a clustering assessment tool. It is
the process of clustering the solutions i.e.,
the cluster partitions themselves, general-
izing the concept of global stability to
local stability (see Fig. 1), that is, the sta-
bility of a solution with respect to a
homogenous subset of the candidates.
The cluster composition then needs
to be explored in detail, for which den-
drograms provide a useful representation,
usually through sequential univariate
splitting of the data space. However, the
interpretation of geometrical relationships
in the data is often best derived from
direct spatial representations. The most
commonly used methods for data visual-
ization are Multi Dimensional Scaling,
Principal Component Analysis (PCA)
and Self-Organizing Maps [16], [17].
Linear methods are particularly inter-
esting since they enable axes projections
to be overlaid on the data. An efficient
methodology for cluster-based linear
visualization is to use a decomposition of
the covariance matrix which explicitly
takes into account cluster membership.
This is the natural tailoring of PCA to the
case where the data are partitioned into
labelled cohorts – an example from breast
cancer proteomics is shown in Fig. 2 [18].
III. Pathway ModelingThe study of the cell functions is mod-
elled from experimental data sets which
are both complex and dynamical [19],
[20]. This has spawned the field of Sys-
tems Biology to understand and describe
how molecular components interact to
manifest the phenotypic behaviour
which is the target of disease interven-
tion [21], illustrated in Fig. 3. In the sys-
tems biology literature there are two
main classes of approaches:
Inverse problems, which involves ❏
mining data from many perturbation
experiments to identify a compatible
structure of the signaling pathway.
Direct problems, to forecast and simu- ❏
late the forward evolution of dynami-
cal systems, using differential equations.
Several methods make the assump-
tion that the expression level of a given
gene can be considered as a random
variable and the mutual relationships
between them can be derived by statisti-
cal dependencies. The resulting system is
embedded in a probabilistic Graphical
Model [22] described for example by a
Bayesian Network [23], a Factor Graph
[24], a Gaussian Graphical Model [25]
or a Mutual Information derived Influ-
ence Network [26].
IV. Cancer Imaging Cancer imaging is a vast field focused
on non-invasive measurement for
1 2
3 4
5
6 7
8
9
10
11
12 13
14 15
16 17 18 19
20
21 22
23 24 25 26 27 28 29 30 31 32 33 34 35
36
37 38 39 40 41
42 43
44
45 46
47
48 49
50 51
52
53 54 55
56 57
58 59
60
61 62
63
64
65 66
67
68 69 70
71 72
73
74 75 76 77 78 79 80 81
82
83 84 85
86 87
88 89
90 91
92
93
94 95 96 97 98
99 100101102103104105 106107108
109110111
112113
114
115116
117
118119120
121
122
123
124
125126127128129
130131
132133
134
135
136137
138
139
140
141
142
143144
145
146
147148149150151
152153
154
155
156
157
158159160161
162
163
164165
166
167
168
169170
171
172
173
174
175
176
177
178
179180
181
182
183
184
185
186
187
188
189190
191
192
193
194195196
197
198
199 200
201
202203204
205
206
207
208 209
210
211
212
213
214
215
216
217
218219
220
221
222
223224225226
227
228229
230
231232
233
234
235
236
237
238
239240
241
242
243
244
245
246
247248 249
250
251
252253254255256
257
258
259260
261262263264265266
267
268
269
270271
272273
274
275
276
277
278
279
280
281282
283
284
285
286
287
288
289
290
291292
293
294
295
296297298
299
300
301
302
303
304
305306
307308
310311
312
313
314
315
316
317318
319
320
321
322
323
324
325
326
327328
329
330331332
333
335336
337
338339
340
341
342
343
344
345346347348
349
350351
352353
354
355
356357
358
359
360
361362363
364
365366367368
369370
371
−0.1 0 0.1 0.2
−0.2
−0.15
−0.1
−0.05
0
0.05
0.1
0.15
0.2
40.5
41
41.5
42
42.5
Distortion
309
334
FIGURE 1 An illustration of meta-clustering. Each point on the map represents a clustering solution and close points represent similar solutions. Colors indicate a measure of quality for the solutions and red circles highlight local stability [15].
Authorized licensed use limited to: LIVERPOOL JOHN MOORES UNIVERSITY. Downloaded on February 12, 2010 at 05:38 from IEEE Xplore. Restrictions apply.
![Page 3: DataMiningInCancerResearch IEEE](https://reader036.vdocuments.mx/reader036/viewer/2022081908/55cf9bf9550346d033a81340/html5/thumbnails/3.jpg)
16 IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE | FEBRUARY 2010
cancer detection, diagnosis and grading,
and to monitor response to therapy by
measuring tumor size and detecting re-
growth. Having started out as mainly
anatomical measurement for anomaly
detection as an indicator of malignancy,
for instance using ultrasound and Mag-
netic Resonance Imaging (MRI), can-
cer imaging has developed into multiple
modalities of physiological imaging
including spectroscopy, with Magnetic
Resonance Spectroscopy (MRS) and
spatially localized Chemical Shift Imag-
es (CSI), with Positron-Emission
Tomography (PET) and Single-Photon
emission CT (SPECT) clinically widely
used to measure glucose metabolism
and therefore find regions with excep-
tionally high energetic rates, which are
characteristic of tumor growth. An
accessible review of machine learning
applications to detection and diagnosis
of disease may be found in [27].
A recent development is the use of
specific functional imaging for tumor
delineation, that is to say, to measure
the contour during surgical planning
and to identify tumor striations that
may not be apparent clinically even
during re-section. Initial efforts in this
direction applied blind signal separation
to single-voxel spectra to separate infil-
trating tumor mass compared with
necrotic tissue [28]. However, tumors
are not homogeneous, often mixing
different types of tissue and sometimes
with different grades of malignancy.
One way to resolve this mixture is to
segment the tumor composition, which
is called nosological imaging, of which
examples of diagnostic applications in
brain are reported in [29], [30]. A relat-
ed and very important clinical area is
follow-up imaging to determine tumor
progression and to differentiate recur-
rent tumor growth from treatment-in-
duced tissue changes [31], [32].
V. Prognostic OutcomeOutcome modeling usually refers to
the analysis of patterns of mortality and
recurrence in longitudinal cohort stud-
ies, where a specific groups of patients
followed-up for up to 20 years, record-
ing the occurrence of events of interest
which may be cancer specific mortality,
overall mortality or local and distal
recurrence. The value of these studies is
closely linked to the design of the fol-
low-up process, highlighting the need
to involve biostatistical expertise from
the very earliest stages in study design.
It is required to model the likeli-
hood of the event of interest, condi-
tional on the event not taking place at
the start of the time interval and taking
into account the loss of patients to fol-
low-up for reasons other than the event
of interest, termed censorship.
The application in medical statistics of
linear-in-the-parameters models to non-
linear effects in outcome data generally
involves discretising the data space to gen-
erate piecewise linear approximations to
the desired response surfaces. However,
this heuristic approach can have important
consequences in cancer research, making
it difficult to mine the hazard function for
insights into the dynamics of metastatic
spread [33] and inducing a categorisation
of the histological grade of the tumor,
which can give rise to important sub-
jective effects that impact on the choice
of therapy given to the patient [34].
Neural network models fit non-lin-
ear, time dependent estimates of the
hazard function, without the need to
make proportionality assumptions or to
discretize the state-space [35], [36] and
their accuracy has been established by
multi-centre, double-blind, evaluation
[37] and this framework has recently
been extended for estimating the joint
model for more than one risk, e.g. local
and distal recurrence [38].
VI. Conclusion Computational intelligence methodol-
ogy has been widely applied to data
–250
–200
–150
–100 –5
0 0 50 100
150
200
250
–350
–300
–250
–200
–150
–100
–50
0
50
100
150
8
24
1812219
11923
Projection onto Axis 1
(a)
–250
–200
–150
–100 –5
0 0 50 100
150
200
250
Projection onto Axis 1
(b)
1020255
Pro
jectio
n o
nto
Axis
2
–500
–400
–300
–200
–100
0
100
200
300
82418
1
2219
119
231020
255
Pro
jectio
n o
nto
Axis
3FIGURE 2 Two views of cluster-based visualization of 25-d protein expression date from breast cancer histology showing (a) the HER-2 positive cohort distinct from the rest and (b) rotating the 3-d view to illustrate the separation between the projections of the remaining seven clusters [18].
Authorized licensed use limited to: LIVERPOOL JOHN MOORES UNIVERSITY. Downloaded on February 12, 2010 at 05:38 from IEEE Xplore. Restrictions apply.
![Page 4: DataMiningInCancerResearch IEEE](https://reader036.vdocuments.mx/reader036/viewer/2022081908/55cf9bf9550346d033a81340/html5/thumbnails/4.jpg)
FEBRUARY 2010 | IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE 17
mining in cancer but open questions
remain before we can fully exploit the
potential of molecular biology, to inte-
grate this into routine imaging and
other non-invasive measurements for
screening and early diagnosis, also to
deliver consistency in laboratory mea-
surements especially in histopathology,
and to more accurately model the bal-
ance between benefit and risk of thera-
py for niche groups of patients, down
to the level of the individual.
New data mining methods need to
be developed to handle high-dimen-
sionality and large data volumes, to more
efficiently model the dynamics of deep
molecular pathways with high levels of
noise and small experimental sample
sizes, and also to robustly estimate haz-
ard functions with censored data with
multiple competing risks. Analytical
models need also to be integrated with
clinical expert knowledge and processes,
potentially by mapping into Boolean
filters [39].
The road ahead is long and bumpy.
This article is provided as a sign-post to
the nature of the key methodological
and application challenges in this grow-
ing and exciting field.
AcknowledgmentsThis work was supported in part by the
Biopattern NoE under FP6/2002/IST/1
and IST-2002-508803. Further support
was provided by MICINN research
project TIN2009-13895-C02-01.
References[1] P. J. G. Lisboa, A. F. G. Taktak, “The use of artif icial neural networks in decision support in cancer: A system-atic review,” Neural Netw., vol. 19, pp. 408–415, 2006. [2] P. J. Campbell, et al., “Identif ication of somatically acquired rearrangements in cancer using genome-wide massively parallel paired-end sequencing,” Nature Genet-ics, vol. 40, pp. 722–729, 2008. [3] N. Belacel, Q. C. Wang, and M. Cuperlovic-Culf, “Clustering methods for microarray gene expression data,” Omics, vol. 10, no. 4, pp. 507–531, 2006.[4] D. Jiang, C. Tang, and A. Zhang, “Cluster analysis for gene expression data: A survey,” IEEE Trans. Knowledge Data Eng., vol. 16, no. 11, pp. 1370–1386, 2004.[5] D. Wunsch and R. Xu, Clustering (IEEE Press Series on Computational Intelligence). Washington, DC: IEEE Com-puter Society Press, 2008. [6] F. Murtagh, G. Downs, and P. Contreras, “Hierarchi-cal clustering of massive, high dimensional data sets by
Thyroid Cancer
Normal Thyrocyte
Normal Thyrocyte
Follicular Adenoma
Follicular Carcinoma
Undifferentiated
Carcinoma
05216 3/31/09
© Kanehisa Laboratories
Papillary
Carcinoma
TRK
REDPTC
Ras
Ras
BRAF
BRAF
PPFP
PPARγ
RXR
BRAFMEK ERK
MEK ERK
Thyroid Follicular Epithelial Cell
Thyroid Follicular Epithelial Cell
+p
+p
+p +p
+p
DNA
ProliferationSurvival
DNA
ProliferationSurvival
DNA
Proliferation
Proliferation
DNA
DNA
DNA
Damage
GenomicInstability
Impaired GICycle Arrest
ReducedApoptosis
MAPK SignalingPathway
p53 SignalingPathway
PPAR SignalingPathway
Ligands
p53
c-Myc
CyclinD1TCF/LEFEACD β-Catenin
AdherensJunction
Wnt SignalingPathway
Downregulation
IncreasedInvasion andCell Motility
Genetic Alterations
Oncogenes: RET/PTC1, RET/PTC3,TRK (TRM3/NTRK1),TRK-T1, T2 (TPR/NTRK1),TRK-T3 (TFG/NTRK1),BRAF, Ras,
PPFP (PAX8/PPARγ), β-Catenin
Tumor Suppressors: p53
FIGURE 3 An excerpt of the papillary thyroid carcinoma pathway from the KEGG database.
Authorized licensed use limited to: LIVERPOOL JOHN MOORES UNIVERSITY. Downloaded on February 12, 2010 at 05:38 from IEEE Xplore. Restrictions apply.
![Page 5: DataMiningInCancerResearch IEEE](https://reader036.vdocuments.mx/reader036/viewer/2022081908/55cf9bf9550346d033a81340/html5/thumbnails/5.jpg)
18 IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE | FEBRUARY 2010
exploiting ultrametric embedding,” SIAM J. Sci. Com-put., vol. 30, pp. 707–730, 2008. [7] S. C. Madeira and A. L. Oliveira, “Biclustering al-gorithms for biological data analysis: A survey,” IEEE/ACM Trans. Computat. Biol. Bioinform., vol. 1, no. 1, pp. 24–45, 2004.[8] L. Carrivick, et al., Identif ication of prognostic sig-natures in breast cancer microarray data using Bayesian techniques,” J. R. Soc. Interface, vol. 3, no. 8, pp. 367–381, 2006. [9]. A. Ben-Hur, A. Elisseeff, and I. Guyon, “A stabil-ity based method for discovering structure in clus-tered data,” in Biocomputing (Proc. Pacific Symp.), vol. 7, R. B. Altman and K. Lauderdalc, Eds. Kauai, Hawaii, USA, 2002, pp. 6–17.[10] M. K. Kerr and G. A. Churchill, “Bootstrapping cluster analysis: Assessing the reliability of conclusions from microarray experiments,” Proc. Natl. Acad. Sci., vol. 98, pp. 8961–8965, 2001. [11] L. I. Kuncheva and D. P. Vetrov, “Evaluation of sta-bility of k-means cluster ensembles with respect to ran-dom initialization,” IEEE Trans. Pattern Anal. Machine Intell., vol. 28, no. 11, pp. 1798–1808, 2006.[12] A. Gionis, H. Mannila, and P. Tsaparas, “Cluster-ing aggregation,” ACM Trans. Knowl. Discov. Data, vol. 1, no. 1, 2007. [13] N. Nguyen and R. Caruana, “Consensus clustering,” in Proc. 7th IEEE Int. Conf. Data Mining (ICDM), 2007, pp. 607–612.[14] R. Caruana, M. Elhawary, N. Nguyen, and C. Smith, “Meta clustering,” in Proc. 6th IEEE Int. Conf. Data Mining (ICDM), 2006, pp. 107–118.[15] A. Ciaramella, et al., “Interactive data analysis and clustering of genomic data,” Neural Netw., vol. 21, pp. 368–378, 2008. [16] J. A. Lee and M. Verleysen, Nonlinear Dimensionality Reduction. New York: Springer-Verlag, 2007. [17] J. Vesanto, “SOM-based data visualization methods,” Intell. Data Anal., vol. 3, pp. 111–126, 1999.
[18] P. J. G. Lisboa, I. O. Ellis, A. R. Green, F. Ambrogi, and M. B. Dias, “Cluster-based visualisation with scat-ter matrices,” Pattern Recogn. Lett., vol. 29, no. 13, pp. 1814–1823, 2008.[19] J. Hasty, D. McMillen, F. Isaacs, and J. J. Collins, “Com-putational studies of gene regulatory networks: In numero molecular biology,” Nature Rev. Gene., vol. 2, pp. 268–279, 2001. [20] H. de Jong, “Modeling and simulation of genetic regulatory systems: A literature review,” J. Computat. Biol., vol. 9, no. 1, pp. 67–103, 2002.[21] E. J. Borowski and J. M. Borwein, Computational Modeling of Genetic and Biochemical Networks. Cambridge, MA: MIT Press, 2001.[22] N. Friedman, “Inferring cellular networks using prob-abilistic graphical models,” Science, vol. 303, p. 799, 2004. [23] C. Needham, J. Bradford, A. Bulpitt, and D. West-head, “A primer on learning in Bayesian networks for computational biology,” PLOS Computat. Biol., vol. 3, no. 8, pp. 1409–1416, 2007.[24] I. Gat-Viks, A. Tanay, D. Raijaman, and R. Shamir, “A probabilistic methodology for integrating knowledge and experiments on biological networks,” J. Computat. Biol., vol. 13, no. 2, pp. 165–181, 2006.[25] J. Schafer and K. Strimmer, “An empirical Bayes ap-proach to inferring large-scale gene association networks,” Bioinformatics, vol. 21, no. 6, pp. 754–764, 2005.[26] K. Basso, et al., “Reverse engineering of regulatory networks in human B cells,” Nature Gene., vol. 37, no. 4, pp. 382–290, 2005. [27] P. Sajda, “Machine learning for detection and di-agnosis of disease,” Annu. Rev. Biomed. Eng., vol. 8, pp. 537–565, 2006. [28] Y. Huang, P. J. G. Lisboa, and W. El-Deredy, “Tu-mour grading from magnetic resonance spectroscopy: A comparison of feature extraction with variable selection,” Stat. Med., vol. 22, no. 1, pp. 147–164, 2003.[29] F. Szabo De Edelenyi, et al., “A new approach for analyzing proton magnetic resonance spectroscopic im-
ages of brain tumors: Nosologic images,” Nature Med., vol. 6, pp. 1287–1289, 2000. [30] M. De Vos, T. Laudadio, A.W. Simonetti, A. Heerschap, and S. Van Huffel, “Fast nosologic imaging of the brain,” J. Magnet. Reson., vol. 184, no. 2, pp. 292–301, 2007.[31] A. H. Jacobs, et al., “Imaging in neurooncology,” J. Amer. Soc. Exp. NeuroTherapeu., vol. 2, pp. 333–347, 2005. [32] S. Cha, “Neuroimaging in neuro-oncology,” J. Amer. Soc. Exp. NeuroTherapeu., vol. 6, pp. 465–477, 2009. [33] R. Demicheli, P. Valagussa, and G. Bonadonna, “Double-peaked time distribution of mortality for breast cancer patients undergoing mastectomy,” Breast Cancer Res. Treat., vol. 75, pp. 127–134, 2002. [34] J. A. Woolgar, “The predictive value of detailed histolog-ical staging of surgical resection specimens in oral cancer,” in Outcome Prediction in Cancer, A. F. G. Taktak and A. C. Fisher, Eds. U.K.: Elsevier, 2007, pp. 3–26.[35] E. Biganzoli, P. Boracchi, L. Mariani, and E. Marubini, “Feed forward neural networks for the analysis of censored survival data: A partial logistic regression approach,” Stat. Med., vol. 17, pp. 1269–1206, 1998. [36] A. Eleuteri, R. Tagliaferri, L. Milano, S. De Placido, and M. De Laurentiis, “A novel neural network-based survival analysis model,” Neural Netw., vol. 16, no. 5–6, pp. 855–864, 2003.[37] A. Taktak, et al., “Double-blind evaluation and bench-marking of survival models in a multi-centre study,” Comput. Biol. Med., vol. 37, no. 8, pp. 1108–1120, 2007. [38] P. J. G. Lisboa, et al., “Partial logistic artif icial neural network for competing risks regularised with automatic relevance determination,” IEEE Trans. Neural Netw., vol. 20, no. 9, pp. 1403–1416, 2009. [39] T. Rögnvaldsson, T. A. Etchells, L. You, D. Garwicz, I. H. Jarman, and P. J. G. Lisboa, “How to find simple and accurate rules for viral protease cleavage specificities,” BMC Bioinform., vol. 10, pp. 149, 2009.
From Imagination to Market
IEEE Expert NowThe Best of IEEE Conferences and Short Courses
Free Trial!Experience IEEE – request a trial for your company.
www.ieee.org/expertnow
An unparalleled education resource that provides the latestin related technologies.
Keep up-to-date on the latest trends in related technologies Interactive content via easy-to-use player-viewer, audio and
video fi les, diagrams, and animations Increases overall knowledge beyond a specifi c discipline 1-hour courses accessible 24/7
“With IEEE, we have24/7 access to the technical informationwe need exactly whenwe need it.”
– Dr. Bin Zhao, Senior Manager, RF/Mixed Signal Design Engineering, Skyworks Solutions
07-PIM-0181f Expert Now Half.ind1 1 7/10/07 10:40:27 AM
Authorized licensed use limited to: LIVERPOOL JOHN MOORES UNIVERSITY. Downloaded on February 12, 2010 at 05:38 from IEEE Xplore. Restrictions apply.