Developing and verifying automatic detection of active pulmonary tuberculosis from multi-slice spiral CT images based on deep learning

Deep learning on longitudinal CT scans: automated prediction of treatment outcomes in hospitalized tuberculosis patients

Three deep learning (DL)-based prediction models (PMs) using longitudinal CT images were developed to predict tuberculosis (TB) treatment outcomes. The internal dataset consists of 493 bacteriologically confirmed TB patients who completed the anti-tuberculosis treatment with three-time CT scans, including a pretreatment CT scan and two follow-up CT scans. PM1 was trained using only pretreatment CT scans, and PM2 and PM3 were developed by adding follow-up scans. An independent testing was performed on external dataset comprising 86 TB patients. The area under the curve for classifying success and drug-resistant (DR)-TB was improved on both internal (0.609 vs. 0.625 vs. 0.815) and external (0.627 vs. 0.705 vs. 0.735) dataset by adding follow-up scans. The accuracy and F1-score also showed an increasing tendency in the external test. Regular follow-up CT scans can aid in the treatment prediction, and special attention should be given to early intensive phase of treatment to identify high-risk DR-TB patients.

A prospective multicenter clinical research study validating the effectiveness and safety of a chest X-ray-based pulmonary tuberculosis screening software JF CXR-1 built on a convolutional neural network algorithm

Background

Chest radiography (chest X-ray or CXR) plays an important role in the early detection of active pulmonary tuberculosis (TB). In areas with a high TB burden that require urgent screening, there is often a shortage of radiologists available to interpret the X-ray results. Computer-aided detection (CAD) software employed with artificial intelligence (AI) systems may have the potential to solve this problem.

Objective

We validated the effectiveness and safety of pulmonary tuberculosis imaging screening software that is based on a convolutional neural network algorithm.

Methods

We conducted prospective multicenter clinical research to validate the performance of pulmonary tuberculosis imaging screening software (JF CXR-1). Volunteers under the age of 15 years, both with or without suspicion of pulmonary tuberculosis, were recruited for CXR photography. The software reported a probability score of TB for each participant. The results were compared with those reported by radiologists. We measured sensitivity, specificity, consistency rate, and the area under the receiver operating characteristic curves (AUC) for the diagnosis of tuberculosis. Besides, adverse events (AE) and severe adverse events (SAE) were also evaluated.

Results

The clinical research was conducted in six general infectious disease hospitals across China. A total of 1,165 participants were enrolled, and 1,161 were enrolled in the full analysis set (FAS). Men accounted for 60.0% (697/1,161). Compared to the results from radiologists on the board, the software showed a sensitivity of 94.2% (95% CI: 92.0-95.8%) and a specificity of 91.2% (95% CI: 88.5-93.2%). The consistency rate was 92.7% (91.1-94.1%), with a Kappa value of 0.854 (P = 0.000). The AUC was 0.98. In the safety set (SS), which consisted of 1,161 participants, 0.3% (3/1,161) had AEs that were not related to the software, and no severe AEs were observed.

Conclusion

The software for tuberculosis screening based on a convolutional neural network algorithm is effective and safe. It is a potential candidate for solving tuberculosis screening problems in areas lacking radiologists with a high TB burden.

Deep learning based CT images automatic analysis model for active/non-active pulmonary tuberculosis differential diagnosis

Active pulmonary tuberculosis (ATB), which is more infectious and has a higher mortality rate compared with non-active pulmonary tuberculosis (non-ATB), needs to be diagnosed accurately and timely to prevent the tuberculosis from spreading and causing deaths. However, traditional differential diagnosis methods of active pulmonary tuberculosis involve bacteriological testing, sputum culturing and radiological images reading, which is time consuming and labour intensive. Therefore, an artificial intelligence model for ATB differential diagnosis would offer great assistance in clinical practice. In this study, computer tomography (CT) scans images and corresponding clinical information of 1160 ATB patients and 1131 patients with non-ATB were collected and divided into training, validation, and testing sets. A 3-dimension (3D) Nested UNet model was utilized to delineate lung field regions in the CT images, and three different pre-trained deep learning models including 3D VGG-16, 3D EfficientNet and 3D ResNet-50 were used for classification and differential diagnosis task. We also collected an external testing set with 100 ATB cases and 100 Non-ATB cases for further validation of the model. In the internal and external testing set, the 3D ResNet-50 model outperformed other models, reaching an AUC of 0.961 and 0.946, respectively. The 3D ResNet-50 model reached even higher levels of diagnostic accuracy than experienced radiologists, while the CT images reading and diagnosing speed was 10 times faster than human experts. The model was also capable of visualizing clinician interpretable lung lesion regions important for differential diagnosis, making it a powerful tool assisting ATB diagnosis. In conclusion, we developed an auxiliary tool to differentiate active and non-active pulmonary tuberculosis, which would have broad prospects in the bedside.

Intelligent tuberculosis activity assessment system based on an ensemble of neural networks

This article proposes a novel approach to assess the degree of activity of pulmonary tuberculosis by active tuberculoma foci. It includes the development of a new method for processing lung CT images using an ensemble of deep convolutional neural networks using such special algorithms: an optimized algorithm for preliminary segmentation and selection of informative scans, a new algorithm for refining segmented masks to improve the final accuracy, an efficient fuzzy inference system for more weighted activity assessment. The approach also includes the use of medical classification of disease activity based on densitometric measures of tuberculomas. The selection and markup of the training sample images were performed manually by qualified pulmonologists from a base of approximately 9,000 CT lung scans of patients who had been enrolled in the dispensary for 15 years. The first basic step of the proposed approach is the developed algorithm for preprocessing CT lung scans. It consists in segmentation of intrapulmonary regions, which contain vessels, bronchi, lung walls to detect complex cases of ingrown tuberculomas. To minimize computational cost, the proposed approach includes a new method for selecting informative lung scans, i.e., those that potentially contain tuberculomas. The main processing step is binary segmentation of tuberculomas, which is proposed to be performed optimally by a certain ensemble of neural networks. Optimization of the ensemble size and its composition is achieved by using an algorithm for calculating individual contributions. A modification of this algorithm using new effective heuristic metrics has been proposed which improves the performance of the algorithm for this problem. A special algorithm was developed for post-processing of tuberculoma masks obtained during the segmentation step. The goal of this step is to refine the calculated mask for the physical placement of the tuberculoma. The algorithm consists in cleaning the mask from noisy formations on the scan, as well as expanding the mask area to maximize the capture of the tuberculoma location area. A simplified fuzzy inference system was developed to provide a more accurate final calculation of the degree of disease activity, which reflects data from current medical studies. The accuracy of the system was also tested on a test sample of independent patients, showing more than 96% correct calculations of disease activity, confirming the effectiveness and feasibility of introducing the system into clinical practice.

Lightweight YOLOv4 with Multiple Receptive Fields for Detection of Pulmonary Tuberculosis

The characteristics of pulmonary tuberculosis are complex, and the cost of manual screening is high. The detection model based on convolutional neural network is an essential method for assisted diagnosis with artificial intelligence. However, it also has the disadvantages of complex structure and a large number of parameters, and the detection accuracy needs to be further improved. Therefore, an improved lightweight YOLOv4 pulmonary tuberculosis detection model named MIP-MY is proposed. Firstly, over 300 actual cases are selected to make a common dataset by professional physicians, which is used to evaluate the performance of the model. Subsequently, by introducing the inverted residual channel attention and the pyramid pooling module, a new structure of MIP is created and used as the backbone extractor of MIP-MY, which could further decrease the number of parameters and fuse context information. Then the multiple receptive field module is added after the three effective feature layers of the backbone extractor, which effectively enhances the information extraction ability of the deep feature layer and reduces the miss detection rate of small pulmonary tuberculosis lesions. Finally, the pulmonary tuberculosis detection model MIP-MY with lightweight and multiple receptive field characteristics is constructed by combining each improved modules with multiscale structure. Compared to the original YOLOv4, the model parameters of MIP-MY is reduced by 47%, while the mAP value is raised to 95.32% and the miss detection rate is decreased to 6%. It is verified that the model can effectively assist radiologists in the diagnosis of pulmonary tuberculosis.

Automatic detection of breast cancer in ultrasound images using Mayfly algorithm optimized handcrafted features

BACKGROUND

The incidence rates of breast cancer in women community is progressively raising and the premature diagnosis is necessary to detect and cure the disease.

OBJECTIVE

To develop a novel automated disuse detection framework to examine the Breast-Ultrasound-Images (BUI).

METHODS

This scheme includes the following stages; (i) Image acquisition and resizing, (ii) Gaussian filter-based pre-processing, (iii) Handcrafted features extraction, (iv) Optimal feature selection with Mayfly Algorithm (MA), (v) Binary classification and validation. The dataset includes BUI extracted from 133 normal, 445 benign and 210 malignant cases. Each BUI is resized to 256×256×1 pixels and the resized BUIs are used to develop and test the new scheme. Handcrafted feature-based cancer detection is employed and the parameters, such as Entropies, Local-Binary-Pattern (LBP) and Hu moments are considered. To avoid the over-fitting problem, a feature reduction procedure is also implemented with MA and the reduced feature sub-set is used to train and validate the classifiers developed in this research.

RESULTS

The experiments were performed to classify BUIs between (i) normal and benign, (ii) normal and malignant, and (iii) benign and malignant cases. The results show that classification accuracy of > 94%, precision of > 92%, sensitivity of > 92% and specificity of > 90% are achieved applying the developed new schemes or framework.

CONCLUSION

In this work, a machine-learning scheme is employed to detect/classify the disease using BUI and achieves promising results. In future, we will test the feasibility of implementing deep-learning method to this framework to further improve detection accuracy.

U-Net: A valuable encoder-decoder architecture for liver tumors segmentation in CT images

This study proposes a new predictive segmentation method for liver tumors detection using computed tomography (CT) liver images. In the medical imaging field, the exact localization of metastasis lesions after acquisition faces persistent problems both for diagnostic aid and treatment effectiveness. Therefore, the improvement in the diagnostic process is substantially crucial in order to increase the success chance of the management and the therapeutic follow-up. The proposed procedure highlights a computerized approach based on an encoder-decoder structure in order to provide volumetric analysis of pathologic tumors. Specifically, we developed an automatic algorithm for the liver tumors defect segmentation through the Seg-Net and U-Net architectures from metastasis CT images. In this study, we collected a dataset of 200 pathologically confirmed metastasis cancer cases. A total of 8,297 CT image slices of these cases were used developing and optimizing the proposed segmentation architecture. The model was trained and validated using 170 and 30 cases or 85% and 15% of the CT image data, respectively. Study results demonstrate the strength of the proposed approach that reveals the superlative segmentation performance as evaluated using following indices including F1-score = 0.9573, Recall = 0.9520, IOU = 0.9654, Binary cross entropy = 0.0032 and p-value <0.05, respectively. In comparison to state-of-the-art techniques, the proposed method yields a higher precision rate by specifying metastasis tumor position.

Spiral Computed Tomography in the Quantitative Measurement of the Adjacent Structure of the Left Atrial Appendage in Patients with Atrial Fibrillation

Cardiac arrhythmias are common clinical cardiovascular diseases. Arrhythmias are abnormalities in the frequency, rhythm, site of origin, conduction velocity, or sequence of excitation of the cardiac impulses. Arrhythmia mechanisms include foldback, altered autonomic rhythm, and triggering mechanisms. It can cause palpitations, dizziness, black dawn, syncope, and angina pectoris and can worsen a preexisting cardiac disease, reduce the quality of life, and increase mortality. Also, by making it one of the constant challenges for the clinical cardiovascular physician, we can get more information. The study included 94 patients with atrial fibers, including 56 men and 38 women aged 57, 46, 11, and 68 years. There are 80 patients with nonatrial fibers, including 44 men and 36 women aged 56, 10, and 83 years. Those who can perform a normal coronary angiography and exclude congenital heart disease, heart valve disease, and other cardiovascular diseases. In both groups, a 256-layer spiral CT examination was performed. A pulmonary vein scanning protocol was applied to the patients with atrial fibrillation, and this can perform normal coronary angiography and exclude those with cardiovascular diseases such as congenital heart disease and valvular heart disease. The purpose of this study is to investigate the anatomical changes of the left atrium and its adjacent structures by applying the 256 nm spiral CT imaging to visualize the left atrium and its adjacent structures and by applying the MPR technology, VR technology, and simulation endoscope techniques.

Application of the Artificial Neural Network in the Diagnosis of Infantile Bronchial Bridge with 64-Slice Multislice Spiral CT

The objective is to study the application of spiral CT in the diagnosis of the trachea in children. In this study, the effect of 64-slice multislice spiral CT in diagnosing infant bronchial bridge was studied based on an artificial neural network. From June 2020 to December 2020, 100 cases of children with the trachea in the outpatient department of our hospital were selected as the research object. They were divided into the study group and the control group, with 50 cases in each group. The results showed that among the 50 cases in the control group, 42 cases were found to have a bronchial foreign body and 8 cases were missed; the detection rate was 84%. There were 7 cases of trachea foreign body, 15 cases of left bronchial foreign body, 14 cases of right bronchial foreign body, and 6 cases of medium bronchial foreign body. The detection rate of the study group was significantly higher than that of the control group, with a statistical significance (P < 0.05). Conclusion. The detection rate of neural networks in MSCT is higher than that of X-ray, and the MSCT based on the artificial neural network can clearly show the morphology, position, and the relationship between the foreign body and trachea, which is worthy of clinical promotion and application.

The Clinical Value of Multislice Spiral Computed Tomography in the Diagnosis of Upper Digestive Tract Diseases

Imaging methods for gastrointestinal diseases were based on X-ray imaging until the 1970s, but the development of fiberoptic endoscopy in the 1980s has replaced X-ray imaging. Endoscopy can directly observe the location, size, scope, and color of lesions and obtain pathological results through biopsy, while ligation and other treatments can be performed on polyps and other lesions. Studies have shown that multilayer spiral computed tomography (CT) examination after standardized gastrointestinal preparation and full use of the advantages of various 3D postprocessing reconstruction techniques are of great clinical value in the detection of gastrointestinal diseases, determination of the nature of lesions, localization of lesions, and staging of gastrointestinal malignancies and can make up for the shortcomings of fiberoptic endoscopy, and various 3D postprocessing reconstruction modes have their own advantages and disadvantages. Among them, conventional CT cross-sectional images are the basic images for the diagnosis of various gastric testicular lesions. Axial images, especially thin-layer axial images, can detect the absolute majority of lesions, but there are limitations in observing the anatomical position of lesions, invasion of surrounding tissues, lymph node metastasis, vascularity, and determination of the stage of malignant tumors.

Deep learning assistance for tuberculosis diagnosis with chest radiography in low-resource settings

Tuberculosis (TB) is a major health issue with high mortality rates worldwide. Recently, tremendous researches of artificial intelligence (AI) have been conducted targeting at TB to reduce the diagnostic burden. However, most researches are conducted in the developed urban areas. The feasibility of applying AI in low-resource settings remains unexplored. In this study, we apply an automated detection (AI) system to screen a large population in an underdeveloped area and evaluate feasibility and contribution of applying AI to help local radiologists detect and diagnose TB using chest X-ray (CXR) images. First, we divide image data into one training dataset including 2627 TB-positive cases and 7375 TB-negative cases and one testing dataset containing 276 TB-positive cases and 619 TB-negative cases, respectively. Next, in building AI system, the experiment includes image labeling and preprocessing, model training and testing. A segmentation model named TB-UNet is also built to detect diseased regions, which uses ResNeXt as the encoder of U-Net. We use AI-generated confidence score to predict the likelihood of each testing case being TB-positive. Then, we conduct two experiments to compare results between the AI system and radiologists with and without AI assistance. Study results show that AI system yields TB detection accuracy of 85%, which is much higher than detection accuracy of radiologists (62%) without AI assistance. In addition, with AI assistance, the TB diagnostic sensitivity of local radiologists is improved by 11.8%. Therefore, this study demonstrates that AI has great potential to help detection, prevention, and control of TB in low-resource settings, particularly in areas with more scant doctors and higher rates of the infected population.