- author = "Ahmet Yardimci",
- title = "Soft computing in medicine",
- journal = "Applied Soft Computing",
- volume = "9",
- number = "3",
- pages = "1029--1043",
- year = "2009",
- ISSN = "1568-4946",
-
DOI = "
doi:10.1016/j.asoc.2009.02.003",
-
URL = "
http://www.sciencedirect.com/science/article/B6W86-4VRP213-2/2/141a98fa1c9600c49c0883f71eb8a711",
- keywords = "genetic algorithms, genetic programming, Soft computing, Fuzzy-neural systems, Medicine",
-
abstract = "Soft computing (SC) is not a new term; we have gotten
used to reading and hearing about it daily. Nowadays,
the term is used often in computer science and
information technology. It is possible to define SC in
different ways. Nonetheless, SC is a consortium of
methodologies which works synergistically and provides,
in one form or another, flexible information processing
capability for handling real life ambiguous situations.
Its aim is to exploit the tolerance for imprecision,
uncertainty, approximate reasoning and partial truth in
order to achieve tractability, robustness and low-cost
solutions. SC includes fuzzy logic (FL), neural
networks (NNs), and genetic algorithm (GA)
methodologies. SC combines these methodologies as FL
and NN (FL-NN), NN and GA (NN-GA) and FL and GA
(FL-GA). Recent years have witnessed the phenomenal
growth of bio-informatics and medical informatics by
using computational techniques for interpretation and
analysis of biological and medical data. Among the
large number of computational techniques used, SC,
which incorporates neural networks, evolutionary
computation, and fuzzy systems, provides unmatched
utility because of its demonstrated strength in
handling imprecise information and providing novel
solutions to hard problems. The aim of this paper is to
introduce briefly the various SC methodologies and to
present various applications in medicine between the
years 2000 and 2008. The scope is to demonstrate the
possibilities of applying SC to medicine-related
problems. The recent published knowledge about use of
SC in medicine is researched in MEDLINE. This study
detects which methodology or methodologies of SC are
used frequently together to solve the special problems
of medicine. According to MEDLINE database searches,
the rates of preference of SC methodologies in medicine
were found as 68percent of FL-NN, 27percent of NN-GA
and 5percent of FL-GA. So far, FL-NN methodology was
significantly used in medicine. The rates of using
FL-NN in clinical science, diagnostic science and basic
science were found as percent83, percent71 and
percent48, respectively. On the other hand NN-GA and
FL-GA methodologies were mostly preferred by basic
science of medicine.
Another message emerging from this survey is that the number of papers which used NN-GA methodology has continuously risen until today. Also search results put the case clearly that FL-GA methodology has not applied well enough to medicine yet. Undeniable interest in studying SC methodologies in genetics, physiology, radiology, cardiology, and neurology disciplines proves that studying SC is very fruitful in these disciplines and it is expected that future researches in medicine will use SC more than it is used today to solve more complex problems.",
- notes = "Only brief mention of GP",
