Objective To develop a deep learning system for CT images to assist in the diagnosis of thoracolumbar fractures and analyze the feasibility of its clinical application. Methods Collected from West China Hospital of Sichuan University from January 2019 to March 2020, a total of 1256 CT images of thoracolumbar fractures were annotated with a unified standard through the Imaging LabelImg system. All CT images were classified according to the AO Spine thoracolumbar spine injury classification. The deep learning system in diagnosing ABC fracture types was optimized using 1039 CT images for training and validation, of which 1004 were used as the training set and 35 as the validation set; the rest 217 CT images were used as the test set to compare the deep learning system with the clinician’s diagnosis. The deep learning system in subtyping A was optimized using 581 CT images for training and validation, of which 556 were used as the training set and 25 as the validation set; the rest 104 CT images were used as the test set to compare the deep learning system with the clinician’s diagnosis. Results The accuracy and Kappa coefficient of the deep learning system in diagnosing ABC fracture types were 89.4% and 0.849 (P<0.001), respectively. The accuracy and Kappa coefficient of subtyping A were 87.5% and 0.817 (P<0.001), respectively. Conclusions The classification accuracy of the deep learning system for thoracolumbar fractures is high. This approach can be used to assist in the intelligent diagnosis of CT images of thoracolumbar fractures and improve the current manual and complex diagnostic process.
ObjectiveTo encourage clinicians to code the major diagnosis of diseases, in order to improve the correct rate of disease major diagnosis coding. MethodsWe analyzed the data of major diagnostic codes by clinicians from January 2012 to June 2013. The group leader of the clinical treatment was designated to be responsible for the disease coding. Disease coders introduced knowledge of international classification of diseases to the clinical department according to the different characteristics of disease in each department and communicated with clinicians on the problems of disease coding. Then, we tried to find out whether this method could improve the correct rate of major diagnosis coding of diseases. ResultsThe rate of disease major coding by clinicians of the whole hospital and pilot departments increased from 94.081% to 98.301%. The correct rate of disease major coding decreased from 75.824% to 67.483% and then reached 81.893%. The correct rate of disease major coding of the Department of Hematology was 83.824% in August 2012 and then decreased with the lowest rate of 68.025%; and the correct rate of disease major coding of the Department of Orthopedics increased rapidly and reached 90% in September 2012. ConclusionsThrough the leader of the clinical treatment being responsible for the disease coding and encouraging clinicians to code the main diagnosis of diseases, the accurate of disease major diagnosis coding has improved. Strengthening the communication between clinical and Medical Record Departments can help our hospital improve the quality of disease major diagnosis coding continuously.
Objective To study the clinical classification and etiologies of uveitis based on 1214 uveitis patients reffered to Zhongshan Ophthalmic Center. Methods A retrospective analysis was made on the patients with uveitis, coming from all over China between January 1996 and December 2001. All kinds of uveitis were classified according to the anatomical criteria and etiological criteria. The relevant data of these patients, such as the age at uveitis onset and sex were also analyzed. Results The total number of the patients is 1214 (male 698, female 516), with the average age at disease onset being 34.43. Anterior uveitis, the most common type, was seen in 546 cases, accounting for 44.98% of all the patients, followed in descending order by panuveitis (530 cases, 43.66%), intermediate uveitis(78 cases, 6.43%) and posterior uveitis(60 cases, 4.94%). Etiological factors and clinical entities were identified in 703 patients, accounting for 57.91% of all the patients, and the other 511 patients were idiopathic ones. The most common types of anterior uveitis were idiopathic uveitis(316 cases, 57.88%), followed by Fuchs syndrome(85 cases) and ankylosing spondylitis(45 cases). BehCcedil;et disease(218 cases, 41.13%) and Vogt-Koyanagi-Harada syndrome(196 cases, 36.98%) were the most common entities in panuveitis. Neither etiological factors nor clinical entities could be identified in the patients with intermediate uveitis and those with posterior uveitis. Conclusions Uveitis occurs mostly in young and middle-aged adults. In general, a predilection was seen in the male as compared with the female in the development of uveitis. Idiopathic anterior uveitis, BehCcedil;et disease and Vogt-Koyanagi-Harada syndrome are the most common entities of uveitis seen in China. Classification based on etiological and anatomical factors may provide a reasonable system for the study of uveitis. (Chin J Ocul Fundus Dis, 2002, 18: 253-255)
Identification of molecular subtypes of malignant tumors plays a vital role in individualized diagnosis, personalized treatment, and prognosis prediction of cancer patients. The continuous improvement of comprehensive tumor genomics database and the ongoing breakthroughs in deep learning technology have driven further advancements in computer-aided tumor classification. Although the existing classification methods based on gene expression omnibus database take the complexity of cancer molecular classification into account, they ignore the internal correlation and synergism of genes. To solve this problem, we propose a multi-layer graph convolutional network model for breast cancer subtype classification combined with hierarchical attention network. This model constructs the graph embedding datasets of patients’ genes, and develops a new end-to-end multi-classification model, which can effectively recognize molecular subtypes of breast cancer. A large number of test data prove the good performance of this new model in the classification of breast cancer subtypes. Compared to the original graph convolutional neural networks and two mainstream graph neural network classification algorithms, the new model has remarkable advantages. The accuracy, weight-F1-score, weight-recall, and weight-precision of our model in seven-category classification has reached 0.851 7, 0.823 5, 0.851 7 and 0.793 6 respectively. In the four-category classification, the results are 0.928 5, 0.894 9, 0.928 5 and 0.865 0 respectively. In addition, compared with the latest breast cancer subtype classification algorithms, the method proposed in this paper also achieved the highest classification accuracy. In summary, the model proposed in this paper may serve as an auxiliary diagnostic technology, providing a reliable option for precise classification of breast cancer subtypes in the future and laying the theoretical foundation for computer-aided tumor classification.
The eye-computer interaction technology based on electro-oculogram provides the users with a convenient way to control the device, which has great social significance. However, the eye-computer interaction is often disturbed by the involuntary eye movements, resulting in misjudgment, affecting the users’ experience, and even causing danger in severe cases. Therefore, this paper starts from the basic concepts and principles of eye-computer interaction, sorts out the current mainstream classification methods of voluntary/involuntary eye movement, and analyzes the characteristics of each technology. The performance analysis is carried out in combination with specific application scenarios, and the problems to be solved are further summarized, which are expected to provide research references for researchers in related fields.
ObjectiveBased on the localization of resource-based relative value scale (RBRVS) in H Hospital, to implement a surgical performance management model reform with the main surgery as the core, and to construct a more scientific and fair surgical performance distribution system. MethodsA surgical performance management model with the main surgery as the core was constructed. Relevant data such as RBRVS, diagnosis related groups (DRG), case mixed index (CMI), and surgical time of 65 915 inpatient elective surgeries in H Hospital in 2023 were collected and organized. Large sample historical data analysis was conducted using SPSS software, and the rationality of the optimized surgical performance management model was verified through key indicators. ResultsThe total coefficient of multiple orders for surgery in the 22 departments included in the study was highly correlated with the main surgery coefficient (γ>0.85), and the matching coefficients for each specialty were significantly different (P<0.001). The surgical performance management model with the main surgery as the core showed a significant improvement in the key indicators (doctor’s time resource investment and surgical risk and difficulty). ConclusionBy implementing a surgical performance management model with the main surgery as the core, we aim to strengthen the performance orientation that reflects the risks and difficulty of diagnosis and treatment, as well as the value of doctor services. This will guide clinical practice to return to the essence of medicine, support the development of discipline construction, and further stimulate the vitality and motivation of clinical work.
How to extract high discriminative features that help classification from complex resting-state fMRI (rs-fMRI) data is the key to improving the accuracy of brain disease recognition such as schizophrenia. In this work, we use a weighted sparse model for brain network construction, and utilize the Kendall correlation coefficient (KCC) to extract the discriminative connectivity features for schizophrenia classification, which is conducted with the linear support vector machine. Experimental results based on the rs-fMRI of 57 schizophrenia patients and 64 healthy controls show that our proposed method is more effective (i.e., achieving a significantly higher classification accuracy, 81.82%) than other competing methods. Specifically, compared with the traditional network construction methods (Pearson’s correlation and sparse representation) and the commonly used feature selection methods (two-sample t-test and Least absolute shrinkage and selection operator (Lasso)), the algorithm proposed in this paper can more effectively extract the discriminative connectivity features between the schizophrenia patients and the healthy controls, and further improve the classification accuracy. At the same time, the discriminative connectivity features extracted in the work could be used as the potential clinical biomarkers to assist the identification of schizophrenia.
ObjectiveTo systematically review the current status of doctor-patient conflicts in China.MethodsWe searched CNKI and CSSCI databases to collect literatures about the doctor-patient conflicts from inception to April 23rd, 2017. The literatures were categorized by the published time, the high-frequency vocabulary, the citation frequency, the researching discipline, the researching facility, the quality of literature, the theme of literature and so on. The current research status of the doctor-patient conflicts was analyzed.ResultsA total of 226 literatures were included, in which 72 defined and classified the doctor-patient conflicts, 122 analyzed the causing reasons of doctor-patient conflicts, and 160 analyzed the governance paths of doctor-patient conflicts. The research disciplinary vision was limited to the policies, regulations and the medical education and so on, and the researches in psychology or economics disciplinary vision were insufficient. The medical and comprehensive universities were the main research units of the studies of the doctor-patient conflicts. The frequency and quality of the researches about doctor-patient conflicts were low.ConclusionThe classified studies of doctor-patient conflicts are insufficient, so the scientific and manageable classified criterions are needed in the further studies. Systematic studies in influential factors of doctor-patient conflicts are insufficient, so the occurring mechanisms of conflicts are needed to be done by systematical researches on patient-centered way. The studies of governance paths of doctor-patient conflicts are insufficient, so the strategies of classified and systematical management which according to the different conflicting forms and entire process of the conflict occurrence should be put forward.