Natural Language Processing Based Approach for Identification of Problems in Medical Image Management Using PACS.

We investigated problems concerning medical imaging management using PACS in medical settings through text analysis. We conducted a questionnaire survey in Hokkaido, Japan, where PACS related problems were described by radiological technologists. After analyzing the descriptions in response to the questionnaire, we detected four main problems in PACS using a word co-occurrence network; i) image management, ii) image fetching error, iii) DICOM violation and iv) name notation.

[An elderly patient with senile dementia showing acquired hemophilia A].

We report a case of an 82-year-old woman with senile dementia who was hospitalized at a specialist dementia hospital for 10 months. She had swelling of the left shoulder joint, subcutaneous hematoma in the left upper arm, and anemia was noted on blood examination. Her serum hemoglobin level was lower than normal at 4.6 g/dL, but there was no sign of gastrointestinal disease or gastrointestinal bleeding, and her stool specimens were negative for occult blood. Hematoma subsequently appeared on her chest and back. She had a low activation level of factor VIII (<1%), a high concentration of acquired inhibitors of factor VIII (18.5 BU/mL), and prolonged activated partial thromboplastin time (83.1 seconds). The possibility of drug-induced anemia or hematoma were ruled out. We diagnosed acquired hemophilia A (AHA), and suspected that this was the cause of her hematomas. We began treatment of her AHA with oral prednisolone and intravenous infusion of factor VIII. The bleeding improved, but she later died due to bacterial pneumonia. AHA is very rare, with a reported annual incidence of 0.1/100,000 in Japan. However, it is necessary to consider such a rare disease when we encounter bleeding in elderly patients.