Graphic Visualization of the Co-Occurrence Analysis Network of Lung Cancer In-Patient Nursing Record

Abstract

Although the nursing record provides a complete account of a patient's information, it is not fully utilized. The relevant information including laboratory results, remarks made by doctors and nurses is not taken into consideration. Knowledge concerning the condition and treatment of patients has been determined in a twofold manner: a text mining technique has identified relations between feature vocabularies seen in past lung cancer in-patient records accumulated on University of Miyazaki Hospital electronic medical record, and an extraction has been attempted to solve the above-mentioned problem in the present study. The result was an analysis of a qualitative lung cancer in- patients' nursing record that used the text mining technique, and the initial goal was achieved: a visual record of this information. In addition, this enabled the discovery of vocabularies relating to the proper methods of treatment, resulting in a concise summary of the vocabularies extracted from the content of the lung cancer in-patients' nursing record. Important vocabularies characterizing each nursing record were also revealed.