Osama Mohammed

Building a diseases symptoms ontology for medical diagnosis: An integrative approach

Medical ontologies are valuable and effective methods of representing medical knowledge. In this direction, they are much stronger than biomedical vocabularies. In the process of medical diagnosis, each disease has several symptoms associated with it. There are currently no ontologies that relate diseases and symptoms and only attempts at their infancy along with some simple proposed models. However, well established ontologies for diseases and for symptoms were already developed in isolation. In this article, we are proposing an alignment algorithm to align the diseases ontology (DOID) with the symptoms ontology (SYMP) creating a core diseases symptoms ontology (DSO) that can scale to any number of diseases and symptoms The core DSO links a few diseases to their symptoms.

Evaluation of Stream Mining Classifiers for Real-Time Clinical Decision Support System: A Case Study of Blood Glucose Prediction in Diabetes Therapy

Earlier on, a conceptual design on the real-time clinical decision support system (rt-CDSS) with data stream mining was proposed and published. The new system is introduced that can analyze medical data streams and can make real-time prediction. This system is based on a stream mining algorithm called VFDT. The VFDT is extended with the capability of using pointers to allow the decision tree to remember the mapping relationship between leaf nodes and the history records. In this paper, which is a sequel to the rt-CDSS design, several popular machine learning algorithms are investigated for their suitability to be a candidate in the implementation of classifier at the rt-CDSS. A classifier essentially needs to accurately map the events inputted to the system into one of the several predefined classes of assessments, such that the rt-CDSS can follow up with the prescribed remedies being recommended to the clinicians. For a real-time system like rt-CDSS, the major technological challenges lie in the capability of the classifier to process, analyze and classify the dynamic input data, quickly and upmost reliably. An experimental comparison is conducted. This paper contributes to the insight of choosing and embedding a stream mining classifier into rt-CDSS with a case study of diabetes therapy.

Developing a Semantic Web Model for Medical Differential Diagnosis Recommendation

In this paper we describe a novel model for differential diagnosis designed to make recommendations by utilizing semantic web technologies. The model is a response to a number of requirements, ranging from incorporating essential clinical diagnostic semantics to the integration of data mining for the process of identifying candidate diseases that best explain a set of clinical features. We introduce two major components, which we find essential to the construction of an integral differential diagnosis recommendation model: the evidence-based recommender component and the proximity-based recommender component. Both approaches are driven by disease diagnosis ontologies designed specifically to enable the process of generating diagnostic recommendations. These ontologies are the disease symptom ontology and the patient ontology. The evidence-based diagnosis process develops dynamic rules based on standardized clinical pathways. The proximity–based component employs data mining to provide clinicians with diagnosis predictions, as well as generates new diagnosis rules from provided training datasets. This article describes the integration between these two components along with the developed diagnosis ontologies to form a novel medical differential diagnosis recommendation model. This article also provides test cases from the implementation of the overall model, which shows quite promising diagnostic recommendation results.

Mining twitterspace for information: Classifying sentiments programmatically using Java

People increasingly use Twitter to share advice, opinions, news, moods, concerns, facts, rumors, and everything else imaginable. Much of that data is public and available for mining. However, classifying automatically the sentiment of the Twitter messages into either positive or negative with respect to a query term represents a new research challenge. Variety of approaches that use natural language and statistical techniques failed to report high accuracy of tweets classification due to the nature of these tweets containing large number of abbreviations, emoticons and ill structured grammar. In this article we are presenting a programming approach that uses the Weka data mining APIs to classify tweets. Using this programming approach we can experiment on how to train the classifiers and determine which one is more effective than the others. In our experiments, the K* classifier is found to report a high degree of accuracy in tweets classification.

A time series pre-processing methodology with statistical and spectral analysis for classifying non-stationary stochastic biosignals

Biosignal classification is an important non-invasive diagnosis tool in biomedical application, e.g. electrocardiogram (ECG), electroencephalogram (EEG), and electromyogram (EMG) that helps medical experts to automatically classify whether a sample of biosignal under test/monitor belongs to the normal type or otherwise. Most biosignals are stochastic and non-stationary in nature, that means their values are time dependent and their statistics vary over different points of time. However, most classification algorithms in data mining are designed to work with data that possess multiple attributes to capture the non-linear relationships between the values of the attributes to the predicted target class. Therefore, it has been a crucial research topic for transforming univariate time series to multivariate dataset to fit into classification algorithms. For this, we propose a pre-processing methodology called statistical feature extraction (SFX). Using the SFX we can faithfully remodel statistical characteristics of the time series via a sequence of piecewise transform functions. The new methodology is tested through simulation experiments over three representative types of biosignals, namely EEG, ECG and EMG. The experiments yield encouraging results supporting the fact that SFX indeed produces better performance in biosignal classification than traditional analysis techniques like Wavelets and LPC-CC.

RULE GENERATION AND EVALUATION BY DATA MINING ENSEMBLES FOR CLINICAL DECISION SUPPORT

Clinical decision support systems (CDSS) often base on rules that are inferred from collected patients’ histories, together with expert judgements and consented medical guidelines. This type of advisor system is known as rulebased reasoning system or expert system which classifies a given test instance into a particular outcome from the learned rules. The test instance carries multiple attributes which are usually the values of diagnostic tests. In this paper, we propose a classifier ensemble-based method for supporting disease diagnosis. The ensemble data mining learning methods are applied for rule generation, and a multi-criteria evaluation approach is used for selecting reliable rules over the results of the ensemble methods. The efficacy of the proposed methodology is illustrated via an example of a thyroid disease classification.