YOLOX-SwinT algorithm improves the accuracy of AO/OTA classification of intertrochanteric fractures by orthopedic trauma surgeons.

PURPOSE: Intertrochanteric fracture (ITF) classification is crucial for surgical decision-making. However, orthopedic trauma surgeons have shown lower accuracy in ITF classification than expected. The objective of this study was to utilize an artificial intelligence (AI) method to improve the accuracy of ITF classification. METHODS: We trained a network called YOLOX-SwinT, which is based on the You Only Look Once X (YOLOX) object detection network with Swin Transformer (SwinT) as the backbone architecture, using 762 radiographic ITF examinations as the training set. Subsequently, we recruited 5 senior orthopedic trauma surgeons (SOTS) and 5 junior orthopedic trauma surgeons (JOTS) to classify the 85 original images in the test set, as well as the images with the prediction results of the network model in sequence. Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) 20.0 (IBM Corp., Armonk, NY, USA) to compare the differences among the SOTS, JOTS, SOTS + AI, JOTS + AI, SOTS + JOTS, and SOTS + JOTS + AI groups. All images were classified according to the AO/OTA 2018 classification system by 2 experienced trauma surgeons and verified by another expert in this field. Based on the actual clinical needs, after discussion, we integrated 8 subgroups into 5 new subgroups, and the dataset was divided into training, validation, and test sets by the ratio of 8:1:1. RESULTS: The mean average precision at the intersection over union (IoU) of 0.5 (mAP50) for subgroup detection reached 90.29%. The classification accuracy values of SOTS, JOTS, SOTS + AI, and JOTS + AI groups were 56.24% ± 4.02%, 35.29% ± 18.07%, 79.53% ± 7.14%, and 71.53% ± 5.22%, respectively. The paired t-test results showed that the difference between the SOTS and SOTS + AI groups was statistically significant, as well as the difference between the JOTS and JOTS + AI groups, and the SOTS + JOTS and SOTS + JOTS + AI groups. Moreover, the difference between the SOTS + JOTS and SOTS + JOTS + AI groups in each subgroup was statistically significant, with all p < 0.05. The independent samples t-test results showed that the difference between the SOTS and JOTS groups was statistically significant, while the difference between the SOTS + AI and JOTS + AI groups was not statistically significant. With the assistance of AI, the subgroup classification accuracy of both SOTS and JOTS was significantly improved, and JOTS achieved the same level as SOTS. CONCLUSION: In conclusion, the YOLOX-SwinT network algorithm enhances the accuracy of AO/OTA subgroups classification of ITF by orthopedic trauma surgeons.

Study of the Immediately Detection of Mild Traumatic Brain Injury by Feature Engineering on Electroencephalography.

The electroencephalographic (EEG) diagnosis of mild traumatic brain injury (mTBI) is not usually timely, and the detection is often performed several hours or days after the trauma, leading to a decrease in the accuracy of its detection. In this study, EEG signals are recorded immediately after mTBI by connecting a bipolar single lead to injured animals. And three types of EEG features, namely time domain, frequency domain, and nonlinear dynamics, are screened for optimal feature subset in mTBI detection. First, EEG signals of animals are recorded before and after establishing the animal model of mTBI. Second, signal preprocessing, feature extraction, and feature preprocessing are performed to obtain the full-feature dataset, and 1442 feature subsets are obtained by 15 feature reduction algorithms extracted from combinations of 47 features. Ultimately, the support vector machines and K-nearest neighbor algorithms are trained and tested respectively, and their performance is comprehensively compared to determine the optimal feature subset for mTBI detection. In the EEG dataset collected in this study, a total of eight feature subsets extracted from combinations of original 47 features and classification models with 100% accuracy are obtained. This study shows the perspective of immediately detecting mTBI based on a bipolar single-lead EEG.

Functional plasticity abnormalities over the lifespan of first-episode patients with major depressive disorder: a resting state fMRI study.

BACKGROUND: Neurodevelopmental and neurodegenerative theories of depression suggest that patients with major depressive disorder (MDD) may follow abnormal developmental, maturational, and aging processes. However, a lack of lifespan studies has precluded verification of these theories. Herein, we analyzed functional magnetic resonance imaging (fMRI) data to comprehensively characterize age-related functional trajectories, as measured by the fractional amplitude of low frequency fluctuations (fALFF), over the course of MDD. METHODS: In total, 235 MDD patients with age-differentiated onsets and 235 age- and sex-matched healthy controls (HC) were included in this study. We determined the pattern of age-related fALFF changes by cross-sectionally establishing the general linear model (GLM) between fALFF and age over a lifespan. Furthermore, the subjects were divided into four age groups to assess age-related neural changes in detail. Inter-group fALFF comparison (MDD vs. HC) was conducted in each age group and Granger causal analysis (GCA) was applied to investigate effective connectivity between regions. RESULTS: Compared with the HC, no significant quadratic or linear age effects were found in MDD over the entire lifespan, suggesting that depression affects the normal developmental, maturational, and degenerative process. Inter-group differences in fALFF values varied significantly at different ages of onset. This implies that MDD may impact brain functions in a highly dynamic way, with different patterns of alterations at different stages of life. Moreover, the GCA analysis results indicated that MDD followed a distinct pattern of effective connectivity relative to HC, and this may be the neural basis of MDD with age-differentiated onsets. CONCLUSIONS: Our findings provide evidence that normal developmental, maturational, and ageing processes were affected by MDD. Most strikingly, functional plasticity changes in MDD with different ages of onset involved dynamic interactions between neuropathological processes in a tract-specific manner.

Preliminary injury risk estimation for occupants involved in frontal crashes by combining computer simulations and real crashes.

The fatality rate can be dramatically reduced with the help of emergency medical services. The purpose of this study was to establish a computational algorithm to predict the injury severity, so as to improve the timeliness, appropriateness, and efficacy of medical care provided. The computer simulations of full-frontal crashes with rigid wall were carried out using LS-DYNA and MADYMO under different collision speeds, airbag deployment time, and seatbelt wearing condition, in which a total of 84 times simulation was conducted. Then an artificial neural network is adopted to construct relevance between head and chest injuries and the injury risk factors; 37 accident cases with Event Data Recorder data and information on occupant injury were collected to validate the model accuracy through receiver operating characteristic analysis. The results showed that delta-v, seatbelt wearing condition, and airbag deployment time were important factors in the occupant's head and chest injuries. When delta-v increased, the occupant had significantly higher level of severe injury on the head and chest; there is a significant difference of Head Injury Criterion and Combined Thoracic Index whether the occupant wore seatbelt. When the airbag deployment time was less than 20 ms, the severity of head and chest injuries did not significantly vary with the increase of deployment time. However, when the deployment time exceeded 20 ms, the severity of head and chest injuries significantly increased with increase in deployment time. The validation result of the algorithm showed that area under the curve = 0.747, p < 0.05, indicating a medium level of accuracy, nearly to previous model. The computer simulation and artificial neural network have a great potential for developing injury risk estimation algorithms suitable for Advanced Automatic Crash Notification applications, which could assist in medical decision-making and medical care.

A 6-year survey of road traffic accidents in Southwest China: Emphasis on traumatic brain injury.

BACKGROUND: The objective of this study is to provide an up-to-date overview of the patterns of injuries, especially traumatic brain injury (TBI) caused by RTAs and to discuss some of the public health consequences. METHODS: A scientific team was established to collect road traffic accidents occurring between 2013 and 2018 in Chongqing, Southwest China. For each accident, the environment-, vehicle-, and person- variables were analyzed and determined. The overall injury distribution and TBI patterns of four types of road users (driver, passenger, motorcyclist and pedestrian) were compared. The environmental and time distribution of accidents with TBI were shown by bar and pie chart. The risks of severe brain injury whether motorcyclist wearing helmets or not were compared and the risk factors of severe TBI in pedestrian were determined by odds ratio analysis. RESULTS: This study enrolled 2131 accidents with 2741 persons of all kind of traffic participants, 1149 of them suffered AIS1+ head injury and 1598(58%) died in 7 days. The most common cause of deaths is due to head injury with 714(85%) and 1266(79%) persons died within 2 hours. Among 423 persons suffered both skull fracture and intracranial injury, 102 (24.1%) have an intracranial injury but no skull fractures, while none of the skull fractures without intracranial injury was found. Besides, motorcyclists without a helmet were at higher risks for all the brain injury categories. The risk of pedestrian suffering severe TBI at an impact speed of more than 70 km/h is 100 times higher than that with an impact speed of less than 40 km/h. CONCLUSION: It is urgently needed to develop a more reliable brain injury evaluation criterion for better protection of the road users. We believe that strengthening the emergency care to head injury at the scene is the most effective way to reduce traffic fatality.

Parameter sensitivity analysis of pedestrian head dynamic response and injuries based on coupling simulations.

There are a very limited number of reports studying on the dynamic response and injuries of pedestrian head in the scenarios with head hitting windshield. This study aims to investigate the significant factors that affect the dynamic response and injuries of pedestrian head through finite element-multi-body coupling simulations. Two finite element vehicle models and two multi-body pedestrian human models were used to build the coupling simulations. Orthogonal experimental design and analysis of variance were used for parameter combination and data analysis. This study demonstrated that the dynamic response of pedestrian head and HIC15 were strongly associated with collision speed and pedestrian orientation. Vehicle type had a significant influence on the dynamic response of pedestrian head and HIC15, while there was no significant relationship between the dynamic response of pedestrian head and HIC15 and the size of pedestrian human models. Collision speed, pedestrian orientation, and vehicle type should be prioritized over the other collision parameters in the study of head injury mechanism and reconstruction of vehicle-pedestrian collisions in the scenarios with head hitting windshield.

Automatic identification of compressions and ventilations during CPR based on the fuzzy c-means clustering and deep belief network.

BACKGROUND: A transthoracic impedance (TTI) signal is an important indicator of the quality of chest compressions (CCs) during cardiopulmonary resuscitation (CPR). We proposed an automatic detection algorithm including the wavelet decomposition, fuzzy c-means (FCM) clustering, and deep belief network (DBN) to identify the compression and ventilation waveforms for evaluating the quality of CPR. METHODS: TTI signals were collected from a cardiac arrest model that electrically induced cardiac arrest in pigs. All signals were denoised using the wavelet and morphology method. The potential compression and ventilation waveforms were marked using an algorithm with a multi-resolution window. The compressions and ventilations in these waveforms were identified and classified using the FCM clustering and DBN methods. RESULTS: Using the FCM clustering method, the positive predictive values (PPVs) for compressions and ventilations were 99.7% and 95.7%, respectively. The sensitivities of recognition were 99.8% for compressions and 95.1% for ventilations. The DBN approach exhibited similar PPV and sensitivity results to the FCM clustering method. The time cost was satisfactory using either of these techniques. CONCLUSIONS: Our findings suggest that FCM clustering and DBN can be utilized to effectively and accurately evaluate CPR quality, and provide information for improving the success rate of CPR. Our real-time algorithms using FCM clustering and DBN eliminated most distortions and noises effectively, and correctly identified the compression and ventilation waveforms with a low time cost.