著者
甲斐 拓海 清原 克哲 中森 友啓 木村 勇太 木村 加奈子 嘉戸 直樹 鈴木 俊明
出版者
関西理学療法学会
雑誌
関西理学療法 (ISSN:13469606)
巻号頁・発行日
vol.22, pp.95-100, 2022 (Released:2022-12-23)
参考文献数
6

We report the case of a man in his 50s with decreased endurance in gait after brain tumor resection. His main complaints were that he felt pain on the right side of his waist and that he could not walk for long periods. We observed that during the left loading-response phase, he could not adduct his left hip joint, nor could he move his pelvis to the left. Therefore, during the initial swing phase, he put his right plantar foot down quickly and his trunk tilted forward. This resulted in flexion of the thoracolumbar transition during the right loading-response phase and extension of the thoracolumbar transition during the right mid-stance phase. The patient repeated this gait pattern, and after about 3,000 steps, he developed pain in the right lumbar region. We considered the main problem to be weakness of the left hip adductor muscle, and treated the patient with physical therapy. As a result, left hip adduction became possible during the left loading-response phase, and the pain in the patient’s right lumbar region improved, resulting in improved endurance of walking. In this case, physical therapy with a focus on the left adductor magnus muscle was effective at improving the pain in the patient’s right lumbar back and the functioning of the hip adductor muscles required during the left loading-response phase.
著者
木村 勇太 伊藤 毅志
出版者
一般社団法人 人工知能学会
雑誌
人工知能学会全国大会論文集 第32回 (2018)
巻号頁・発行日
pp.1H1OS13a03, 2018 (Released:2018-07-30)

従来の人狼AIでは、役職の推定は製作者によりヒューリスティックに決められてきた.本研究では既存エージェントの推定部分に機械学習の一つであるSVR(Support Vector Regression)を導入しその有効性を検証した.その結果,中盤以降で従来の手法を上回る推定率を実現することが出来た.また,序盤では従来の手法,中盤以降では提案手法を用いるエージェントを構築することで,従来のエージェントを有意に上回る勝率を実現することが出来た.