Objective To discuss the treatment of craniocerebral injuries caused by earthquake. Methods Retrospective analysis of clinical information for 256 patients with craniocerebral injury caused by an earthquake. Results The ‘Classification and Treatment’ was applied to the patients, whether or not they were operated on. A total of 146 patients were cured, 68 improved, 24 remained dependent on the care of others, and 8 died. The mortality rate was 3.13%. Conclusion Applying the ‘Classification and Treatment’ to patients with craniocerebral injury following an earthquake supported the use of medical resources and was associated with a low rate of death and disability.
Objective To review the research progress of the risk factors for slip progression and the pathogenesis of lumbosacral spondylolisthesis, and to discuss the value of Spinal Deformity Study Group (SDSG) classification system for lumbosacral spondylolisthesis. Methods Recent articles about the risk factors for slip progression and the pathogenesis of lumbosacral spondylolisthesis were reviewed and comprehensively analyzed with SDSG classification system of lumbosacral spondylolisthesis. Results Pelvic incidence (PI) is the key pathogenic factor of lumbosacral spondylolisthesis. The Meyerding grade of slip, PI, sacro-pelvic balance, and spino-pelvic balance not only are the fundamental risk factors of slip progression, but also are the key factors to determine how to treat and influence the prognosis. Therefore, compared with Wiltse, Marchetti-Bartolozzi, and Mac-Thiong-Labelle classification systems of lumbosacral spondylolisthesis, SDSG classification based on these factors mentioned above, has better homogeneity between the subjects of subgroup, and better reliability, moreover, could better guide operative plan and judge the prognosis. Conclusion It is suggested that the SDSG classification system should be the standard classification for lumbosacral spondylolisthesis for the clinical and research work.
Feature extraction methods and classifier selection are two critical steps in heart sound classification. To capture the pathological features of heart sound signals, this paper introduces a feature extraction method that combines mel-frequency cepstral coefficients (MFCC) and power spectral density (PSD). Unlike conventional classifiers, the adaptive neuro-fuzzy inference system (ANFIS) was chosen as the classifier for this study. In terms of experimental design, we compared different PSDs across various time intervals and frequency ranges, selecting the characteristics with the most effective classification outcomes. We compared four statistical properties, including mean PSD, standard deviation PSD, variance PSD, and median PSD. Through experimental comparisons, we found that combining the features of median PSD and MFCC with heart sound systolic period of 100–300 Hz yielded the best results. The accuracy, precision, sensitivity, specificity, and F1 score were determined to be 96.50%, 99.27%, 93.35%, 99.60%, and 96.35%, respectively. These results demonstrate the algorithm’s significant potential for aiding in the diagnosis of congenital heart disease.
One hundred and thirty patients with uveitis in north-western zone of our country were analyzed based on anatomical classification and their causes. It was found that anterior uveitis was the commonest type in uveitis,accounting for 86.15% of total patients. Intermediate uveitis, pan-uveitis and posterior uveitis accounted repectively for 6.92%, 3.85%and3.08% of the total patients. Rheumatic arthritis was the most frequently accompanied systemic disease in patients with uveitis,showing a possibly causative link between them in their pathogenesis. （Chin J Ocul Fundus Dis,1994,10:156-158）
ObjectiveTo study the classification criteria of adult cuboid fracture and its guidance feasibility and effect of treatment. MethodsA retrospective analysis was made on the clinical data of 415 adult patients (416 feet) with cuboid fractures who had complete CT data treated between May 2009 and April 2014. There were 337 males and 78 females, aged 19 to 64 years (mean, 38.8 years). The left foot, right foot, and bilateral feet were involved in 220 cases, 194 cases, and 1 case respectively. The causes of injury were sprain in 106 cases, traffic accident in 65 cases, falling from height in 129 cases, and heavy crushing in 115 cases. The interval of injury and hospitalization was 2 hours to 3 days (mean, 8.5 hours). Based on CT findings, the classification criteria of cuboid fracture was proposed and methods of treatment was statistically analyzed. The external fixation surgery was performed in patients of type I (285 feet), type IIa (18 feet), and type III (5 feet); open reduction and internal fixation were performed in patients of type IIb (41 feet) and type III (67 feet), and bone grafting was used to repair defects in 58 feet (type III). ResultsAll patients were followed up 1 year to 5 years and 11 months (mean, 2 years and 3 months). Primary healing of incision was obtained. In patients with type I fracture, fracture healed in 165 feet at 4-6 weeks (mean, 5.5 weeks), fracture did not heal in the other 120 feet; the American Orthopaedic Foot and Ankle Society (AOFAS) score was 95-100(mean, 96.7) at last follow-up. In patients with type II fracture, fracture healed in all feet at 6-8 weeks (mean, 6.5 weeks); the AOFAS score was 92-100(mean, 95.5) at last follow-up. In patients with type III fracture, malunion was observed at 6-8 weeks in 5 feet undergoing external fixation, and in 9 feet undergoing open reduction and internal fixation with foot lateral column shortening, forefoot abduction deformity, osteoarthritis, lateral foot pain; fracture healed at 8-12 weeks in 58 feet undergoing open reduction and internal fixation, without osteoarthritis, cuboid bone shortening, and pain at cuboid bone; and AOFAS score was 75-97(mean,93.5) at last follow-up. ConclusionThe classification criteria of cuboid fracture proposed based on CT examination is feasible and has guiding significance to the choice of treatment method.
Hepatocellular carcinoma (HCC) is the most common liver malignancy, where HCC segmentation and prediction of the degree of pathological differentiation are two important tasks in surgical treatment and prognosis evaluation. Existing methods usually solve these two problems independently without considering the correlation of the two tasks. In this paper, we propose a multi-task learning model that aims to accomplish the segmentation task and classification task simultaneously. The model consists of a segmentation subnet and a classification subnet. A multi-scale feature fusion method is proposed in the classification subnet to improve the classification accuracy, and a boundary-aware attention is designed in the segmentation subnet to solve the problem of tumor over-segmentation. A dynamic weighted average multi-task loss is used to make the model achieve optimal performance in both tasks simultaneously. The experimental results of this method on 295 HCC patients are superior to other multi-task learning methods, with a Dice similarity coefficient (Dice) of (83.9 ± 0.88)% on the segmentation task, while the average recall is (86.08 ± 0.83)% and an F1 score is (80.05 ± 1.7)% on the classification task. The results show that the multi-task learning method proposed in this paper can perform the classification task and segmentation task well at the same time, which can provide theoretical reference for clinical diagnosis and treatment of HCC patients.
Objective To recognize the different phases of Korotkoff sounds through deep learning technology, so as to improve the accuracy of blood pressure measurement in different populations. Methods A classification model of the Korotkoff sounds phases was designed, which fused attention mechanism (Attention), residual network (ResNet) and bidirectional long short-term memory (BiLSTM). First, a single Korotkoff sound signal was extracted from the whole Korotkoff sounds signals beat by beat, and each Korotkoff sound signal was converted into a Mel spectrogram. Then, the local feature extraction of Mel spectrogram was processed by using the Attention mechanism and ResNet network, and BiLSTM network was used to deal with the temporal relations between features, and full-connection layer network was applied in reducing the dimension of features. Finally, the classification was completed by SoftMax function. The dataset used in this study was collected from 44 volunteers (24 females, 20 males with an average age of 36 years), and the model performance was verified using 10-fold cross-validation. Results The classification accuracy of the established model for the 5 types of Korotkoff sounds phases was 93.4%, which was higher than that of other models. Conclusion This study proves that the deep learning method can accurately classify Korotkoff sounds phases, which lays a strong technical foundation for the subsequent design of automatic blood pressure measurement methods based on the classification of the Korotkoff sounds phases.
Speech expression is an important high-level cognitive behavior of human beings. The realization of this behavior is closely related to human brain activity. Both true speech expression and speech imagination can activate part of the same brain area. Therefore, speech imagery becomes a new paradigm of brain-computer interaction. Brain-computer interface (BCI) based on speech imagery has the advantages of spontaneous generation, no training, and friendliness to subjects, so it has attracted the attention of many scholars. However, this interactive technology is not mature in the design of experimental paradigms and the choice of imagination materials, and there are many issues that need to be discussed urgently. Therefore, in response to these problems, this article first expounds the neural mechanism of speech imagery. Then, by reviewing the previous BCI research of speech imagery, the mainstream methods and core technologies of experimental paradigm, imagination materials, data processing and so on are systematically analyzed. Finally, the key problems and main challenges that restrict the development of this type of BCI are discussed. And the future development and application perspective of the speech imaginary BCI system are prospected.
Objective To investigate new classification and repair methods for the traverse amputated fingertip. Methods From March 2000 to October 2006, 20 cases of 20 fingers with traverse amputated fingertip, including 13 males and 7 females aged 17-47 years, were treated. Twenty patients (9 crush injuries, 5 cutting injuries and 6 sawing injuries) were classified into 4 types, namely type I (the distal one third of nail bed), type II (the middle of nail bed), type III (the poximal one third of nail bed), and type IV (the root of nail bed). There were 3 patients (2 index fingers and 1 l ittle finger) of type I, 8 patients (2 thumbs, 3 index fingers and 3 middle fingers) of type II, 5 patients (3 index fingers, 1 ring finger and 1 l ittle finger)of type III, and 4 patients (2 thumbs, 1 middle finger and 1 l ittle finger) of type IV. The soft tissue defect ranged from 1.2 cm × 1.2 cm to 1.5 cm × 1.2 cm. The time from injury to surgery was 3-10 hours. Fingers of type I and type II were treated with forward flow axial flap and modified nail bed lengthening. Fingers of type III and type IV were treated with forward flow axial flap and partial nail bed replantation as well as modified nail bed lengthening. The flaps ranged in size from 1.5 cm × 1.2 cm to 2.0 cm × 1.4 cm. Results Twenty patients incisions healed by first intention and the flaps, nails and skin grafting survived. All donor sites healed by first intention. All patients were followed up for 2-6 months (4 months on average). The appearances of fingertips were good. The texture of the flap was soft, and the fingers had no tenderness and motor disturbance. The two-point discrimination was 4.5-6.5 mm．The finger nails of type I and type II extended 3-4 mm after operation, while the finger nails of type III and type IV extended 8-10 mm after operation. All finger nails were smooth and flat without pain. Hook nail happened in 1 case 6 months after operation. Conclusion Classification of the injured fingers according to the condition of the amputation base is helpful in choosing repair methods, and is conducive to maximize the recovery of the function and shape of fingertips.
Post-traumatic stress disorder (PTSD) presents with complex and diverse clinical manifestations, making accurate and objective diagnosis challenging when relying solely on clinical assessments. Therefore, there is an urgent need to develop reliable and objective auxiliary diagnostic models to provide effective diagnosis for PTSD patients. Currently, the application of graph neural networks for representing PTSD is limited by the expressiveness of existing models, which does not yield optimal classification results. To address this, we proposed a multi-graph multi-kernel graph convolutional network (MK-GCN) model for classifying PTSD data. First, we constructed functional connectivity matrices at different scales for the same subjects using different atlases, followed by employing the k-nearest neighbors algorithm to build the graphs. Second, we introduced the MK-GCN methodology to enhance the feature extraction capability of brain structures at different scales for the same subjects. Finally, we classified the extracted features from multiple scales and utilized graph class activation mapping to identify the top 10 brain regions contributing to classification. Experimental results on seismic-induced PTSD data demonstrated that our model achieved an accuracy of 84.75%, a specificity of 84.02%, and an AUC of 85% in the classification task distinguishing between PTSD patients and non-affected subjects. The findings provide robust evidence for the auxiliary diagnosis of PTSD following earthquakes and hold promise for reliably identifying specific brain regions in other PTSD diagnostic contexts, offering valuable references for clinicians.