Escolar Documentos
Profissional Documentos
Cultura Documentos
Neural Networks
in Medical Diagnosis
Włodzisław Duch
Dept. of Informatics,
Nicholas Copernicus University,
Toruń, Poland
What is it about?
• Data is precious! But also overwhelming ...
• Statistical methods are important but new
techniques may frequently be more accurate
and give more insight into the data.
• Data analysis requires intelligence.
• Inspirations come from many sources,
including biology: artificial neural networks,
evolutionary computing, immune systems ...
Computational Intelligence
Neural
networks Evolutionary
Pattern algorithms
Recognition
Visuali-
Computational Intelligence zation
Fuzzy Data + Knowledge
logic Artificial Intelligence
Multivariate
statistics
Expert
systems Machine Probabilistic
learning methods
What do these methods do?
• Provide non-parametric models of data.
• Allow to classify new data to pre-defined
categories, supporting diagnosis & prognosis.
• Allow to discover new categories.
• Allow to understand the data, creating fuzzy
or crisp logical rules.
• Help to visualize multi-dimensional
relationships among data samples.
• Help to model real neural networks!
GhostMiner
GhostMiner Philosophy
Philosophy
GhostMiner, data mining tools from our lab.
Synapses Synapses
(weights)
Axon
Neural
Neural network
network for
for MI
MI diagnosis
diagnosis
~ p(MI|X) 0.7 Myocardial Infarction
Output
weights
Input
weights
Inputs: 65 1 5 3 1
5 o 6 i
FMI X Wij W jk X k
i 1 k 1
Such networks are universal approximators:
they may learn any mapping X => Y
Learning
Learning dynamics
dynamics
Decision regions shown every 200 training epochs in x3, x4
coordinates; borders are optimally placed with wide margins.
Neurofuzzy
Neurofuzzy system
systemss
Fuzzy: x(no/yes) replaced by a degree
x. Triangular, trapezoidal, Gaussian ... MF.
M.f-s in many
dimensions:
Input
layer
Output:
one node
per class.
Single rule:
IF (nodes-involved [0,2] degree-malignant = 3
THEN recurrence, ELSE no-recurrence
Interactive multidimensional
scaling: zooming on the new
case to inspect its similarity to
other cases.
Summary
Neural networks and other computational
intelligence methods are useful additions to the
multivariate statistical tools.
They support diagnosis, predictions, and data
understanding: extracting rules, prototypes.