Can AI Help Pediatricians? Diagnosing Kawasaki Disease Using DRSA.

The DRSA method (dominance-based rough set approach) was used to create decision-making rules based on the results of physical examination and additional laboratory tests in the differential diagnosis of Kawasaki disease (KD), infectious mononucleosis and S. pyogenes pharyngitis in children. The study was conducted retrospectively. The search was based on the ICD-10 (International Classification of Diseases) codes of final diagnosis. Demographic and laboratory data from one Polish hospital (Poznan) were collected. Traditional statistical methods and the DRSA method were applied in data analysis. The algorithm formed 45 decision rules recognizing KD. The rules with the highest sensitivity (number of false negatives equals zero) were based on the presence of conjunctivitis and CRP (C-reactive Protein) ≥ 40.1 mg/L, thrombocytosis and ESR (Erythrocyte Sedimentation Rate) ≥ 77 mm/h; fair general condition and fever ≥ 5 days and rash; fair general condition and fever ≥ 5 days and conjunctivitis; fever ≥ 5 days and rash and CRP ≥ 7.05 mg/L. The DRSA analysis may be helpful in diagnosing KD at an early stage of the disease. It can be used even with a small amount of clinical or laboratory data.

Structure-Activity Relationships of the Imidazolium Compounds as Antibacterials of Staphylococcus aureus and Pseudomonas aeruginosa.

This paper presents the results of structure-activity relationship (SAR) studies of 140 3,3'-(α,ω-dioxaalkan)bis(1-alkylimidazolium) chlorides. In the SAR analysis, the dominance-based rough set approach (DRSA) was used. For analyzed compounds, minimum inhibitory concentration (MIC) against strains of Staphylococcus aureus and Pseudomonas aeruginosa was determined. In order to perform the SAR analysis, a tabular information system was formed, in which tested compounds were described by means of condition attributes, characterizing the structure (substructure parameters and molecular descriptors) and their surface properties, and a decision attribute, classifying compounds with respect to values of MIC. DRSA allows to induce decision rules from data describing the compounds in terms of condition and decision attributes, and to rank condition attributes with respect to relevance using a Bayesian confirmation measure. Decision rules present the most important relationships between structure and surface properties of the compounds on one hand, and their antibacterial activity on the other hand. They also indicate directions of synthesizing more efficient antibacterial compounds. Moreover, the analysis showed differences in the application of various parameters for Gram-positive and Gram-negative strains, respectively.

Supporting contaminated sites management with Multiple Criteria Decision Analysis: Demonstration of a regulation-consistent approach.

This study proposes a set of key decision-making features of the contaminated site remediation process to assist in selecting the most appropriate decision support method(s). Using a case study consistent with the requirements of the U.S. regulation for contaminated sites management, this article shows that suitable Multiple Criteria Decision Analysis methods can be selected based on a dynamic and evolving problem structuring. The selected methods belong to the family of PROMETHEE methods and can provide ranking recommendations of the considered alternatives using variable structures of the criteria, evaluation of the alternatives and exploitation of the preference model. It was found that in order to support a quick and up-to-date application of powerful decision support techniques in the process of remediation of contaminated sites, decision analysts and stakeholders should interact and co-develop the process. This research also displays how such interactions can guarantee a transparent and traceable decision recommendation so that stakeholders can better understand why some alternatives perform comprehensively better than others when a multitude of inputs is used in the decision-making process.

Application of Dominance-Based Rough Set Approach for Optimization of Pellets Tableting Process.

Multiple-unit pellet systems (MUPS) offer many advantages over conventional solid dosage forms both for the manufacturers and patients. Coated pellets can be efficiently compressed into MUPS in classic tableting process and enable controlled release of active pharmaceutical ingredient (APIs). For patients MUPS are divisible without affecting drug release and convenient to swallow. However, maintaining API release profile during the compression process can be a challenge. The aim of this work was to explore and discover relationships between data describing: composition, properties, process parameters (condition attributes) and quality (decision attribute, expressed as similarity factor f2) of MUPS containing pellets with verapamil hydrochloride as API, by applying a dominance-based rough ret approach (DRSA) mathematical data mining technique. DRSA generated decision rules representing cause-effect relationships between condition attributes and decision attribute. Similar API release profiles from pellets before and after tableting can be ensured by proper polymer coating (Eudragit® NE, absence of ethyl cellulose), compression force higher than 6 kN, microcrystalline cellulose (Avicel® 102) as excipient and tablet hardness ≥42.4 N. DRSA can be useful for analysis of complex technological data. Decision rules with high values of confirmation measures can help technologist in optimal formulation development.

How to support the application of multiple criteria decision analysis? Let us start with a comprehensive taxonomy.

Decision making is a complex task that involves a multitude of perspectives, constraints, and variables. Multiple Criteria Decision Analysis (MCDA) is a process that has been used for several decades to support decision making. It includes a series of steps that systematically help Decision Maker(s) (DM(s)) and stakeholders in structuring a decision making problem, identifying their preferences, and building a decision recommendation consistent with those preferences. Over the last decades, many studies have demonstrated the conduct of the MCDA process and how to select an MCDA method. Until now, there has not been a review of these studies, nor a proposal of a unified and comprehensive high-level representation of the MCDA process characteristics (i.e., features), which is the goal of this paper. We introduce a review of the research that defines how to conduct the MCDA process, compares MCDA methods, and presents Decision Support Systems (DSSs) to recommend a relevant MCDA method or a subset of methods. We then synthesize this research into a taxonomy of characteristics of the MCDA process, grouped into three main phases, (i) problem formulation, (ii) construction of the decision recommendation, and (iii) qualitative features and technical support. Each of these phases includes a subset of the 10 characteristics that helps the analyst implementing the MCDA process, while also being aware of the implication of these choices at each step. By showing how decision making can be split into manageable and justifiable steps, we reduce the risk of overwhelming the analyst, as well as the DMs/stakeholders during the MCDA process. A questioning strategy is also proposed to demonstrate how to apply the taxonomy to map MCDA methods and select the most relevant one(s) using real case studies. Additionally, we show how the DSSs for MCDA method recommendation can be grouped into three main clusters. This proposal can enhance a traceable and categorizable development of such systems.

Machine-learned models using hematological inflammation markers in the prediction of short-term acute coronary syndrome outcomes.

BACKGROUND: Increased systemic and local inflammation play a vital role in the pathophysiology of acute coronary syndrome. This study aimed to assess the usefulness of selected machine learning methods and hematological markers of inflammation in predicting short-term outcomes of acute coronary syndrome (ACS). METHODS: We analyzed the predictive importance of laboratory and clinical features in 6769 hospitalizations of patients with ACS. Two binary classifications were considered: significant coronary lesion (SCL) or lack of SCL, and in-hospital death or survival. SCL was observed in 73% of patients. In-hospital mortality was observed in 1.4% of patients and it was higher in the case of patients with SCL. Ensembles of decision trees and decision rule models were trained to predict these classifications. RESULTS: The best performing model for in-hospital mortality was based on the dominance-based rough set approach and the full set of laboratory as well as clinical features. This model achieved 81 ± 2.4% sensitivity and 81.1 ± 0.5% specificity in the detection of in-hospital mortality. The models trained for SCL performed considerably worse. The best performing model for detecting SCL achieved 56.9 ± 0.2% sensitivity and 66.9 ± 0.2% specificity. Dominance rough set approach classifier operating on the full set of clinical and laboratory features identifies presence or absence of diabetes, systolic and diastolic blood pressure and prothrombin time as having the highest confirmation measures (best predictive value) in the detection of in-hospital mortality. When we used the limited set of variables, neutrophil count, age, systolic and diastolic pressure and heart rate (taken at admission) achieved the high feature importance scores (provided by the gradient boosted trees classifier) as well as the positive confirmation measures (provided by the dominance-based rough set approach classifier). CONCLUSIONS: Machine learned models can rely on the association between the elevated inflammatory markers and the short-term ACS outcomes to provide accurate predictions. Moreover, such models can help assess the usefulness of laboratory and clinical features in predicting the in-hospital mortality of ACS patients.

Optimization of pellets manufacturing process using rough set theory.

Pharmaceutical pellets are spherical agglomerates manufactured in extrusion/spheronization process. The composition of the pellets, the amount of active pharmaceutical ingredient (API) and the type of used excipients have an influence on the shape and quality of dosage form. A proper quality of the pellets can also be achieved by identifying the most important technological process parameters. In this paper, a knowledge discovery method, called dominance-based rough set approach (DRSA) has been applied to evaluate critical process parameters in pellets manufacturing. For this purpose, a set of condition attributes (amount of API; type and amount of excipient used; process parameters such as screw and rotation speed, time and temperature of spheronization) and a decision attribute (quality of the pellets defined by the aspect ratio) were used to set up an information system. The DRSA analysis allowed to induce decision rules containing information about process parameters which have a significant impact on the quality of manufactured pellets. Those rules can be used to optimize the process of pellets manufacturing.

Learning ensemble classifiers for diabetic retinopathy assessment.

Diabetic retinopathy is one of the most common comorbidities of diabetes. Unfortunately, the recommended annual screening of the eye fundus of diabetic patients is too resource-consuming. Therefore, it is necessary to develop tools that may help doctors to determine the risk of each patient to attain this condition, so that patients with a low risk may be screened less frequently and the use of resources can be improved. This paper explores the use of two kinds of ensemble classifiers learned from data: fuzzy random forest and dominance-based rough set balanced rule ensemble. These classifiers use a small set of attributes which represent main risk factors to determine whether a patient is in risk of developing diabetic retinopathy. The levels of specificity and sensitivity obtained in the presented study are over 80%. This study is thus a first successful step towards the construction of a personalized decision support system that could help physicians in daily clinical practice.

With a little help from a computer: discriminating between bacterial and viral meningitis based on dominance-based rough set approach analysis.

Differential Diagnosis of bacterial and viral meningitis remains an important clinical problem. A number of methods to assist in the diagnoses of meningitis have been developed, but none of them have been found to have high specificity with 100% sensitivity.We conducted a retrospective analysis of the medical records of 148 children hospitalized in St. Joseph Children's Hospital in Poznan. In this study, we applied for the first time the original methodology of dominance-based rough set approach (DRSA) to diagnostic patterns of meningitis data and represented them by decision rules useful in discriminating between bacterial and viral meningitis. The induction algorithm is called VC-DomLEM; it has been implemented as software package called jMAF (http://www.cs.put.poznan.pl/jblaszczynski/Site/jRS.html), based on java Rough Set (jRS) library.In the studied group, there were 148 patients (78 boys and 70 girls), and the mean age was 85 months. We analyzed 14 attributes, of which only 4 were used to generate the 6 rules, with C-reactive protein (CRP) being the most valuable.Factors associated with bacterial meningitis were: CRP level ≥86 mg/L, number of leukocytes in cerebrospinal fluid (CSF) ≥4481 µL, symptoms duration no longer than 2 days, or age less than 1 month. Factors associated with viral meningitis were CRP level not higher than 19 mg/L, or CRP level not higher than 84 mg/L in a patient older than 11 months with no more than 1100 µL leukocytes in CSF.We established the minimum set of attributes significant for classification of patients with meningitis. This is new set of rules, which, although intuitively anticipated by some clinicians, has not been formally demonstrated until now.

Robustness analysis of a green chemistry-based model for the classification of silver nanoparticles synthesis processes.

This paper proposes a robustness analysis based on Multiple Criteria Decision Aiding (MCDA). The ensuing model was used to assess the implementation of green chemistry principles in the synthesis of silver nanoparticles. Its recommendations were also compared to an earlier developed model for the same purpose to investigate concordance between the models and potential decision support synergies. A three-phase procedure was adopted to achieve the research objectives. Firstly, an ordinal ranking of the evaluation criteria used to characterize the implementation of green chemistry principles was identified through relative ranking analysis. Secondly, a structured selection process for an MCDA classification method was conducted, which ensued in the identification of Stochastic Multi-Criteria Acceptability Analysis (SMAA). Lastly, the agreement of the classifications by the two MCDA models and the resulting synergistic role of decision recommendations were studied. This comparison showed that the results of the two models agree between 76% and 93% of the simulation set-ups and it confirmed that different MCDA models provide a more inclusive and transparent set of recommendations. This integrative research confirmed the beneficial complementary use of MCDA methods to aid responsible development of nanosynthesis, by accounting for multiple objectives and helping communication of complex information in a comprehensive and traceable format, suitable for stakeholders and/or decision-makers with diverse backgrounds.

Prediction of antifungal activity of gemini imidazolium compounds.

The progress of antimicrobial therapy contributes to the development of strains of fungi resistant to antimicrobial drugs. Since cationic surfactants have been described as good antifungals, we present a SAR study of a novel homologous series of 140 bis-quaternary imidazolium chlorides and analyze them with respect to their biological activity against Candida albicans as one of the major opportunistic pathogens causing a wide spectrum of diseases in human beings. We characterize a set of features of these compounds, concerning their structure, molecular descriptors, and surface active properties. SAR study was conducted with the help of the Dominance-Based Rough Set Approach (DRSA), which involves identification of relevant features and relevant combinations of features being in strong relationship with a high antifungal activity of the compounds. The SAR study shows, moreover, that the antifungal activity is dependent on the type of substituents and their position at the chloride moiety, as well as on the surface active properties of the compounds. We also show that molecular descriptors MlogP, HOMO-LUMO gap, total structure connectivity index, and Wiener index may be useful in prediction of antifungal activity of new chemical compounds.

Antimicrobial activity and SAR study of new gemini imidazolium-based chlorides.

A series of 70 new 3,3'(α,ω-dioxaalkyl)bis(1-alkylimidazolium) chlorides were synthesized. They were characterized with respect to surface active properties and antimicrobial activity against the following pathogens: Staphylococcus aureus, Enterococcus faecalis, Acinetobacter baumannii, Escherichia coli, Klebsiella pneumoniae, Enterobacter cloacae, Pseudomonas aeruginosa, Candida krusei, and Candida albicans. In this article, besides description of the synthesis, we characterize a set of features of these compounds, concerning their structure (described by the length of the dioxaalkan spacer and the length of the alkyl substituent in the aromatic ring) and surface active properties (critical micelle concentration, value of surface tension at critical micelle concentration, value of surface excess, molecular area of a single particle, and free energy of adsorption of molecule). Then, we present a SAR study for Staphylococcus aureus, as one of the most widespread pathogenic strains, conducted with the help of the Dominance-based Rough Set Approach (DRSA), that involves identification of relevant features and relevant combinations of features being in strong relationship with a high antimicrobial activity of the compounds. The SAR study shows, moreover, that the antimicrobial activity is dependent on the type of substituents and their position at the chloride moiety, as well as on the surface active properties of the compounds.

MET system: a new approach to m-health in emergency triage.

The MET (Mobile Emergency Triage) system is an m-health application that supports emergency triage of various types of acute pain at the point of care. The system is designed for use in the Emergency Department (ED) of a hospital and to aid physicians in disposition decisions. Given patient's condition, MET recommends a triage by consulting decision rules stored in the system's knowledge base. The rules have been created using a data mining method (based on rough set methodology) applied to data collected during a retrospective chart study and verified by the clinicians. MET is designed following the extended client-server architecture, suited for weak-connectivity conditions, where stable connection between clients and a server cannot be provided. The MET server interacts with the hospital's patient information system in order to retrieve information about patients admitted to the ED. It also stores current patients' demographic and clinical data to be exchanged with mobile clients. The MET mobile client, running on a Personal Digital Assistant (PDA), is used for collecting clinical data and supporting triage decisions. The support function runs solely on the client side, thus it can be invoked anytime and anywhere, even if there is no communication link with the server (e.g., there is no wireless network available in the ED). Due to implementation on PDAs and working in weak-connectivity conditions, the MET system is very well suited for use in the ED and fits seamlessly into the regular clinical workflow without introducing any hindrances or disruptions that are often reported when using stationary (i.e., working on desktop computers) clinical systems. The system facilitates patient-centered service and timely, high quality patient management. It provides recommendations using a limited amount of clinical data, normally available at the point of care. Furthermore, it provides a possibility for the structured evaluation of this data by an attending physician.