ABSTRACT
BACKGROUND: Gait disturbances may appear prior to cognitive dysfunction in the early stage of silent cerebrovascular disease (SCD). Subtle changes in gait characteristics may provide an early warning of later cognitive decline. Our team has proposed a vision-based artificial intelligent gait analyzer for the rapid detection of spatiotemporal parameters and walking pattern based on videos of the Timed Up and Go (TUG) test. The primary objective of this study is to investigate the relationship between gait features assessed by our artificial intelligent gait analyzer and cognitive function changes in patients with SCD. METHODS: This will be a multicenter prospective cohort study involving a total of 14 hospitals from Shanghai and Guizhou. One thousand and six hundred patients with SCD aged 60-85 years will be consecutively recruited. Eligible patients will undergo the intelligent gait assessment and neuropsychological evaluation at baseline and at 1-year follow-up. The intelligent gait analyzer will divide participant into normal gait group and abnormal gait group according to their walking performance in the TUG videos at baseline. All participants will be naturally observed during 1-year follow-up period. Primary outcome are the changes in Mini-Mental State Examination (MMSE) score. Secondary outcomes include the changes in intelligent gait spatiotemporal parameters (step length, gait speed, step frequency, step width, standing up time, and turning back time), the changes in scores on other neuropsychological tests (Montreal Cognitive Assessment, the Stroop Color Word Test, and Digit Span Test), falls events, and cerebrovascular events. We hypothesize that both groups will show a decline in MMSE score, but the decrease of MMSE score in the abnormal gait group will be more significant. CONCLUSION: This study will be the first to explore the relationship between gait features assessed by an artificial intelligent gait analyzer and cognitive decline in patients with SCD. It will demonstrate whether subtle gait abnormalities detected by the artificial intelligent gait analyzer can act as a cognitive-related marker for patients with SCD. TRIAL REGISTRATION: This trial was registered at ClinicalTrials.gov ( NCT04456348 ; 2 July 2020).
Subject(s)
Cerebrovascular Disorders , Cognitive Dysfunction , Cerebrovascular Disorders/complications , Cerebrovascular Disorders/diagnosis , China , Cognition , Cognitive Dysfunction/diagnosis , Gait , Humans , Multicenter Studies as Topic , Prospective StudiesABSTRACT
BACKGROUND: Early detection of gait abnormalities is critical for preventing severe injuries in future falls. The timed up and go (TUG) test is a commonly used clinical gait screening test; however, the interpretation of its results is limited to the TUG total time. RESEARCH QUESTION: What is diagnostic accuracy of the low-cost, markerless, automated gait analyzer, with the aid of vision-based artificial intelligence technology, which extract gait spatiotemporal features and screen for abnormal walking patterns through video recordings of the TUG test? METHODS: Our dataset contained retrospective data from outpatients from the Department of Neurology or Rehabilitation of two tertiary hospitals in Shanghai. A panel of three expert neurologists specialized in movement disorders reviewed the gait performance in each TUG video, and labeled them separately, with the most commonly assigned label being used as the reference standard. The gait analyzer performed the AlphaPose algorithm to track the human joint position and calculated the spatiotemporal parameters by filtering and double-threshold signal detection. Gait spatiotemporal features and expert labels were input into machine learning models, and the accuracy of each model was tested with leave-one-out cross-validation (LOOCV). RESULTS: A total of 284 participants were recruited. Among these, 100 were labeled as having abnormal gait performance by experts. The Naive Bayes classifier achieved the best performance with a full-data accuracy of 90.14% and a LOOCV accuracy of 89.08% for screening abnormal gait performance. SIGNIFICANCE: This study is the first to investigate the accuracy of a vision-based intelligent gait analyzer for screening abnormal clinical gait performance. By virtue of a pose estimation algorithm and machine learning models, our intelligent gait analyzer can detect abnormal walking patterns approximate to judgements made by experienced neurologists, which is expected to be a supplementary gait assessment protocol for basic-level doctors in the future.
Subject(s)
Artificial Intelligence , Movement Disorders , Bayes Theorem , China , Gait , Humans , Retrospective StudiesABSTRACT
Background: Idiopathic normal pressure hydrocephalus (iNPH) is a common disease in elderly adults. Patients with iNPH are generally characterized by progressive gait impairment, cognitive deficits, and urinary urgency and/or incontinence. A number of radiographic studies have shown that iNPH patients have enlarged ventricles and altered brain morphology; however, few studies have focused on the relationships between altered brain structure and gait dysfunction due to iNPH. Thus, this study aimed to evaluate the abnormalities of white matter (WM) correlated with gait impairment in iNPH patients and to gain a better understanding of its underlying pathology. Methods: Fifteen iNPH patients (five women, 10 men) were enrolled in this study, and each patient's demographic and gait indices were collected. First, we performed a correlation analysis between the demographic and gait indices. Then, all gait indices were grouped according to the number of WM hyperintensities (WMH) among each WM tract (JHU WM tractography atlas), to perform comparative analysis. Results: Considering sex and illness duration as covariates, correlation analysis showed a significantly negative correlation between step length (r = -0.80, p = 0.001), pace (r = -0.84, p = 2.96e-4), and age. After removing the effects of age, sex, and illness duration, correlation analysis showed negative correlation between step length (r = -0.73, p = 0.007), pace (r = -0.74, p = 0.005), and clinical-grade score and positive correlation between 3-m round trip time (r = 0.66, p = 0.021), rising time (r = 0.76, p = 0.004), and clinical-grade score. Based on WMH of each white matter tract, gait indices showed significant differences (p < 0.05/48, corrected by Bonferroni) between fewer WMH patients and more WMH in the middle cerebellar peduncle, left medial lemniscus, left posterior limb of the internal capsule (IC), and right posterior limb of the IC. Conclusions: Our results indicated that iNPH patients exhibited gait-related WM abnormalities located in motor and sensory pathways around the ventricle, which is beneficial to understand the underlying pathology of iNPH.
ABSTRACT
OBJECTIVE: To evaluate the therapeutic effects of recombinant mutant human tumor necrosis factor-related apoptosis-inducing ligand (rmhTRAIL) on non-small cell cancer (NSCLC). METHODS: NSCLC cells of the line NCI-H460 were cultured and underwent agarose gel electrophoresis. rmhTRAIL of different concentrations was added into the culture fluid to observe the influence of rmhTRAIL. Nude BALB/c rats were transplanted with rat NCI-H460 tumor. Then the rats carrying xenografts were randomly divided into 5 groups of 8 rats: negative control group to be injected intraperitoneally with normal saline; positive control group to be injected with vancristine; low-dose rmhTRAIL group, to be injected with 1.7 mg/kg rmhTRAIL, middle-dose rmhTRAIL group to be injected with rmhTRAIL 5.0 mg/kg, and high-dose rmhTRAIL group to be injected with 15.0 mg/kg rmhTRAIL. The size of tumor was measured every 3 approximately 4 days. Ten days after the administration of different drugs the rats were killed and the tumors were taken out to undergo TUNEL staining for microscopy. RESULTS: The relative tumor volume of the low-dose rmhTRAIL group and high-dose rmhTRAIL group were 3.19 +/- 2.05 and 1.47 +/- 0.77 respectively, both significantly smaller than that of the negative group (8.48 +/- 5.87, P < 0.05 and P < 0.01). The relative tumor growth rate of the low-dose rmhTRAIL group and high-dose rmhTRAIL group were 37.6% and 17.3% respectively. The tumor weight of low-dose rmhTRAIL group and high-dose rmhTRAIL group were 1.09 g +/- 0.55 g and 0.31 g +/- 0.09 g respectively, both significantly lighter than that of the negative control group (2.78 +/- 0.77, both P < 0.01). Large amount of apoptotic cells were seen in the tumor tissues of the rmhTRAIL -treated rats and tumors cells cultured with rmhTRAIL. Agarose gel electrophoresis showed apoptosis-characteristic ladder in the DNA extracted from the rmhTRAIL -treated tumors cells. CONCLUSION: rmhTRAIL dose-dependently inhibit the growth of NSCLC cells, primarily by inducing apoptosis of tumor cells.