[Development of the method for the similarity detection of descriptions in radiology reports using structuring technology].

Radiologists often spend much time for re-reading some of the past free-text radiology reports and determining interval changes in the physical findings when creating a report for long term cases. The aim of this study was to propose the method to detect semantic similar descriptions in the free-text reports using the structuring method based on text-mining technology. In a previous study, we had developed the structuring method that can semantically analyze the free-text reports and convert them into the description unit consisting of five items: finding/diagnosis, modifier, region, regional modifier, and confidence. Our developed prototype system extracted similar descriptions from the free-text reports by calculating the similarity index between description units. We confirmed similar descriptions extracted by the system applied to free-text reports of cases which had more than one chest CT examination written in actual clinical situation. As a result, it became available to identify candidates of similar descriptions from free-text reports. In some cases regarding practical use, the similar descriptions could not be identified in the sentences which used paraphrasing or where the findings had status changes. A solution of identifying similarity in these cases was necessary to improve the method. With the presented method here, it is expected that interval changes in the findings can be visualized and applied it to support diagnosis.

Structuring of free-text diagnostic report.

PURPOSE: It is useful to convert free-text diagnostic reports into structured diagnostic reports by semantic analysis for the secondary investigation of their contents. In this study, we propose a system in which description units are automatically extracted to create structured text reports and we evaluated its usefulness. METHODS: We defined the rules to create description units and developed the system that can automatically extract these description units from free-text diagnostic reports. We applied this system to reports of cerebral perfusion scintigrams and obtained 5 dictionaries of description units, increasing the number of scintigrams from 100 to 500 in increments of 100. Each dictionary was used to analyze another 100 scintigrams. The results obtained using each dictionary were compared with the results of physicians' interpretation. RESULTS: The recall rate of this system to the physicians' interpretation increased when correlated with the number of scintigrams but with 300 cases was almost saturated at 85%. CONCLUSION: We propose a semantic analysis system and show its usefulness in the semantic evaluation of the reports of cerebral perfusion scintigrams.

Development of built-in type and noninvasive sensor systems for smart artificial heart.

It is very important to grasp the artificial heart condition and the physiologic conditions for the implantable artificial heart. In our laboratory, a smart artificial heart (SAH) has been proposed and developed. An SAH is an artificial heart with a noninvasive sensor; it is a sensorized and intelligent artificial heart for safe and effective treatment. In this study, the following sensor systems for SAH are described: noninvasive blood temperature sensor system, noninvasive blood pressure sensor system, and noninvasive small blood flow sensor system. These noninvasive sensor systems are integrated and included around the artificial heart to evaluate these sensor systems for SAH by the mockup experiments and the animal experiments. The blood temperature could be measured stably by the temperature sensor system. Aortic pressure was estimated, and sucking condition was detected by the pressure sensor system. The blood flow was measured by the flow meter system within 10% error. As a result of these experiments, we confirmed the effectiveness of the sensor systems for SAH.