Computer aided fuzzy medical diagnosis

Centre for Computational Intelligence, Department of Computer Science, De Montfort University, The Gateway, Leicester LE1 9BH, UK
Inf. Sci 05/2004; 162(2):81-104. DOI: 10.1016/j.ins.2004.03.003
Source: DBLP

ABSTRACT This paper describes a fuzzy approach to computer aided medical diagnosis in a clinical context. It introduces a formal view of diagnosis in clinical settings and shows the relevance and possible uses of fuzzy cognitive maps and fuzzy logic. A constraint satisfaction method is introduced which uses the temporal uncertainty in symptom durations that may occur with particular diseases. Together with fuzzy symptom descriptions, the method results in an estimate of the stage of a disease if the temporal constraints of the disease in relation to the occurrence of the symptoms are satisfied. The approach is evaluated through simulation experiments showing the effects of symptom ordering, temporal uncertainty and symptom strengths on the diagnosis efficiency. The method is effective and can be developed further using second order (Type 2) fuzzy logic to better represent uncertainty in the clinical context thus improving differential diagnosis accuracy.

1 Follower
8 Reads
  • Source
    • "Iakovidis and Papageorgio (2011) propose to use intuitionistic fuzzy cognitive maps to account for the hesitancy in the doctors' evaluation on the relationship between symptoms and possible diseases. Innocent and John (2004) developed a similar system using type-2 fuzzy sets and they found that accounting for this higher level of imprecision can improve the accuracy of diagnosis. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Mobile applications and specifically wellness applications are used increasingly by different age-segments of the general population. This is facilitated by the large amount of data collected through various built-in sensors in the smartphone or other mobile devises, e.g. smart watches. Young-elderly cohort (60-75 year old individual) is probably one of the most potential user groups that would benefit from using mobile health and wellness applications, if their needs and preferences are precisely addressed. General knowledge is limited on understanding to what extent mobile wellness applications can and should provide precise recommendations which improve the users’ health and physical conditions. To address this problem, the current study identifies the potential benefits of utilizing fuzzy optimization tools to design recommendation systems that can take into consideration the (i) imprecision in the data and (ii) the imprecision by which one can estimate the effect of a recommendation on the user of the system. The proposed approach, depending on the context of use, identifies a set of actions to be taken by the users in order to optimize the physical or mental condition from various perspectives. The model is illustrated through the example of walking speed optimization which is an important issue for the young-elderly.
    The 28th Bled eConference #Wellbeing, Bled, Slovenia; 06/2015
  • Source
    • "The human diagnosis is based on observations, thus uncertain observations may hinder diagnosis accuracy. Several techniques exist to treat such uncertainty: the Bayesian network based approaches [5] [20] [22] [24], the probability based approaches [7] [17] [18], the subjective evaluation based approaches [41], the belief based approaches [25] [30] [32], the possibility based approaches [6] [12] [14] [15], the evidential network based approaches [3] [19], etc. This paper is about uncertainty treatment using an evidential network for multiviewpoint abductive diagnosis. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The paper proposes an approach to support human abductive reasoning in the diagnosis of a multiviewpoint system. The novelty of this work lies on the capability of the approach to treat the uncertainty held by the agent performing the diagnosis. To do so, we make use of evidential networks to represent and propagate the uncertain evidence gathered by the agent. Using forward and backward propagation of the information, the impact of the evidence over the different symptoms and causes of failure is quantified. The agent can then make use of this information as additional hints in an iterative diagnosis process until a desired degree of certainty is obtained. The model is compared with a deterministic one in which evidence is represented by binary states, that is, a symptom is either observed or not.
    Information Sciences 12/2013; 253:110-125. DOI:10.1016/j.ins.2013.07.014 · 4.04 Impact Factor
  • Source
    • "Expert System is software which has the ability to replicate the thinking and reasoning capacity of humans based on some facts and rules presented to it. The use of expert systems finds its place in diverse sectors like medical diagnosis [1], decision support systems [2], biology [3], educational [4], chemical [5], and etc. For example, the use of expert system in medical diagnosis dates back to the early 70's when MYCIN [6], an expert system for identifying bacteria causing diseases was developed at Stanford University. "
    [Show abstract] [Hide abstract]
    ABSTRACT: One of the basic tasks which is responded for head of each university department, is employing lecturers based on some default factors such as experience, evidences, qualifies and etc. In this respect, to help the heads, some automatic systems have been proposed until now using machine learning methods, decision support systems (DSS) and etc. According to advantages and disadvantages of the previous methods, a full automatic system is designed in this paper using expert systems. The proposed system is included two main steps. In the first one, the human expert's knowledge is designed as decision trees. The second step is included an expert system which is evaluated using extracted rules of these decision trees. Also, to improve the quality of the proposed system, a majority voting algorithm is proposed as post processing step to choose the best lecturer which satisfied more expert's decision trees for each course. The results are shown that the designed system average accuracy is 78.88. Low computational complexity, simplicity to program and are some of other advantages of the proposed system.
Show more