Kento Sabashi

Dr. Kento Sabashi


Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
Verified
Journal Article
Open access
PDF file
KIKoji Iwasaki
Koji Iwasaki
KSKento Sabashi
Kento Sabashi
HKHidenori Koyano
Hidenori Koyano
YKYuji Kodama
Yuji Kodama
SSShigeyuki Sakurai
Shigeyuki Sakurai
NINorimasa Iwasaki
Norimasa Iwasaki
Background: Osteotomy around the knee (OAK) is a joint-preserving surgery for knee osteoarthritis, yet some patients experience suboptimal outcomes. Preoperative identification of high-risk patients remains challenging. This study aimed to develop a machine learning model to predict clinical outcomes after OAK using preoperative gait acceleration data from inertial measurement units (IMUs). Methods: This multicenter prospective study enrolled patients undergoing OAK. Preoperative gait was recorded using synchronized IMUs placed on the lumbar spine and tibia. Lumbar and tibial signals were used for gait-cycle segmentation, while wavelet-based time–frequency features were extracted from tibial acceleration only. Outcomes were defined by achievement of the minimal clinically important difference in ≥3 KOOS subscales at 2-year follow-up (Good vs. Poor). Continuous wavelet transform features (5–20 Hz) were summarized as mean and standard deviation across six stance subphases. A Random Undersampling Boost classifier was trained and evaluated using nested leave-one-subject-out cross-validation. A sensitivity analysis using logistic regression confirmed that the IMU-based prediction score was independently associated with outcome after adjustment for baseline KOOS (p = 0.047). Results: Of 67 enrolled patients, 37 were classified as Good and 30 as Poor outcome. For machine learning analysis, 1173 tibial acceleration gait-cycle waveforms were usable. The model achieved an AUC of 0.744 (95% CI, 0.610–0.860) using a median of 15 features (range, 5–25) with sensitivity of 0.69 and specificity of 0.72. The most informative predictors were the mean magnitude in the 5–8 Hz band during loading response (0–17%) and variability in the 5–8 Hz band during late stance (67–83%). No significant differences in baseline demographics or radiographic parameters were found between outcome groups. Conclusions: Preoperative IMU-derived gait acceleration features showed moderate-to-good discrimination between outcome groups and may support preoperative risk stratification and individualized perioperative management.

15 April 2026 in Biomechanics
Verified
Journal Article
Open access
PDF file
KSKento Sabashi
Kento Sabashi
RURyo Ueno
Ryo Ueno
Background/Objectives: The Gait Variable Score (GVS) and Gait Abnormality Score (GAS) have been proposed as methods for quantitatively evaluating deviations from normal gait patterns. This study aimed to investigate whether the GVS or GAS is more useful for evaluating gait in stroke survivors. Methods: We used open-access motion capture datasets from 43 stroke survivors and 82 healthy individuals. Nine kinematics and seven muscle activities were extracted. The GVS was calculated as the root mean square difference between the pathological and healthy gait patterns. The modified GAS (mGAS) newly defined in this study was calculated as the mean value of the absolute differences between the pathological and healthy gait patterns divided by the standard deviation of healthy gait patterns. The amplitudes of kinematics and muscle activities were calculated. Results: Both the GVS and mGAS were significantly higher in stroke survivors than in healthy individuals. A significant strong correlation for 16 variables (nine kinematics and seven muscle activities) was observed between the GVS and amplitude (r = 0.921), but no significant correlation was found between the mGAS and amplitude (r = 0.167). Conclusions: As the mGAS is not affected by the amplitude of kinematics and muscle activities, it allows for a comprehensive comparison of abnormalities in both kinematics and muscle activities. The mGAS may be more useful than the GVS for evaluating gait abnormalities in stroke survivors.

14 September 2026 in Bioengineering
Verified
Journal Article
Open access
PDF file
KSKento Sabashi
Kento Sabashi
RURyo Ueno
Ryo Ueno
MSMina Samukawa
Mina Samukawa
HTHarukazu Tohyama
Harukazu Tohyama
This study aimed to compare the accuracy of video-based kinematic measurements using artificial intelligence (AI) between patients with hip osteoarthritis (OA) and healthy participants, and to compare the gait abnormality estimated from video data with that calculated using optical three-dimensional motion capture (MOCAP) data. This study used open-access datasets of MOCAP and video data including 20 patients with hip OA and 20 healthy participants. Video data were processed using the MYoACT application to extract marker data. The mean absolute error (MAE) of joint angles between MOCAP and MYoACT data was calculated. In patients with hip OA, the modified Gait Abnormality Score (mGAS) for each joint angle was calculated by averaging the absolute differences between each patient’s value and the mean value of healthy participants, divided by the standard deviation of healthy participants, across the entire gait cycle. No significant difference in the MAE was observed between groups. In patients with hip OA, the mGAS showed no significant difference between MYoACT and MOCAP data. These findings suggest that AI-driven video-based kinematic measurements show no clear difference in the accuracy between patients with hip OA and healthy participants, and demonstrate the potential to identify gait abnormalities.

5 August 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
KSKento Sabashi
Kento Sabashi
RURyo Ueno
Ryo Ueno
MSMina Samukawa
Mina Samukawa
HTHarukazu Tohyama
Harukazu Tohyama
This study investigated whether the accuracy of video-based kinematic measurements using artificial intelligence (AI) in patients with hip osteoarthritis (OA) is comparable to that in healthy participants, and whether the gait abnormality estimated from video data is comparable to that calculated using optical three-dimensional motion capture (MOCAP) data. This study used open-access datasets of MOCAP and video data including 20 patients with hip OA and 20 healthy participants. Video data were processed using the MYoACT application to extract marker data. The mean absolute error (MAE) of joint angles between MOCAP and MYoACT data was calculated. In patients with hip OA, the modified Gait Abnormality Score (mGAS) for each joint angle was calculated by averaging the absolute differences between each patient’s value and the mean value of healthy participants, divided by the standard deviation of healthy participants, across the entire gait cycle. No significant difference in the MAE was observed between groups. In patients with hip OA, the mGAS showed no significant difference between MYoACT and MOCAP data. These findings suggest that AI-driven video-based kinematic measurements provide comparable accuracy between patients with hip OA and healthy participants, and can identify gait abnormalities at a level comparable to the MOCAP.