대한한방내과학회지 제4 0 권 2 호( 2 0 1 9 년 5 월) J .I n t .K o r e a nMe d .2 0 1 9 ; 4 0 ( 2 ) : 2 7 0 -2 7 8 h t t p : / / d x . d o i . o r g / 1 0 . 2 2 2 4 6 / j i k m. 2 0 1 9 . 4 0 . 2 . 2 7 0 2 7 0 ․투고일:2 0 1 9 . 0 3 . 2 8 ,심사일:2 0 1 9 . 0 5 . 3 1 ,게재확정일:2 0 1 9 . 0 5 . 3 1 ․교신저자:권승원 서울시 동대문구 회기동 1 번지 경희대학교 한방병원 순환신경내과학교실 TEL:0 2 -9 5 8 -9 2 7 5 FAX:0 2 -9 5 8 -9 1 3 2 E-ma i l :k k o k k o t t u n g @h a n ma i l . n e t

In this study, five cases of post-stroke vision disorders are presented. The patients had no distinctive change in vision disorders before Korean medical treatment, which included herbal medicine, acupuncture, and electroacupuncture. After 11 to 28 days of Korean medical treatment, the patients experienced the first improvement of vision disorders after stroke. Also, patients who received Korean medical treatment after 30 days from onset, the duration in which the most improvement of post-stroke vision disorders occurs, progressed their vision disorders by the start of Korean medical treatment. Korean medical treatment might be effective in treating vision disorders after stroke.

The purpose of this case report is to evaluate the effect of electroacupuncture on Pal-sa (EX-UE9) for unilateral motor disturbance of the hand after cerebral infarction.Methods: One patient with unilateral motor disturbance of the hand following cerebral infarction (right basal ganglion and corona radiata) was treated with acupuncture, herbal medication, and electroacupuncture on Pal-sa (EX-UE9) once daily from June 30, 2016 to July 4, 2016. We evaluated improvement using the Box and Block Test (BTT) and 10-seconds Test, including the Finger Individual Movement Test (FIMT), the Hand Pronation and Supination Test (HPST), and the Finger Tapping Test (FTT).Results: After five days of treatment, increase of FIMT and FTT scores was observed after electroacupuncture on Pal-sa (EX-UE9). However, no increase was observed in BBT or HPST scores. Conclusions:This study suggests that electroacupuncture on Pal-sa (EX-UE9) can help treat motor disturbance of the hand after cerebral infarction.

Deep learning has recently achieved remarkable results in the field of medical imaging. However, as a deep learning network becomes deeper to improve its performance, it becomes more difficult to interpret the processes within. This can especially be a critical problem in medical fields where diagnostic decisions are directly related to a patient's survival. In order to solve this, explainable artificial intelligence techniques are being widely studied, and an attention mechanism was developed as part of this approach. In this paper, attention techniques are divided into two types: post hoc attention, which aims to analyze a network that has already been trained, and trainable attention, which further improves network performance. Detailed comparisons of each method, examples of applications in medical imaging, and future perspectives will be covered.

No abstract