Accurate fall risk classification in elderly using one gait cycle data and machine learning.

BACKGROUND: Falls among the elderly are a major societal problem. While observations of medium-distance walking using inertial sensors identified potential fall predictors, classifying individuals at risk based on single gait cycles remains elusive. This challenge stems from individual variability and step-to-step fluctuations, making accurate classification difficult. METHODS: We recruited 44 participants, equally divided into high and low fall-risk groups. A smartphone secured on their second sacral spinous process recorded data during indoor walking. Features were extracted at each gait cycle from a 6-dimensional time series (tri-axial angular velocity and tri-axial acceleration) and classified using the gradient boosting decision tree algorithm. FINDINGS: Mean accuracy across five-fold cross-validation was 0.936. "Age" was the most influential individual feature, while features related to acceleration in the gait direction held the highest total relative importance when aggregated by axis (0.5365). INTERPRETATION: Combining acceleration, angular velocity data, and the gradient boosting decision tree algorithm enabled accurate fall risk classification in the elderly, previously challenging due to lack of discernible features. We reveal the first-ever identification of three-dimensional pelvic motion characteristics during single gait cycles in the high-risk group. This novel method, requiring only one gait cycle, is valuable for individuals with physical limitations hindering repetitive or long-distance walking or for use in spaces with limited walking areas. Additionally, utilizing readily available smartphones instead of dedicated equipment has potential to improve gait analysis accessibility.

Is radiographic lumbar spinal stenosis associated with the quality of life?: The Wakayama Spine Study.

OBJECTIVES: This prospective study aimed to determine the association between radiographic lumbar spinal stenosis (LSS) and the quality of life (QOL) in the general Japanese population. METHODS: The severity of radiographic LSS was qualitatively graded on axial magnetic resonance images as follows: no stenosis, mild stenosis with ≤1/3 narrowing, moderate stenosis with a narrowing between 1/3 and 2/3, and severe stenosis with > 2/3 narrowing. Patients less than 40 years of age and those who had undergone previous lumbar spine surgery were excluded from the study. The Oswestry Disability Index (ODI), which includes 10 sections, was used to assess the QOL. One-way analysis of variance was performed to determine the statistical relationship between radiographic LSS and ODI. Further, logistic regression analysis adjusted for gender, age, and body mass index was performed to detect the relationship. RESULTS: Complete data were available for 907 patients (300 men and 607 women; mean age, 67.3±12.4 years). The prevalence of severe, moderate, and non-mild/non-radiographic were 30%, 48%, and 22%, respectively. In addition, the mean values of ODI in each group were 12.9%, 13.1%, and 11.7%, respectively, and there was no statistically significant difference between the three groups in logistic analysis (P = 0.55). In addition, no significant differences in any section of the ODI were observed among the groups. However, severe radiographic LSS was associated with low back pain in the "severe" group as determined by logistic analysis adjusted for gender, age, and body mass index (odds ratio: 1.53, confidence interval: 1.13-2.07) compared with the non-severe group. CONCLUSION: In this general population study, severe radiographic LSS was associated with low back pain (LBP), but did not affect ODI.

The Impact of Heavy Snowfall on Home Care: A 2018 Case Study in Fukui City, Japan.

OBJECTIVE: Little is known regarding how home care is affected by extreme weather. In February 2018, Fukui City in Japan experienced unprecedented levels of snowfall. We examined snowfall impact on the provision of home care to elucidate whether patients incurred any harm. METHODS: A retrospective observational study using clinical and administrative records from a clinic in Fukui City was conducted on 294 patients (mean age = 69.9 ± 27.7 years; women = 60.5%, median age = 81). The study period was from February 5 to February 18, 2018. We analysed the patients' characteristics, daily trend of planned/actual patient visits, emergency transportation situations, and local snow accumulation. We summarized the situation in the clinic. RESULTS: There were 326 planned home visits, however only 121 (37%) occurred. Despite this, there were only 2 emergency transfers. Although the available clinical staff was limited, they managed to contact most patients via telephone and social networking services. CONCLUSION: Although the number of home visits dramatically decreased, the number of emergency transfers did not increase. This study therefore highlights the necessity for effective disaster preparation, such as assessment training, or use of telemedicine, and on-site decision-making to maintain home care during disasters.

The survival time of end-of-life home care patients in Fukui prefecture, Japan: A retrospective observational study.

ABSTRACT: End-of-life advance care planning (ACP) has become increasingly important in home care setting. In facilitating ACP discussion in home care setting, accurate understanding of patients' survival would be beneficial because it would facilitate healthcare professionals to individualize ACP discussion. However, little is known about survival outcome of home care patients. This study aimed to clarify the outcome of patients and identify factors to better predict the survival outcome of home care patients with the focus on patients' primary diseases.We conducted a retrospective analysis using data from 277 patients managed at a home care clinic in Japan and first treated in 2017 or 2018. Data regarding sociodemographic and clinical characteristics, and clinical outcome on December 31, 2019 were extracted. Using Kaplan-Meier product-limit method, we estimated the overall 30 days, 90 days, 1 year, and 3 year survival probabilities among the entire patients and their differences according to their primary disease. We also evaluated whether outcomes differed based on the primary disease or other factors using the hazard ratio and Cox proportional hazards regression.The overall survival probability was 82.5% at 30 days, 67.8% at 90 days, 52.7% at 1 year, and 39.1% at 3 years. The survival rates at 30 days, 90 days, 1 year, and 3 years were 64.6%, 33.4%, 9.5%, and 4.1% among cancer patients; 91.9%, 86.4%, 78.1%, and 47.0% among dementia patients; and 91.9%, 86.4%, 78.1%, and 47.0% among patients with other nervous and cerebrovascular diseases, respectively. Cox proportional hazard regression clarified that cancer patients (hazard ratio 6.53 [95% CI 4.16-10.28]) and older adults (hazard ratio 1.01 [95% CI 1.00-1.02]) were significantly more likely to die than dementia patients and young patients, respectively.Primary disease had a significant influence on the prediction of survival time and could be a useful indicator to individualize ACP in home care setting.

Feature selection to classify lameness using a smartphone-based inertial measurement unit.

BACKGROUND AND OBJECTIVES: Gait can be severely affected by pain, muscle weakness, and aging resulting in lameness. Despite the high incidence of lameness, there are no studies on the features that are useful for classifying lameness patterns. Therefore, we aimed to identify features of high importance for classifying population differences in lameness patterns using an inertial measurement unit mounted above the sacral region. METHODS: Features computed exhaustively for multidimensional time series consisting of three-axis angular velocities and three-axis acceleration were carefully selected using the Benjamini-Yekutieli procedure, and multiclass classification was performed using LightGBM (Microsoft Corp., Redmond, WA, USA). We calculated the relative importance of the features that contributed to the classification task in machine learning. RESULTS: The most important feature was found to be the absolute value of the Fourier coefficients of the second frequency calculated by the one-dimensional discrete Fourier transform for real input. This was determined by the fast Fourier transformation algorithm using data of a single gait cycle of the yaw angular velocity of the pelvic region. CONCLUSIONS: Using an inertial measurement unit worn over the sacral region, we determined a set of features of high importance for classifying differences in lameness patterns based on different factors. This completely new set of indicators can be used to advance the understanding of lameness.

Videoconferencing for Home Care Delivery in Japan: Observational Study.

BACKGROUND: Telemedicine has been increasingly used in many health care fields, including home care, where patients receive medical care at home. Owing to the current COVID-19 crisis, the value of telemedicine via videoconferencing is more recognized, particularly in allowing immobile patients to continue receiving care. However, the efficacy of telemedicine in home care settings in Japan remains to be fully appraised. OBJECTIVE: This study aims to identify the use and impact of telemedicine in a singular home care delivery setting in Japan. METHODS: A retrospective observational study was conducted using patient and other administrative records from a home care clinic. We considered patients who were involved in videoconferencing with home care physicians and telepresenters serving patients during 2018 and 2019. We extracted sociodemographic data of the patients and details of the videoconferencing and descriptively illustrated some specific cases. RESULTS: In a home care clinic in Japan, videoconferencing was conducted in 17 cases (involving 14 patients) over a 2-year period. Of all the cases, 12% (2/17) required emergency transfers and were hospitalized. A total of 88% (15/17) of cases remained; 71% (12/17) of cases were found to need extra medication or to go to a medical facility for consultation, whereas 18% (3/17) of cases were found not to be in need of urgent attention and were asked to rest. Problematic symptoms subsequently improved in 82% (14/17) of cases, and only 6% (1/17) of cases were later hospitalized. CONCLUSIONS: Telemedicine was deemed effective for assessing patients' conditions in the home care setting in situations where home visits by a physician cannot be carried out. Our findings indicate that consultations via videoconferencing are safe and effective, suggesting more active use of videoconferencing in other clinical contexts.

Emergency transfers of home care patients in Fukui Prefecture, Japan: A retrospective observational study.

Little is known about how emergency transfers take place and what outcomes they lead to in the patients who receive home care in Japan. We aimed to assess outcomes of emergency transfers and factors associated with such outcomes in the Japanese home care setting.A retrospective analysis of patient data from a home care clinic in Fukui, Japan, included all patients who experienced emergency transfers which were reported to the clinic during 2018 and 2019. We collected data on patients' sociodemographic and clinical characteristics, as well as the transfer process and its outcome, using patient charts and other administrative records. We first analyzed the overall outcome and then evaluated whether transfer outcomes would differ according to by whom and from where the emergency medical service (EMS) was called, by univariate and multivariate analyses.We considered 63 patients who experienced emergency transfers during the study period. Of the total, 10 (15.9%) returned to their residences without being admitted or being dead on arrival. Although only 2.6% (1/39) of patients whose transfers were determined by health care professionals (HCPs) returned home without being admitted, a direct return was observed for 37.5% (9/24) of patients whose transfer was determined by those other than HCPs (odds ratio of direct return to residences 22.80, 95% confidence interval 2.65-195.87). There was no other variable which was significantly associated with the outcomes after the emergency transfers, although all the patients who have no available caregivers resulted in hospitalization.In this preliminary analysis in the Japanese home care setting, only a small proportion of patients returned to their residences without being admitted following emergency transfers. Patients whose EMS transfer was requested by an HCP usually resulted in an admission to the clinic, whereas transfers requested by non-HCPs frequently did not.