Ashok Deshpande

Modeling Paradigms for Medical Diagnostic Decision Support: A Survey and Future Directions

Use of computer based decision tools to aid clinical decision making, has been a primary goal of research in biomedical informatics. Research in the last five decades has led to the development of Medical Decision Support (MDS) applications using a variety of modeling techniques, for a diverse range of medical decision problems. This paper surveys literature on modeling techniques for diagnostic decision support, with a focus on decision accuracy. Trends and shortcomings of research in this area are discussed and future directions are provided. The authors suggest that—(i) Improvement in the accuracy of MDS application may be possible by modeling of vague and temporal data, research on inference algorithms, integration of patient information from diverse sources and improvement in gene profiling algorithms; (ii) MDS research would be facilitated by public release of de-identified medical datasets, and development of opensource data-mining tool kits; (iii) Comparative evaluations of different modeling techniques are required to understand characteristics of the techniques, which can guide developers in choice of technique for a particular medical decision problem; and (iv) Evaluations of MDS applications in clinical setting are necessary to foster physicians’ utilization of these decision aids.

Fuzzy relation based modeling for medical diagnostic decision support: Case studies

This paper investigates a variation to Adlassnigs fuzzy relation based model for medical diagnosis. The proposed model is an attempt to closely replicate a physicians perceptions of symptom-disease associations and his approximate-reasoning for diagnosis. For proof of principle, the algorithm is evaluated in two sample studies. First case study relates to selected cardiovascular diseases, wherein the required parameters are estimated by interviewing physicians, and an evaluation is performed on a dataset of 79 cases. In the second study, the algorithm is implemented using an alternative semiautomatic approach for a more complex problem of diagnosing common infectious diseases, wherein the parameters are derived from a dataset of 92 case records; for evaluation, jack-knife is performed along with a comparison with Independence Bayes, considered here as the reference standard. The proposed algorithm was found to be as accurate as Independence Bayes for diagnosing common infectious diseases from the small dataset. This result may indicate the utility of proposed algorithm to optimally model the diagnostic process for small datasets; especially, due to its computational simplicity. Further studies on a variety of datasets are needed to establish such a utility.

Estimating premature mortality attributable to PM2.5 exposure and benefit of air pollution control policies in China for 2020

In past decade of rapid industrial development and urbanization, China has witnessed increasingly persistent severe haze and smog episodes, posing serious health hazards to the Chinese population, especially in densely populated cities. Quantification of health impacts attributable to PM2.5 (particulates with aerodynamic diameter≤2.5μm) has important policy implications to tackle air pollution. The Chinese national monitoring network has recently included direct measurements of ground level PM2.5, providing a potentially more reliable source for exposure assessment. This study reports PM2.5-related long-term mortality of year 2015 in 161 cities of nine regions across China using integrated exposure risk (IER) model for PM2.5 exposure-response functions (ERF). It further provides an estimate of the potential health benefits by year 2020 with a realization of the goals of Air Pollution Prevention and Control Action Plan (APPCAP) and the three interim targets (ITs) and Air Quality Guidelines (AQG) for PM2.5 by the World Health Organization (WHO). PM2.5-related premature mortality in 161 cities was 652 thousand, about 6.92% of total deaths in China during year 2015. Among all premature deaths, contributions of cerebrovascular disease (stroke), ischemic heart disease (IHD), chronic obstructive pulmonary disease (COPD), lung cancer (LC) and acute lower respiratory infections (ALRIs) were 51.70, 26.26, 11.77, 9.45 and 0.82%, respectively. The premature mortality in densely populated cities is very high, such as Tianjin (12,533/year), Beijing (18,817/year), Baoding (10,932/year), Shanghai (18,679/year), Chongqing (23,561/year), Chengdu (11,809/year), Harbin (9037/year) and Linyi (9141/year). The potential health benefits will be 4.4, 16.2, 34.5, 63.6 and 81.5% of the total present premature mortality when PM2.5 concentrations in China meet the APPCAP, WHO IT-1, IT-2, IT-3 and AQG respectively, by the year 2020. In the current situation, by the end of year 2030, even if Chines government fulfills its own target to meet national ambient air quality standard of PM2.5 (35μg/m3), total premature mortality attributable to PM2.5 will be 574 thousand across 161 cities. The present methodology will greatly help policy makers and pollution control authorities to further analyze cost and benefits of air pollution management programs in China.

Disability-adjusted life years and economic cost assessment of the health effects related to PM2.5 and PM10 pollution in Mumbai and Delhi, in India from 1991 to 2015

Particulate air pollution is becoming a serious public health concern in urban cities in India due to air pollution-related health effects associated with disability-adjusted life years (DALYs) and economic loss. To obtain the quantitative result of health impact of particulate matter (PM) in most populated Mumbai City and most polluted Delhi City in India, an epidemiology-based exposure–response function has been used to calculate the attributable number of mortality and morbidity cases from 1991 to 2015 in a 5-year interval and the subsequent DALYs, and economic cost is estimated of the health damage based on unit values of the health outcomes. Here, we report the attributable number of mortality due to PM10 in Mumbai and Delhi increased to 32,014 and 48,651 in 2015 compared with 19,291 and 19,716 in year 1995. And annual average mortality due to PM2.5 in Mumbai and Delhi was 10,880 and 10,900. Premature cerebrovascular disease (CEV), ischemic heart disease (IHD), and chronic obstructive pulmonary disease (COPD) causes are about 35.3, 33.3, and 22.9% of PM2.5-attributable mortalities. Total DALYs due to PM10 increased from 0.34 million to 0.51 million in Mumbai and 0.34 million to 0.75 million in Delhi from average year 1995 to 2015. Among all health outcomes, mortality and chronic bronchitis shared about 95% of the total DALYs. Due to PM10, the estimated total economic cost at constant price year 2005 US

Fuzzy fault tree analysis: revisited

increased from 2680.87 million to 4269.60 million for Mumbai City and 2714.10 million to 6394.74 million for Delhi City, from 1995 to 2015, and the total amount accounting about 1.01% of India’s gross domestic product (GDP). A crucial presumption is that in 2030, PM10 levels would have to decline by 44% (Mumbai) and 67% (Delhi) absolutely to maintain the same health outcomes in year 2015 levels. The results will help policy makers from pollution control board for further cost–benefit analyses of air pollution management programs in Mumbai and Delhi.

Can Fuzzy Logic Bring Complex Environmental Problems into Focus

Fault tree analysis is one of the most effective techniques for estimating the frequency of occurrence of hazardous events in probabilistic risk assessment study. The analyst needs to study a multi component system and identifies vulnerable sections of hazardous plant based on formalized procedure. In this sequel, we have made an attempt to demonstrate the effective ness of using fuzzy fault tree formalism over the conventional approach and the concept is termed as Fuzzy Top Event Probability (FTEP). The case study 1 relates to flash vessel in an ammonia tank, while in case study 2, using fuzzy set theory, the FTEP has been estimated for the existing ammonia storage tank in a large fertilizer complex located in Mumbai India. In addition, a new formalism for estimating the possibility of a fuzzy event is presented with application.

Evaluation of Fuzzy Relation Method for Medical Decision Support

In everyday life and field such as environmental health /environmental impacts people deal with concepts that involve factors that defy classification into crisp sets safe/minimal, harmful/ very high negative impacts, acceptable with mitigation measures, and so on. A classic example is a regulator carefully explaining the result of a detailed quantitative risk assessment/environmental impact assessment report to a community group, only to be asked over and over again. But are we safe? / But are environmental impacts minimal? In this case, safe/minimal defies crisp classification because it is a multivariate state with gradation that varies among different individuals and groups. Furthermore, it is hard to define the terms like health, environment, and hazardous, safe, air and water quality, risk and alike as these are vague or fuzzy terms based on perception. In July 1964, Professor Lotfi Zadeh, while working on the problems in pattern classification and system analysis thought of the use of imprecise categories for classification, and the idea of grade of membership, which is the concept that became the backbone of fuzzy set theory, occurred to him then. This important event led to the publication of his seminal paper: Fuzzy Sets (1965) and the birth of fuzzy logic technology. In this sequel, we consider how fuzzy logic applies to two important issues of environment management systems: 1] river water quality classification and 2] ranking of industries based on their hazardous pollution potential. The presentation is primarily centered on fuzzy sets and fuzzy rule based systems, aimed at straightly defining one of the components of environmental quality straightway in linguistic terms with degree of certainty. Would decision makers and the public accept expressions environmental quality goals in straightway linguistic terms with computed degrees of certainty? Resistance is likely. In many regions, such as the United States and European Union, both decision makers and members of the public seem more comfortable with the current system—in which government agencies avoid confronting uncertainties by setting guidelines that are crisp and often fail to communicate uncertainty. Perhaps someday a more comprehensive approach that includes exposure surveys, toxicological data, and epidemiological studies coupled with fuzzy modeling (could be termed as hybrid fuzzy- Probability modeling) will go a long way toward resolving some of the conflict, divisiveness, and controversy in the current regulatory paradigm.

Maintenance of industrial equipment

The potential of computer based tools to assist physicians in medical decision making, was envisaged five decades ago. Apart from factors like usability, integration with work-flow and natural language processing, lack of decision accuracy of the tools has hindered their utility. Hence, research to develop accurate algorithms for medical decision support tools, is required. Pioneering research in last two decades, has demonstrated the utility of fuzzy set theory for medical domain. Recently, Wagholikar and Deshpande proposed a fuzzy relation based method (FR) for medical diagnosis. In their case studies for heart and infectious diseases, the FR method was found to be better than naive bayes (NB). However, the datasets in their studies were small and included only categorical symptoms. Hence, more evaluative studies are required for drawing general conclusions. In the present paper, we compare the classification performance of FR with NB, for a variety of medical datasets. Our results indicate that the FR method is useful for classification problems in the medical domain, and that FR is marginally better than NB. However, the performance of FR is significantly better for datasets having high proportion of unknown attribute values. Such datasets occur in problems involving linguistic information, where FR can be particularly useful. Our empirical study will benefit medical researchers in the choice of algorithms for decision support tools.

Fuzzy naive bayesian model for medical diagnostic decision support

Purpose – The purpose of this paper is to ascertain overall system health and maintenance needs with degree of certainty using condition‐monitoring data with hierarchical fuzzy inference system.Design/methodology/approach – In process plants, equipment condition is ascertained using condition‐monitoring data for each condition indicator. For large systems with multiple condition indicators, estimating the overall system health becomes cumbersome. The decision of selecting the equipment for an overhaul is mostly determined by generic guidelines, and seldom backed up by condition‐monitoring data. The proposed approach uses a hierarchical system health assessment using fuzzy inference on condition‐monitoring data collected over a period. Each subsystem health is ascertained with degree of certainty using degree of match operation performed on fuzzy sets of condition‐monitoring data and expert opinion. Fuzzy sets and approximate reasoning are used to handle the uncertainty/imprecision in data and subjectivity...

Availability assessment of a two-unit stand-by pumping system

This work relates to the development of computational algorithms to provide decision support to physicians. The authors propose a Fuzzy Naive Bayesian (FNB) model for medical diagnosis, which extends the Fuzzy Bayesian approach proposed by Okuda. A physician’s interview based method is described to define a orthogonal fuzzy symptom information system, required to apply the model. For the purpose of elaboration and elicitation of characteristics, the algorithm is applied to a simple simulated dataset, and compared with conventional Naive Bayes (NB) approach. As a preliminary evaluation of FNB in real world scenario, the comparison is repeated on a real fuzzy dataset of 81 patients diagnosed with infectious diseases. The case study on simulated dataset elucidates that FNB can be optimal over NB for diagnosing patients with imprecise-fuzzy information, on account of the following characteristics — 1) it can model the information that, values of some attributes are semantically closer than values of other attributes, and 2) it offers a mechanism to temper exaggerations in patient information. Although the algorithm requires precise training data, its utility for fuzzy training data is argued for. This is supported by the case study on infectious disease dataset, which indicates optimality of FNB over NB for the infectious disease domain. Further case studies on large datasets are required to establish utility of FNB.