Extracting Information from medical reports : European Hospital-Gaza Strip (case study)‎

العناوين الأخرى

استخراج المعلومات من التقارير الطبية : المستشفى الأوروبي-قطاع غزة (دراسة حالة)‎

مقدم أطروحة جامعية

al-Hajj, Maali Abd al-Razzaq Muhammad

مشرف أطروحة جامعية

al-Hulays, Ala Mustafa

الجامعة

الجامعة الإسلامية

الكلية

كلية تكنولوجيا المعلومات

القسم الأكاديمي

تكنولوجيا المعلومات

دولة الجامعة

فلسطين (قطاع غزة)

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2017

الملخص الإنجليزي

Developing a computer system to process and utilize text is significant kinds of research in text mining field.

But the challenge in how to built this system optimally, especially that these systems process text written in free language.

Textual documents are huge in all fields including a medical field which is the focus of our study in this research.

Recent years have observed the ability to gather a massive amount of data in a large number of domains.

As the data is collected in unprecedented rate, the analysis, rather than the storage of this data becomes a challenge.

According to the IDC estimation 90% of data is unstructured data which is a fastest growing data whereas the remaining is the structured data.

The rapid growth of medical records motivates hospitals and health organizations to use this technology to reduce time and effort, extract important information, and to speed up the process of analyzing and linking information to discover and predict relations between contents as: diseases, medication, medical history and medical recommendations for patients.

Medical reports are the beating heart of the hospital.

They are the most important work space in medical organizations, because of their importance as a bank of information provides support to fields of medicine and management.

In the same context, it became necessary to extract the required entities from the large data in order to shorten time and efforts of access the information in which is known “Information Extraction” technology.

Information Extraction (IE) is the process of finding instances of predefined entities type within the text, this predefine entity are depending on the field of the text and the goal of the system.

In this thesis we described the design and results of the experimental model to extract information (Named Entities) from Arabic textual reports in the medical field.

In another way we describe a model to transform unstructured Arabic medical text in reports to structured information.

We collected the records from Gaza hospitals which are written in Arabic and English Languages.

Then, we investigate some data mining process, association rules to generate useful rules from structured information.

That because, this model can be used to help medical staff to efficiently detect hidden relations between medical information, and making decisions that will improve the medical service for patients.

To evaluate our work we used two approaches objective and subjective.

As an objective evaluation we used support and confident metrics for association rules.

For subjective evaluation, we used a questionnaire to evaluate the generated rules by medical staff.

87% of the experts found that the generated rules are useful.

التخصصات الرئيسية

تكنولوجيا المعلومات وعلم الحاسوب

عدد الصفحات

80

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-905690