RESUMO
BACKGROUND: Neurodegenerative diseases (NDDs) are prevalent among older adults worldwide. Early diagnosis of NDD is challenging yet crucial. Gait status has been identified as an indicator of early-stage NDD changes and can play a significant role in diagnosis, treatment, and rehabilitation. Historically, gait assessment has relied on intricate but imprecise scales by trained professionals or required patients to wear additional equipment, causing discomfort. Advancements in artificial intelligence may completely transform this and offer a novel approach to gait evaluation. OBJECTIVE: This study aimed to use cutting-edge machine learning techniques to offer patients a noninvasive, entirely contactless gait assessment and provide health care professionals with precise gait assessment results covering all common gait-related parameters to assist in diagnosis and rehabilitation planning. METHODS: Data collection involved motion data from 41 different participants aged 25 to 85 (mean 57.51, SD 12.93) years captured in motion sequences using the Azure Kinect (Microsoft Corp; a 3D camera with a 30-Hz sampling frequency). Support vector machine (SVM) and bidirectional long short-term memory (Bi-LSTM) classifiers trained using spatiotemporal features extracted from raw data were used to identify gait types in each walking frame. Gait semantics could then be obtained from the frame labels, and all the gait parameters could be calculated accordingly. For optimal generalization performance of the model, the classifiers were trained using a 10-fold cross-validation strategy. The proposed algorithm was also compared with the previous best heuristic method. Qualitative and quantitative feedback from medical staff and patients in actual medical scenarios was extensively collected for usability analysis. RESULTS: The evaluations comprised 3 aspects. Regarding the classification results from the 2 classifiers, Bi-LSTM achieved an average precision, recall, and F1-score of 90.54%, 90.41%, and 90.38%, respectively, whereas these metrics were 86.99%, 86.62%, and 86.67%, respectively, for SVM. Moreover, the Bi-LSTM-based method attained 93.2% accuracy in gait segmentation evaluation (tolerance set to 2), whereas that of the SVM-based method achieved only 77.5% accuracy. For the final gait parameter calculation result, the average error rate of the heuristic method, SVM, and Bi-LSTM was 20.91% (SD 24.69%), 5.85% (SD 5.45%), and 3.17% (SD 2.75%), respectively. CONCLUSIONS: This study demonstrated that the Bi-LSTM-based approach can effectively support accurate gait parameter assessment, assisting medical professionals in making early diagnoses and reasonable rehabilitation plans for patients with NDD.
Assuntos
Aprendizado Profundo , Marcha , Doenças Neurodegenerativas , Idoso , Humanos , Inteligência Artificial , Aprendizado de Máquina , Doenças Neurodegenerativas/diagnósticoRESUMO
Objective:To investigate the distribution and antimicrobial resistance profile of clinical bacteria isolated from blood culture in China.Methods:The clinical bacterial strains isolated from blood culture from member hospitals of Blood Bacterial Resistant Investigation Collaborative System (BRICS) were collected during January 2018 to December 2019. Antibiotic susceptibility tests were conducted with agar dilution or broth dilution methods recommended by US Clinical and Laboratory Standards Institute (CLSI). WHONET 5.6 was used to analyze data.Results:During the study period, 14 778 bacterial strains were collected from 50 hospitals, of which 4 117 (27.9%) were Gram-positive bacteria and 10 661(72.1%) were Gram-negative bacteria. The top 10 bacterial species were Escherichia coli (37.2%), Klebsiella pneumoniae (17.0%), Staphylococcus aureus (9.7%), coagulase-negative Staphylococci (8.7%), Pseudomonas aeruginosa (3.7%), Enterococcus faecium (3.4%), Acinetobacter baumannii(3.4%), Enterobacter cloacae (2.9%), Streptococci(2.8%) and Enterococcus faecalis (2.3%). The the prevalence of methicillin-resistant S. aureus (MRSA) and methicillin-resistant coagulase-negative Staphylococcus were 27.4% (394/1 438) and 70.4% (905/1 285), respectively. No glycopeptide-resistant Staphylococcus was detected. More than 95% of S. aureus were sensitive to amikacin, rifampicin and SMZco. The resistance rate of E. faecium to vancomycin was 0.4% (2/504), and no vancomycin-resistant E. faecalis was detected. The ESBLs-producing rates in no carbapenem-resistance E. coli, carbapenem sensitive K. pneumoniae and Proteus were 50.4% (2 731/5 415), 24.6% (493/2001) and 35.2% (31/88), respectively. The prevalence of carbapenem-resistance in E. coli and K. pneumoniae were 1.5% (85/5 500), 20.6% (518/2 519), respectively. 8.3% (27/325) of carbapenem-resistance K. pneumoniae was resistant to ceftazidime/avibactam combination. The resistance rates of A. baumannii to polymyxin and tigecycline were 2.8% (14/501) and 3.4% (17/501) respectively, and that of P. aeruginosa to carbapenem were 18.9% (103/546). Conclusions:The surveillance results from 2018 to 2019 showed that the main pathogens of bloodstream infection in China were gram-negative bacteria, while E. coli was the most common pathogen, and ESBLs-producing strains were in majority; the MRSA incidence is getting lower in China; carbapenem-resistant E. coli keeps at a low level, while carbapenem-resistant K. pneumoniae is on the rise obviously.
RESUMO
Objective:To investigate the distribution and antimicrobial resistance profile of clinical bacteria isolated from blood culture in China.Methods:The clinical bacterial strains isolated from blood culture from member hospitals of Blood Bacterial Resistant Investigation Collaborative System (BRICS) were collected during January 2016 to December 2017. Antibiotic susceptibility tests were conducted by agar dilution or broth dilution methods recommended by US Clinical and Laboratory Standards Institute (CLSI) 2019. WHONET 5.6 was used to analyze data.Results:During the study period, 8 154 bacterial strains were collected from 33 hospitals, of which 2 325 (28.5%) were Gram-positive bacteria and 5 829 (71.5%) were Gram-negative bacteria. The top 10 bacterial species were Escherichia coli (34.7%), Klebsiella pneumoniae (15.8%), Staphylococcus aureus (11.3%), coagulase-negative Staphylococci (7.4%), Acinetobacter baumannii (4.6%), Pseudomonas aeruginosa (3.9%), Enterococcus faecium (3.8%), Streptococci (2.9%), Enterobacter cloacae (2.7%) and Enterococcus faecalis (2.5%). Methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-resistant coagulase-negative Staphylococcus (MRCNS) accounted for 34.2%(315/922) and 77.7%(470/605), respectively. No vancomycin-resistant Staphylococcus was detected. The resistance rate of Enterococcus faecium to vancomycin was 0.6%(2/312), and no vancomycin-resistant Enterococcus faecium was detected. The ESBLs-producing rates in Escherichia coli, Klebsiella pneumoniae and Proteus were 55.7%(1 576/2 831), 29.9%(386/1 289) and 38.5%(15/39), respectively. The incidences of carbapenem-resistance in Escherichia coli, Klebsiella pneumoniae were 1.2%(33/2 831), 17.5%(226/1 289), respectively. The resistance rates of Acinetobacter baumannii to polymyxin and tigecycline were 14.8%(55/372) and 5.9%(22/372) respectively, and those of Pseudomonas aeruginosa to polymyxin and carbapenem were 1.3%(4/315) and 18.7%(59/315), respectively. Conclusion:The surveillance results from 2016 to 2017 showed that the main pathogens of blood stream infection in China were gram-negative bacteria, while Escherichia coli was the most common pathogen; the MRSA incidence was lower than other surveillance data in the same period in China; carbapenem-resistant Escherichia coli was at a low level during this surveillance, while carbapenem-resistant Klebsiella pneumoniae is on the rise.
RESUMO
Objective To analyze the distribution and antimicrobial resistance profile of clinical bacterial strains isolated from blood culture in China.Methods Clinical bacterial strains isolated from blood culture from participating hospitals of Blood Bacterial Resistance Investigation Collaborative System (BRICS) during January 2014 to December 2015 were collected.Antibiotic susceptibility tests were conducted with agar dilution or broth dilution methods as recommended by US Clinical and Laboratory Standards Institute(CLSI)2018.The data were analyzed with Whonet 5.6 software.Results During the study period,4 801 clinical bacterial isolates were collected from 26 hospitals,of which 1 798 (37.5％) were Gram-positive bacteria and 3 003 (62.5％) were gram-negative bacteria.The top 10 isolates were Escherichia coli (33.8％),coagulase-negative Staphylococcus (19.0％),Klebsiella pneumoniae (11.9％),Staphylococcus aureus (10.1％),Acinetobacter baumannii (4.0％),Pseudomonas aeruginosa (3.8％),Streptococcus (3.0％),Enterobacter sulcus (2.9％),Enterococcus faecium (2.8％) and Enterococcus faecalis (1.8％).Methicillin-resistant Staphylococcus aureus (MRSA) and methicillinresistant coagulase-negative Staphylococcus (MRCNS) accounted for 33.9％ (165/487) and 56.9％ (520/913) of Staphylococcus aureus and coagulase-negative Staphylococcus respectively.No vancomycinresistant Staphylococcus was detected.The resistance rate of Enterococcus faecium to vancomycin was 0.7％ (1/135),and no vancomycin-resistant Enterococcus faecaliss was detected.The positive rates of extendedspectrum β-1actamases(ESBLs)-producing Escherichia coli,Klebsiella pneumoniae and Proteus were 56.9％ (923/1 621),30.1％ (172/572) and 29.2％ (7/24),respectively.The positive rates of carbapenemresistant Escherichia coli,Klebsiella pneumoniae,Enterobacter,Salmonella and Citrobacter were 1.2％ (20/1 621),7.2％ (41/572),4.3％ (6/141),1.5％ (1/67) and 2.9％ (1/34),respectively.The resistance rates of Acinetobacter baumannii to polymyxin and tegacycline were 2.6％ (5/190) and 8.9％ (17/190)respectively,and that of Pseudomonas aeruginosa to polymyxin and fosfomycin were 1.1％ (2/183)and 0.6％ (1/183),respectively.Conclusions The surveillance results from 2014 to 2015 show that the main pathogens of blood stream infection in China are Gram-negative bacteria,while Escherichia coli is the most common pathogen,the detection rate of MRSA is lower than other surveillance data in the same period in China;carbapenem-resistant Klebsiella pneumoniae and Escherichia coli are at a low level as shown in this surveillance.
