Medical conferences, in which doctors and other clinical practitioners consider treatment policies for patients, are instrumental in providing advanced and safe medical care at large medical institutions. However, they face the following challenges. 1) Because existing medical information systems are ineffective in extracting information alluding to patients discussed during medical conferences and in integrated management of patient information related to diverse aspects, extraction tasks burden the personnel involved. These personnel must gather the clinical course, examination results, and treatment details in advance, as well as other materials necessary to understand and assess a medical condition. 2) Because these data are distributed and stored in various medical information systems, considerable time and effort are spent in manipulating multiple systems simultaneously. This necessity also hinders the grasp of information at a medical conference. In this study, we constructed a conference support information system that retrieves information from multiple medical information systems in a cross-sectional manner. The system extracts relevant information about the target patient of a medical conference, and provides an integrated user experience (UX). We also attempted to resolve the afore-mentioned problems. Testing and evaluating our conference support information system showed that medical conference preparation time was shortened. Our system also contributed to improving the conference itself, through the effectiveness of the system’s inter-system information gathering capabilities and integrated UX at the medical conference. These results suggest the possibility that public application programming interface in medical information systems may generate new support services.

We previously developed a method to bend thin catheter using acoustic radiation force. We succeeded in bending a thin catheter in the direction perpendicular to ultrasound propagation, which is the most challenging condition, by forming two focal points with opposite phases. However, since the acoustic energy was dispersed, it was difficult to obtain sufficient displacement. Therefore, we attempted to bend the thin catheter in the direction perpendicular to ultrasound propagation using tempo-spatial division emission of a single focal point. In this study, we used a two-dimensional array transducer with 256 elements, central frequency of 1MHz, and radius of curvature of 120mm. The thin catheter made of perfluoroalkoxy had an outer diameter of 0.2mm and an inner diameter of 0.05mm. The tip of the catheter was bent by transition at the focal point, where the temporal interval ranged from 0.01 to 5 s and the spatial interval ranged from 0.05 to 1.5mm, with totally 10 patterns of the focal point. As a result, the catheter was bent with a maximum displacement of 0.7mm, which was almost three times greater than the outer diameter of the catheter.

Lung cancer is one of the most important diseases to overcome, and chest radiography and low-dose computed tomography (CT) have been used for lung cancer screening. In chest radiographs, nodules overlying the ribs and mediastinum are occasionally difficult to detect. Therefore, CT is used as the gold standard modality to diagnose lung cancer, despite the high radiation exposure it entails. Recently, tomosynthesis has been used to detect pulmonary nodules, but mechanical scanning and a long data acquisition time (approximately 10 seconds) are required to obtain the image. Under this circumstance, tomosynthesis using a small number of projections has been anticipated to allow omission of the mechanical system and shortening of the exposure time. A chest phantom with artificial pulmonary nodules of various sizes was imaged in rotating 360-degree projections. Tomosynthesis images were reconstructed using back projection (BP) from ±32-degrees projection data, and the projection numbers were set at 4, 6, 8, 16, 32 and 64. The images were evaluated visually and quantitatively by measuring the contrast noise ratio (CNR) and artifact spread function (ASF). Tomosynthesis images reconstructed from 4 projection images allowed visualization of an artificial nodule 10mm in diameter without overlapping the ribs or the mediastinum. The CNR of a 4-projection image normalized to that of a 64-projection image was approximately 0.34, and the ASF obtained indicated that fewer number of projections was associated with greater z-axis resolution. Thus, 4-projection tomosynthesis allows visualization of pulmonary nodules 10mm or larger in diameter, and image quality appears to be useful.