PMID- 29625186
OWN - NLM
STAT- MEDLINE
DCOM- 20191030
LR  - 20191030
IS  - 1532-0480 (Electronic)
IS  - 1532-0464 (Linking)
VI  - 83
DP  - 2018 Jul
TI  - Bimodal fuzzy analytic hierarchy process (BFAHP) for coronary heart disease risk 
      assessment.
PG  - 204-216
LID - S1532-0464(18)30064-9 [pii]
LID - 10.1016/j.jbi.2018.03.016 [doi]
AB  - Rooted deeply in medical multiple criteria decision-making (MCDM), risk 
      assessment is very important especially when applied to the risk of being 
      affected by deadly diseases such as coronary heart disease (CHD). CHD risk 
      assessment is a stochastic, uncertain, and highly dynamic process influenced by 
      various known and unknown variables. In recent years, there has been a great 
      interest in fuzzy analytic hierarchy process (FAHP), a popular methodology for 
      dealing with uncertainty in MCDM. This paper proposes a new FAHP, bimodal fuzzy 
      analytic hierarchy process (BFAHP) that augments two aspects of knowledge, 
      probability and validity, to fuzzy numbers to better deal with uncertainty. In 
      BFAHP, fuzzy validity is computed by aggregating the validities of relevant risk 
      factors based on expert knowledge and collective intelligence. By considering 
      both soft and statistical data, we compute the fuzzy probability of risk factors 
      using the Bayesian formulation. In BFAHP approach, these fuzzy validities and 
      fuzzy probabilities are used to construct a reciprocal comparison matrix. We then 
      aggregate fuzzy probabilities and fuzzy validities in a pairwise manner for each 
      risk factor and each alternative. BFAHP decides about being affected and not 
      being affected by ranking of high and low risks. For evaluation, the proposed 
      approach is applied to the risk of being affected by CHD using a real dataset of 
      152 patients of Iranian hospitals. Simulation results confirm that adding 
      validity in a fuzzy manner can accrue more confidence of results and clinically 
      useful especially in the face of incomplete information when compared with actual 
      results. Applying the proposed BFAHP on CHD risk assessment of the dataset, it 
      yields high accuracy rate above 85% for correct prediction. In addition, this 
      paper recognizes that the risk factors of diastolic blood pressure in men and 
      high-density lipoprotein in women are more important in CHD than other risk 
      factors.
CI  - Copyright (c) 2018 Elsevier Inc. All rights reserved.
FAU - Sabahi, Farnaz
AU  - Sabahi F
AD  - Soft Computing Laboratory, Faculty of Electrical and Computer Engineering, Urmia 
      University, Urmia, Iran. Electronic address: f.sabahi@urmia.ac.ir.
LA  - eng
PT  - Journal Article
DEP - 20180404
PL  - United States
TA  - J Biomed Inform
JT  - Journal of biomedical informatics
JID - 100970413
SB  - IM
MH  - Bayes Theorem
MH  - Clinical Decision-Making
MH  - Computer Simulation
MH  - Coronary Disease/*diagnosis
MH  - *Diagnosis, Computer-Assisted
MH  - Female
MH  - *Fuzzy Logic
MH  - Humans
MH  - Male
MH  - Risk Assessment/*methods
MH  - Risk Factors
MH  - Uncertainty
OTO - NOTNLM
OT  - Coronary Heart Disease (CHD)
OT  - Fuzzy Analytic Hierarchy Process (FAHP)
OT  - Multi-criteria Decision-Making (MCDM)
OT  - Validity
EDAT- 2018/04/07 06:00
MHDA- 2019/10/31 06:00
CRDT- 2018/04/07 06:00
PHST- 2017/07/31 00:00 [received]
PHST- 2018/03/13 00:00 [revised]
PHST- 2018/03/31 00:00 [accepted]
PHST- 2018/04/07 06:00 [pubmed]
PHST- 2019/10/31 06:00 [medline]
PHST- 2018/04/07 06:00 [entrez]
AID - S1532-0464(18)30064-9 [pii]
AID - 10.1016/j.jbi.2018.03.016 [doi]
PST - ppublish
SO  - J Biomed Inform. 2018 Jul;83:204-216. doi: 10.1016/j.jbi.2018.03.016. Epub 2018 
      Apr 4.