Tuberculosis is one of the major infectious diseases that seriously endanger human health. Since 2014, it has surpassed human immunodeficiency virus/acquired immunodeficiency syndrome as the first infectious disease in patients with single pathogens. China is the third-largest country in the world in terms of high burden of tuberculosis. In 2016, there were about 900 000 new cases of tuberculosis in China. China is facing a severe tuberculosis epidemic, especially for the early diagnosis of tuberculosis and misdiagnosis of tuberculosis, which leads to delay in treatment and the spread of tuberculosis. With the application of artificial intelligence in the medical field, machine learning and deep learning methods have shown important value in the diagnosis of tuberculosis. This article will explain the application status and future development of machine learning and deep learning in the diagnosis of tuberculosis.
ObjectivesTo explore the effect of the deep learning algorithm convolutional neural network (CNN) in screening of randomized controlled trials (RCTs) in Chinese medical literatures.MethodsLiterature with the topic " oral science” published in 2014 were retrieved from CNKI and exported citations containing title and abstract. RCTs screening was conducted by double independent screening, checking and peer discussion. The final results of the citations were used for CNN algorithm model training. After completing the algorithm model training, a prospective comparative trial was organized by searching all literature with the topic "oral science" published in CNKI from January to March 2018 to compare the sensitivity (SEN) and specificity (SPE) of algorithm with manual screening. The initial results of a single screener represented the performance of manual screening, and the final results after peer discussion were used as the gold standard. The best thresholds of algorithm were determined with the receptor operative characteristic (ROC) curve.ResultsA total of 1 246 RCTs and 4 754 non-RCTs were eventually included for training and testing of CNN algorithm model. 249 RCTs and 949 non-RCTs were included in the prospective trial. The SEN and SPE of manual screening were 98.01% and 98.82%. For the algorithm model, the SEN of RCTs screening decreased with the increase of threshold value while the SPE increased with the increase of threshold value. After 27 changes of threshold value, ROC curve were obtained. The area under the ROC curve was 0.9977, unveiling the optimal accuracy threshold (Threshold=0.4, SEN=98.39%, SPE=98.84%) and high sensitivity threshold (Threshold=0.06, SEN=99.60%, SPE=94.10%).ConclusionsA CNN algorithm model is trained with Chinese RCTs classification database established in this study and shows an excellent classification performance in screening RCTs of Chinese medical literature, which is proved to be comparable to the manual screening performance in the prospective controlled trial.
ObjectiveTo observe and analyze the accuracy of the optic disc positioning and segmentation method of fundus images based on deep learning.MethodsThe model training strategies were training and evaluating deep learning-based optic disc positioning and segmentation methods on the ORIGA dataset. A deep convolutional neural network (CNN) was built on the Caffe framework of deep learning. A sliding window was used to cut the original image of the ORIGA data set into many small pieces of pictures, and the deep CNN was used to determine whether each small piece of picture contained the complete disc structure, so as to find the area of the disc. In order to avoid the influence of blood vessels on the segmentation of the optic disc, the blood vessels in the optic disc area were removed before segmentation of the optic disc boundary. A deep network of optic disc segmentation based on image pixel classification was used to realize the segmentation of the optic disc of fundus images. The accuracy of the optic disc positioning and segmentation method was calculated based on deep learning of fundus images. Positioning accuracy=T/N, T represented the number of fundus images with correct optic disc positioning, and N represented the total number of fundus images used for positioning. The overlap error was used to compare the difference between the segmentation result of the optic disc and the actual boundary of the optic disc.ResultsOn the dataset from ORIGA, the accuracy of the optic disc localization can reach 99.6%, the average overlap error of optic disc segmentation was 7.1%. The calculation errors of the average cup-to-disk ratio for glaucoma images and normal images were 0.066 and 0.049, respectively. Disc segmentation of each image took an average of 10 ms.ConclusionThe algorithm can locate the disc area quickly and accurately, and can also segment the disc boundary more accurately.
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 build a small-sample ultra-widefield fundus images (UWFI) multi-disease classification artificial intelligence model, and initially explore the ability of artificial intelligence to classify UWFI multi-disease tasks. MethodsA retrospective study. From 2016 to 2021, 1 608 images from 1 123 patients who attended the Eye Center of the Renmin Hospital of Wuhan University and underwent UWFI examination were used for UWFI multi-disease classification artificial intelligence model construction. Among them, 320, 330, 319, 268, and 371 images were used for diabetic retinopathy (DR), retinal vein occlusion (RVO), pathological myopia (PM), retinal detachment (RD), and normal fundus images, respectively. 135 images from 106 patients at the Tianjin Medical University Eye Hospital were used as the external test set. EfficientNet-B7 was selected as the backbone network for classification analysis of the included UWFI images. The performance of the UWFI multi-task classification model was assessed using the receiver operating characteristic curve, area under the curve (AUC), sensitivity, specificity, and accuracy. All data were expressed using numerical values and 95% confidence intervals (CI). The datasets were trained on the network models ResNet50 and ResNet101 and tested on an external test set to compare and observe the performance of EfficientNet with the 2 models mentioned above. ResultsThe overall classification accuracy of the UWFI multi-disease classification artificial intelligence model on the internal and external test sets was 92.57% (95%CI 91.13%-92.92%) and 88.89% (95%CI 88.11%-90.02%), respectively. These were 96.62% and 92.59% for normal fundus, 95.95% and 95.56% for DR, 96.62% and 98.52% for RVO, 98.65% and 97.04% for PM, and 97.30% and 94.07% for RD, respectively. The mean AUC on the internal and external test sets was 0.993 and 0.983, respectively, with 0.994 and 0.939 for normal fundus, 0.999 and 0.995 for DR, 0.985 and 1.000 for RVO, 0.991 and 0.993 for PM and 0.995 and 0.990 for RD, respectively. EfficientNet performed better than the ResNet50 and ResNet101 models on both the internal and external test sets. ConclusionThe preliminary UWFI multi-disease classification artificial intelligence model using small samples constructed in this study is able to achieve a high accuracy rate, and the model may have some value in assisting clinical screening and diagnosis.
ObjectiveTo study a deep learning-based dual-modality fundus camera which was used to study retinal blood oxygen saturation and vascular morphology changes in eyes with branch retinal vein occlusion (BRVO). MethodsA prospective study. From May to October 2020, 31 patients (31 eyes) of BRVO (BRVO group) and 20 healthy volunteers (20 eyes) with matched gender and age (control group) were included in the study. Among 31 patients (31 eyes) in BRVO group, 20 patients (20 eyes) received one intravitreal injection of anti-vascular endothelial growth factor drugs before, and 11 patients (11 eyes) did not receive any treatment. They were divided into treatment group and untreated group accordingly. Retinal images were collected with a dual-modality fundus camera; arterial and vein segments were segmented in the macular region of interest (MROI) using deep learning; the optical density ratio was used to calculate retinal blood oxygen saturation (SO2) on the affected and non-involved sides of the eyes in the control group and patients in the BRVO group, and calculated the diameter, curvature, fractal dimension and density of arteriovenous in MROI. Quantitative data were compared between groups using one-way analysis of variance. ResultsThere was a statistically significant difference in arterial SO2 (SO2-A) in the MROI between the affected eyes, the fellow eyes in the BRVO group and the control group (F=4.925, P<0.001), but there was no difference in the venous SO2 (SO2-V) (F=0.607, P=0.178). Compared with the control group, the SO2-A in the MROI of the affected side and the non-involved side of the untreated group was increased, and the difference was statistically significant (F=4.925, P=0.012); there was no significant difference in SO2-V (F=0.607, P=0.550). There was no significant difference in SO2-A and SO2-V in the MROI between the affected side, the non-involved side in the treatment group and the control group (F=0.159, 1.701; P=0.854, 0.197). There was no significant difference in SO2-A and SO2-V in MROI between the affected side of the treatment group, the untreated group and the control group (F=2.553, 0.265; P=0.088, 0.546). The ophthalmic artery diameter, arterial curvature, arterial fractal dimension, vein fractal dimension, arterial density, and vein density were compared in the untreated group, the treatment group, and the control group, and the differences were statistically significant (F=3.527, 3.322, 7.251, 26.128, 4.782, 5.612; P=0.047, 0.044, 0.002, <0.001, 0.013, 0.006); there was no significant difference in vein diameter and vein curvature (F=2.132, 1.199; P=0.143, 0.321). ConclusionArterial SO2 in BRVO patients is higher than that in healthy eyes, it decreases after anti-anti-vascular endothelial growth factor drugs treatment, SO2-V is unchanged.
Steady-state visual evoked potential (SSVEP) is one of the commonly used control signals in brain-computer interface (BCI) systems. The SSVEP-based BCI has the advantages of high information transmission rate and short training time, which has become an important branch of BCI research field. In this review paper, the main progress on frequency recognition algorithm for SSVEP in past five years are summarized from three aspects, i.e., unsupervised learning algorithms, supervised learning algorithms and deep learning algorithms. Finally, some frontier topics and potential directions are explored.
In recent years, epileptic seizure detection based on electroencephalogram (EEG) has attracted the widespread attention of the academic. However, it is difficult to collect data from epileptic seizure, and it is easy to cause over fitting phenomenon under the condition of few training data. In order to solve this problem, this paper took the CHB-MIT epilepsy EEG dataset from Boston Children's Hospital as the research object, and applied wavelet transform for data augmentation by setting different wavelet transform scale factors. In addition, by combining deep learning, ensemble learning, transfer learning and other methods, an epilepsy detection method with high accuracy for specific epilepsy patients was proposed under the condition of insufficient learning samples. In test, the wavelet transform scale factors 2, 4 and 8 were set for experimental comparison and verification. When the wavelet scale factor was 8, the average accuracy, average sensitivity and average specificity was 95.47%, 93.89% and 96.48%, respectively. Through comparative experiments with recent relevant literatures, the advantages of the proposed method were verified. Our results might provide reference for the clinical application of epilepsy detection.
The diagnosis of hypertrophic cardiomyopathy (HCM) is of great significance for the early risk classification of sudden cardiac death and the screening of family genetic diseases. This research proposed a HCM automatic detection method based on convolution neural network (CNN) model, using single-lead electrocardiogram (ECG) signal as the research object. Firstly, the R-wave peak locations of single-lead ECG signal were determined, followed by the ECG signal segmentation and resample in units of heart beats, then a CNN model was built to automatically extract the deep features in the ECG signal and perform automatic classification and HCM detection. The experimental data is derived from 108 ECG records extracted from three public databases provided by PhysioNet, the database established in this research consists of 14,459 heartbeats, and each heartbeat contains 128 sampling points. The results revealed that the optimized CNN model could effectively detect HCM, the accuracy, sensitivity and specificity were 95.98%, 98.03% and 95.79% respectively. In this research, the deep learning method was introduced for the analysis of single-lead ECG of HCM patients, which could not only overcome the technical limitations of conventional detection methods based on multi-lead ECG, but also has important application value for assisting doctor in fast and convenient large-scale HCM preliminary screening.
Lung cancer is the most threatening tumor disease to human health. Early detection is crucial to improve the survival rate and recovery rate of lung cancer patients. Existing methods use the two-dimensional multi-view framework to learn lung nodules features and simply integrate multi-view features to achieve the classification of benign and malignant lung nodules. However, these methods suffer from the problems of not capturing the spatial features effectively and ignoring the variability of multi-views. Therefore, this paper proposes a three-dimensional (3D) multi-view convolutional neural network (MVCNN) framework. To further solve the problem of different views in the multi-view model, a 3D multi-view squeeze-and-excitation convolution neural network (MVSECNN) model is constructed by introducing the squeeze-and-excitation (SE) module in the feature fusion stage. Finally, statistical methods are used to analyze model predictions and doctor annotations. In the independent test set, the classification accuracy and sensitivity of the model were 96.04% and 98.59% respectively, which were higher than other state-of-the-art methods. The consistency score between the predictions of the model and the pathological diagnosis results was 0.948, which is significantly higher than that between the doctor annotations and the pathological diagnosis results. The methods presented in this paper can effectively learn the spatial heterogeneity of lung nodules and solve the problem of multi-view differences. At the same time, the classification of benign and malignant lung nodules can be achieved, which is of great significance for assisting doctors in clinical diagnosis.