Computational Intelligence in Early Diabetes Diagnosis: A Review

Department of Biotechnology, Birla Institute of Technology, Mesra, Ranchi, India.
The Review of Diabetic Studies 01/2010; 7(4):252-62. DOI: 10.1900/RDS.2010.7.252
Source: PubMed


The development of an effective diabetes diagnosis system by taking advantage of computational intelligence is regarded as a primary goal nowadays. Many approaches based on artificial network and machine learning algorithms have been developed and tested against diabetes datasets, which were mostly related to individuals of Pima Indian origin. Yet, despite high accuracies of up to 99% in predicting the correct diabetes diagnosis, none of these approaches have reached clinical application so far. One reason for this failure may be that diabetologists or clinical investigators are sparsely informed about, or trained in the use of, computational diagnosis tools. Therefore, this article aims at sketching out an outline of the wide range of options, recent developments, and potentials in machine learning algorithms as diabetes diagnosis tools. One focus is on supervised and unsupervised methods, which have made significant impacts in the detection and diagnosis of diabetes at primary and advanced stages. Particular attention is paid to algorithms that show promise in improving diabetes diagnosis. A key advance has been the development of a more in-depth understanding and theoretical analysis of critical issues related to algorithmic construction and learning theory. These include trade-offs for maximizing generalization performance, use of physically realistic constraints, and incorporation of prior knowledge and uncertainty. The review presents and explains the most accurate algorithms, and discusses advantages and pitfalls of methodologies. This should provide a good resource for researchers from all backgrounds interested in computational intelligence-based diabetes diagnosis methods, and allows them to extend their knowledge into this kind of research.

Download full-text


Available from: Shankaracharya
  • Source
    • "ANNs have proven useful in the analysis of blood and urine samples of diabetic patients[20], leukemia classification[21]and EEG recording for prediction of epileptic[22]. In addition, ANNs have also been applied in the diagnosis of colorectal cancer[23], colon cancer[24]and early diabetes[25]. In the present research, we use two different architectures of supervised artificial neural networks, namely multi-layer perceptron (MLP) with backpropagation and radial basis function RBF networks to discriminate EEG power spectra from the four different environmental shapes. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The present work introduces a new method for discriminating electroencephalogram (EEG) power spectra of rat's brain housed in different architectural shapes. The ability of neural networks for discrimination is used to describe the effect of different environments on brain activity. In this research the rats were divided into four groups according to the type of environmental shapes as: control (normal cage), pyramidal, inverted pyramidal and circular. The brain activities (EEG) were recorded from rats of each group. Fast Fourier Transform (FFT) analysis of EEG signals was carried out to obtain power spectra. Two different neural networks are used as classifiers for power spectra of the different 4 groups: multi-layer perceptron (MLP) with backpropagation and radial basis function RBF networks with unsupervised K means clustering algorithm. Experimental studies have shown that the proposed algorithms give good results when applied and tested on the four groups. The multilayer with backpropagation and radial basis function networks achieved a performance rate reaching 94.4 % and 96.67% respectively.
    Full-text · Article · Dec 2015
  • Source
    • "The early contributions can be found on the neural networks, it provides a new significant way for intelligent medical diagnosis. Based on this idea, artificial neural networks have been applied in the diagnosis of: (i) pancreatic disease [1], (ii) gynecological diseases [2], (iii) early diabetes [3], (iv) colorectal cancer [4], and (v) multiple sclerosis lesions [5]. While this kind of method to set up the achievements of medical diagnostic system is still limited, the main reasons are that the learning algorithm cannot calculate the right results when the required algorithm to set up neural network model solves the larger, multi-features disease diagnosis problems. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background: According to the analysis of the challenges faced by the current public health circumstances such as the sharp increase in elderly patients, limited medical personnel, resources and technology, the agent-based intelligent medical diagnosis system for patients (AIMDS) is proposed in this research. Objective: Based on advanced sensing technology and professional medical knowledge, the AIMDS can output the appropriate medical prescriptions and food prohibition when the physical signs and symptoms of the patient are inputted. Methods: Three core modules are designed include sensing module, intuition-based fuzzy set theory/medical diagnosis module, and medical knowledge module. Results: The result shows that the optimized prescription can reach the desired level, with great curative effect for patient disease, through a case study simulation. Conclusion: The presented AIMDS can integrate sensor technique and intelligent medical diagnosis methods to make an accurate diagnosis, resulting in three-type of optimized descriptions for patient selection.
    Preview · Article · Jun 2015 · Technology and health care: official journal of the European Society for Engineering and Medicine
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The concept of classification of clinical data can be utilized in the development of an effective diagnosis system by taking the advantage of computational intelligence. Diabetes disease diagnosis via proper interpretation of the diabetes data is an important problem in neural networks. Unfortunately, although several classification studies have been carried out with significant performance, many of the current methods often fail to reach out to patients. Graphical user interface-enabled tools need to be developed through which medical practitioners can simply enter the health profiles of their patients and receive an instant diabetes prediction with an acceptable degree of confidence. In this study, the neural network approach was used for a dataset of 768 persons from a Pima Indian population living near Phoenix, AZ. A neural network mixture of experts model was trained with these data using the expectation-minimization algorithm. The mixture of experts method was used to train the algorithm with 97% accuracy. A graphical user interface was developed that would work in conjunction with the trained network to provide the output in a presentable format. This study provides a machine-implementable approach that can be used by physicians and patients to minimize the extent of error in diagnosis. The authors are hopeful that replication of results of this study in other populations may lead to improved diagnosis. Physicians can simply enter the health profile of patients and get the diagnosis for diabetes type 2.
    Full-text · Article · Nov 2011 · Diabetes Technology & Therapeutics
Show more