An efficient information retrieval technique for e-health systems

In the health domain, the adoption of computer systems introduces better services, reduces human errors, and provides reliable services with nearly zero down time. In general, data in computer systems is stored in coded format; however, certain data, like user comments, cannot be coded. Hence, it is stored in the form of free text. Based on the results of the performed literature review, it was identified that the free text contains invaluable information; however, extracting such information is a challenging task due to the complexity of the stored data. In this paper, a Latent Semantic Indexing (LSI) algorithm is developed and applied on The Health Improvement Network (THIN). The algorithm utilizes the computational power provided by the multi-processor/multi-core system in performing the IR process. Further to that, the paper investigates the representation of the patient's data in the Term Document Matrix (TDM) to enhance the accuracy of the retrieved information.