Academic journal article Issues in Informing Science & Information Technology

Meta-Analysis of Clinical Cardiovascular Data towards Evidential Reasoning for Cardiovascular Life Cycle Management

Academic journal article Issues in Informing Science & Information Technology

Meta-Analysis of Clinical Cardiovascular Data towards Evidential Reasoning for Cardiovascular Life Cycle Management

Article excerpt

Introduction

Computers and information technology (IT) have been widely used in medical research, particularly for providing several types of service. A remarkable advantage of IT is that long-term investigation on patient-care, diagnosis and treatment of diseases can be done efficiently through largescale data analysis and data mining of personalized historical data sets. Thus for instance, a clinical decision support system provides diagnosis, treatment, follow-up on a number of issues, including the option to create information anchor desks for a wide variety of practitioners, such as, general practitioners, pharmacists, nutritionists, and the like (Kaplan, 2001, Trivedi, 2004, Payne, 2000, Sim, 2001, Purcell, 2005). Such a system will not only benefits patient care, but also provide physicians a platform to study the medical issues at a broader level. However, few systems have been successfully developed and tailored to handle a broad spectrum of diseases (Economou, 2001, Cornelia, 2003, Bobb, 2004). This is mainly due to the fact that designing and providing a reliable and knowledge-based medical informatics environment, like medical disease decision support system, is still very challenging in terms of collecting large amount of medical data, large scale data mining and management, information extraction including text and graphics, Human Computer Interface (HCI), development of statistical tools, and decision making tools (Kuperman, 2003, Clercq, 2004). In particular, when such tools are put together over a hardware-software architecture, it requires a competitive multi-disciplinary team with experts drawn from each field to manage, massage and exploit data sets effectively and efficiently.

From the medical perspective, cardiovascular disease is one of the serious and life-threatening diseases in the developed world. Causes of cardiovascular disease are many, including genetics, food habits, hardening of the arteries, high blood pressure and others. In recent years, a number of drugs have been discovered or synthesized, with capabilities to soften the arteries and control cholesterol and blood pressure, and some of them are widely used in medical treatment. However, it has been observed that the effectiveness and outcomes of the treatment are surprisingly inconsistent, even among "similar patients". More and more literature show that such diverse reasons result in many problems in understanding the causes of the disease, patient treatment, drug design and disease prevention in individual level (Reddy, 1998, Fuster, 2005, Lewis, 2002). For example, Lipitor (Black, 1998)--a cholesterol-reducing drug/satin has profound impact on certain patients, but limited or no impact on other patients. Further, certain post-operative patients, who perform routine exercise show significant improvement in their life-style, but others who maintain such or similar kind of exercise regimen, seem to have life style problems. Further critical research on data available openly on landmark studies in cardiology reveal that medical treatment needs to be greatly tailored to specific individual patients. However, these pathologies and mechanisms are still under discovery, which seriously affects the accuracy of decision-making on treatment; occasionally such trial and error processes may also lead to medical accidents. Theoretically, such a study can be prosecuted through a careful and complex investigation of patients using computational and statistical tools. However, in practice, it has many technical difficulties, two of them being: i) the need to focus any study to a few measurable parameters and ii) ipso facto the need to manage inconsistent medical information and data from several clinical studies. The former problem requires meta-analyses procedures to be developed as envisaged in this proposal, and the latter needs metadata converters (called form fillers here) so that medical information data from different landmark studies may be combined in a manner that meets the needs of cardiovascular disease management purposes. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.