الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 190 |  |  |
| | | from | 190 | | |
Abstract
Healthcare is a big topic which presents even bigger challenges for the leaders. Past few years; healthcare experts used decision support systems (DSS) to find solutions to multiple complex issues which impact the ability to successfully implement cost-effective programs, maintain efficient operations and services. Traditional DSSs was enhanced based on artificial intelligence and expert system techniques. Also, traditional DSSs based on data mining that is focused on the analysis of large data sets. Process mining, an interdisciplinary research field between data mining and business process management aims to discover, monitor and improve real-world processes by extracting knowledge from activity (event) logs.
The main objective of this research is to develop, implement and evaluate a decision support system based on process-mining to enhance the performance of healthcare business processes. The thesis has four main contributions that employed as modules of the recommender the system framework:
First contribution; the thesis implements process mining approach in analyzing the patient’s journey for the patient, as he registered to the hospital to the end and leaving the hospital. We have used a dataset from Egyptian hospital based on the event logs of the cardiac patients. The proposed methodology has employed four discovery algorithms to mine a structured process model from unstructured care flows, an evaluation of the quality metrics among the output models from the discovery miners was conducted to select the best process model that describe care
II
flows of all groups of patients, the evaluation was based on fitness, precision, generalization, and simplicity metrics. To evaluate the simplicity metric of extracted models; we suggested a method to quantify the simplicity metric and decide if the process model is simple or not. The method with two steps: checking the sound- ness of every model resulted from the miner algorithms, then secondly; measuring the metrics of control-flow complexity. The matching rate between the discovered process model and the standard one was a 95%. We derived the insights from the care flows and the event log by utilizing organizational and performance analysis.
Second contribution; Decision mining approach within process mining to analyze how decisions impact treatment outcomes. We identify areas where the decision-makers are able to improve to enhance the care-flows, to achieve the satisfaction for the patient.
Third contribution; Convolutional Neural Network-based next-activity prediction of an event in a business process using process mining and data analytics. Initially, each trace of the historical events is converted into a set of prefix traces, which are then mapped into two-dimensional images. These are called “spatial data”. The process data engineering approach is used to convert the temporal data for an event into spatial data, treating them as an image. These are then trained with the CNN to create a model that can predict the next activity in the running processes of a business process.
Fourth contribution; Design a process-aware recommender system based on CNN predictive model (second contribution) and the analysis of healthcare processes
III
(first contribution), the process-aware recommender system can advise the knowledge worker with the best next activities. At the stage of process mining analysis, there are discovering of the process models and KPI (discovering bottlenecks) and compare it with the output of CNN model to avoid the risky activities that causes bottlenecks.