Chest Computed Tomography Volumes Research Articles

Deep learning for precise diagnosis and subtype triage of drug-resistant tuberculosis on chest computed tomography.

Deep learning, transforming input data into target prediction through intricate network structures, has inspired novel exploration in automated diagnosis based on medical images. The distinct morphological characteristics of chest abnormalities between drug-resistant tuberculosis (DR-TB) and drug-sensitive tuberculosis (DS-TB) on chest computed tomography (CT) are of potential value in differential diagnosis, which is challenging in the clinic. Hence, based on 1176 chest CT volumes from the equal number of patients with tuberculosis (TB), we presented a Deep learning-based system for TB drug resistance identification and subtype classification (DeepTB), which could automatically diagnose DR-TB and classify crucial subtypes, including rifampicin-resistant tuberculosis, multidrug-resistant tuberculosis, and extensively drug-resistant tuberculosis. Moreover, chest lesions were manually annotated to endow the model with robust power to assist radiologists in image interpretation and the Circos revealed the relationship between chest abnormalities and specific types of DR-TB. Finally, DeepTB achieved an area under the curve (AUC) up to 0.930 for thoracic abnormality detection and 0.943 for DR-TB diagnosis. Notably, the system demonstrated instructive value in DR-TB subtype classification with AUCs ranging from 0.880 to 0.928. Meanwhile, class activation maps were generated to express a human-understandable visual concept. Together, showing a prominent performance, DeepTB would be impactful in clinical decision-making for DR-TB.

Dual Multiscale Mean Teacher Network for Semi-Supervised Infection Segmentation in Chest CT Volume for COVID-19.

Automated detecting lung infections from computed tomography (CT) data plays an important role for combating coronavirus 2019 (COVID-19). However, there are still some challenges for developing AI system: 1) most current COVID-19 infection segmentation methods mainly relied on 2-D CT images, which lack 3-D sequential constraint; 2) existing 3-D CT segmentation methods focus on single-scale representations, which do not achieve the multiple level receptive field sizes on 3-D volume; and 3) the emergent breaking out of COVID-19 makes it hard to annotate sufficient CT volumes for training deep model. To address these issues, we first build a multiple dimensional-attention convolutional neural network (MDA-CNN) to aggregate multiscale information along different dimension of input feature maps and impose supervision on multiple predictions from different convolutional neural networks (CNNs) layers. Second, we assign this MDA-CNN as a basic network into a novel dual multiscale mean teacher network (DM [Formula: see text]-Net) for semi-supervised COVID-19 lung infection segmentation on CT volumes by leveraging unlabeled data and exploring the multiscale information. Our DM [Formula: see text]-Net encourages multiple predictions at different CNN layers from the student and teacher networks to be consistent for computing a multiscale consistency loss on unlabeled data, which is then added to the supervised loss on the labeled data from multiple predictions of MDA-CNN. Third, we collect two COVID-19 segmentation datasets to evaluate our method. The experimental results show that our network consistently outperforms the compared state-of-the-art methods.

TRCT-GAN: CT reconstruction from biplane X-rays using transformer and generative adversarial networks

Computed tomography (CT) provides a three-dimensional view of a patient's internal organs. Compared to CT volumes, X-ray imaging can significantly reduce the patient's exposure to ionizing radiation. Moreover, X-ray images are more economical and widely applied in surgical procedures. However, X-ray images can only provide two-dimensional information. In this paper, an end-to-end GAN network, named TRCT-GAN, is proposed to reconstruct chest CT volumes from biplane X-ray images. In the GAN network, the Transformer network module is employed to enhance the feature representation of X-ray images. Moreover, a dynamic attention module is added to exploit some 2D feature maps and 3D feature maps to enhance the contextual association. The experimental results demonstrate that the proposed network can effectively produce high-quality CT reconstructions from X-ray images.

Bilateral adaptive graph convolutional network on CT based Covid-19 diagnosis with uncertainty-aware consensus-assisted multiple instance learning.

Coronavirus disease (COVID-19) has caused a worldwide pandemic, putting millions of people's health and lives in jeopardy. Detecting infected patients early on chest computed tomography (CT) is critical in combating COVID-19. Harnessing uncertainty-aware consensus-assisted multiple instance learning (UC-MIL), we propose to diagnose COVID-19 using a new bilateral adaptive graph-based (BA-GCN) model that can use both 2D and 3D discriminative information in 3D CT volumes with arbitrary number of slices. Given the importance of lung segmentation for this task, we have created the largest manual annotation dataset so far with 7,768 slices from COVID-19 patients, and have used it to train a 2D segmentation model to segment the lungs from individual slices and mask the lungs as the regions of interest for the subsequent analyses. We then used the UC-MIL model to estimate the uncertainty of each prediction and the consensus between multiple predictions on each CT slice to automatically select a fixed number of CT slices with reliable predictions for the subsequent model reasoning. Finally, we adaptively constructed a BA-GCN with vertices from different granularity levels (2D and 3D) to aggregate multi-level features for the final diagnosis with the benefits of the graph convolution network's superiority to tackle cross-granularity relationships. Experimental results on three largest COVID-19 CT datasets demonstrated that our model can produce reliable and accurate COVID-19 predictions using CT volumes with any number of slices, which outperforms existing approaches in terms of learning and generalisation ability. To promote reproducible research, we have made the datasets, including the manual annotations and cleaned CT dataset, as well as the implementation code, available at https://doi.org/10.5281/zenodo.6361963.

Effective deep learning approaches for predicting COVID-19 outcomes from chest computed tomography volumes

The rapid evolution of the novel coronavirus disease (COVID-19) pandemic has resulted in an urgent need for effective clinical tools to reduce transmission and manage severe illness. Numerous teams are quickly developing artificial intelligence approaches to these problems, including using deep learning to predict COVID-19 diagnosis and prognosis from chest computed tomography (CT) imaging data. In this work, we assess the value of aggregated chest CT data for COVID-19 prognosis compared to clinical metadata alone. We develop a novel patient-level algorithm to aggregate the chest CT volume into a 2D representation that can be easily integrated with clinical metadata to distinguish COVID-19 pneumonia from chest CT volumes from healthy participants and participants with other viral pneumonia. Furthermore, we present a multitask model for joint segmentation of different classes of pulmonary lesions present in COVID-19 infected lungs that can outperform individual segmentation models for each task. We directly compare this multitask segmentation approach to combining feature-agnostic volumetric CT classification feature maps with clinical metadata for predicting mortality. We show that the combination of features derived from the chest CT volumes improve the AUC performance to 0.80 from the 0.52 obtained by using patients’ clinical data alone. These approaches enable the automated extraction of clinically relevant features from chest CT volumes for risk stratification of COVID-19 patients.

Automated deep learning-based segmentation of COVID-19 lesions from chest computed tomography images.

PurposeThe novel coronavirus COVID-19, which spread globally in late December 2019, is a global health crisis. Chest computed tomography (CT) has played a pivotal role in providing useful information for clinicians to detect COVID-19. However, segmenting COVID-19-infected regions from chest CT results is challenging. Therefore, it is desirable to develop an efficient tool for automated segmentation of COVID-19 lesions using chest CT. Hence, we aimed to propose 2D deep-learning algorithms to automatically segment COVID-19-infected regions from chest CT slices and evaluate their performance.Material and methodsHerein, 3 known deep learning networks: U-Net, U-Net++, and Res-Unet, were trained from scratch for automated segmenting of COVID-19 lesions using chest CT images. The dataset consists of 20 labelled COVID-19 chest CT volumes. A total of 2112 images were used. The dataset was split into 80% for training and validation and 20% for testing the proposed models. Segmentation performance was assessed using Dice similarity coefficient, average symmetric surface distance (ASSD), mean absolute error (MAE), sensitivity, specificity, and precision.ResultsAll proposed models achieved good performance for COVID-19 lesion segmentation. Compared with Res-Unet, the U-Net and U-Net++ models provided better results, with a mean Dice value of 85.0%. Compared with all models, U-Net gained the highest segmentation performance, with 86.0% sensitivity and 2.22 mm ASSD. The U-Net model obtained 1%, 2%, and 0.66 mm improvement over the Res-Unet model in the Dice, sensitivity, and ASSD, respectively. Compared with Res-Unet, U-Net++ achieved 1%, 2%, 0.1 mm, and 0.23 mm improvement in the Dice, sensitivity, ASSD, and MAE, respectively.ConclusionsOur data indicated that the proposed models achieve an average Dice value greater than 84.0%. Two-dimensional deep learning models were able to accurately segment COVID-19 lesions from chest CT images, assisting the radiologists in faster screening and quantification of the lesion regions for further treatment. Nevertheless, further studies will be required to evaluate the clinical performance and robustness of the proposed models for COVID-19 semantic segmentation.

Automatic Sequence-Based Network for Lung Diseases Detection in Chest CT.

ObjectiveTo develop an accurate and rapid computed tomography (CT)-based interpretable AI system for the diagnosis of lung diseases.BackgroundMost existing AI systems only focus on viral pneumonia (e.g., COVID-19), specifically, ignoring other similar lung diseases: e.g., bacterial pneumonia (BP), which should also be detected during CT screening. In this paper, we propose a unified sequence-based pneumonia classification network, called SLP-Net, which utilizes consecutiveness information for the differential diagnosis of viral pneumonia (VP), BP, and normal control cases from chest CT volumes.MethodsConsidering consecutive images of a CT volume as a time sequence input, compared with previous 2D slice-based or 3D volume-based methods, our SLP-Net can effectively use the spatial information and does not need a large amount of training data to avoid overfitting. Specifically, sequential convolutional neural networks (CNNs) with multi-scale receptive fields are first utilized to extract a set of higher-level representations, which are then fed into a convolutional long short-term memory (ConvLSTM) module to construct axial dimensional feature maps. A novel adaptive-weighted cross-entropy loss (ACE) is introduced to optimize the output of the SLP-Net with a view to ensuring that as many valid features from the previous images as possible are encoded into the later CT image. In addition, we employ sequence attention maps for auxiliary classification to enhance the confidence level of the results and produce a case-level prediction.ResultsFor evaluation, we constructed a dataset of 258 chest CT volumes with 153 VP, 42 BP, and 63 normal control cases, for a total of 43,421 slices. We implemented a comprehensive comparison between our SLP-Net and several state-of-the-art methods across the dataset. Our proposed method obtained significant performance without a large amount of data, outperformed other slice-based and volume-based approaches. The superior evaluation performance achieved in the classification experiments demonstrated the ability of our model in the differential diagnosis of VP, BP and normal cases.

Detection of COVID-19 Cases with Fuzzy Classifiers Using Chest Computed Tomography

The novel coronavirus 2019 (COVID-19) is still spreading rapidly since it first appeared in Wuhan city of China in December 2019, resulting in a worldwide pandemic. Early detection of positive cases plays a key role in preventing the further spread of the epidemic which leads to the development of diagnostic methods that give rapid and accurate responses for the detection of COVID-19. Previous studies confirmed that chest computed tomography (CT) is an indispensable tool for early screening and diagnosing of COVID-19 cases. As a result of examinations on CT scans, a radiological finding that is called ground-glass opacity, causing color, and texture change, was found in the lung of a person with COVID-19. Due to the carelessness of radiologists who work long hours and the misdiagnosis resulting in confusion of the findings with different diseases, an automatic system that helps radiologists is needed. In this paper, we present a new approach based on fuzzy classification for the detection of COVID-19 using 3D CT volumes. In the proposed approach, the skewness, kurtosis, and average statistical features of 3D CT images of patients consisting of two classes, COVID and Normal, are calculated and the value ranges are determined for both classes. Three statistical features and value ranges are used as membership functions in the development of fuzzy logic classifier. The proposed approach provides rapid and accurate diagnostics in terms of COVID vs. Normal (binary classification) under a user-friendly interface. Experimental evaluations demonstrate that our approach has great potential for radiologists to validate their initial screening and improve early diagnosis, isolation, and treatment, which contributes to infection prevention and control of the epidemic.

Deep COVID DeteCT: an international experience on COVID-19 lung detection and prognosis using chest CT

The Coronavirus disease 2019 (COVID-19) presents open questions in how we clinically diagnose and assess disease course. Recently, chest computed tomography (CT) has shown utility for COVID-19 diagnosis. In this study, we developed Deep COVID DeteCT (DCD), a deep learning convolutional neural network (CNN) that uses the entire chest CT volume to automatically predict COVID-19 (COVID+) from non-COVID-19 (COVID−) pneumonia and normal controls. We discuss training strategies and differences in performance across 13 international institutions and 8 countries. The inclusion of non-China sites in training significantly improved classification performance with area under the curve (AUCs) and accuracies above 0.8 on most test sites. Furthermore, using available follow-up scans, we investigate methods to track patient disease course and predict prognosis.

Machine-learning-based multiple abnormality prediction with large-scale chest computed tomography volumes.

Machine learning models for radiology benefit from large-scale data sets with high quality labels for abnormalities. We curated and analyzed a chest computed tomography (CT) data set of 36,316 volumes from 19,993 unique patients. This is the largest multiply-annotated volumetric medical imaging data set reported. To annotate this data set, we developed a rule-based method for automatically extracting abnormality labels from free-text radiology reports with an average F-score of 0.976 (min 0.941, max 1.0). We also developed a model for multi-organ, multi-disease classification of chest CT volumes that uses a deep convolutional neural network (CNN). This model reached a classification performance of AUROC >0.90 for 18 abnormalities, with an average AUROC of 0.773 for all 83 abnormalities, demonstrating the feasibility of learning from unfiltered whole volume CT data. We show that training on more labels improves performance significantly: for a subset of 9 labels - nodule, opacity, atelectasis, pleural effusion, consolidation, mass, pericardial effusion, cardiomegaly, and pneumothorax - the model's average AUROC increased by 10% when the number of training labels was increased from 9 to all 83. All code for volume preprocessing, automated label extraction, and the volume abnormality prediction model is publicly available. The 36,316 CT volumes and labels will also be made publicly available pending institutional approval.

Volumetric assessment of solid pulmonary nodules on ultralow-dose CT: a phantom study.

To reduce the radiation exposure from chest computed tomography (CT), ultralow-dose CT (ULDCT) protocols performed at sub-millisievert levels were previously tested for the evaluation of pulmonary nodules (PNs). The purpose of our study was to investigate the effect of ULDCT and iterative image reconstruction on volumetric measurements of solid PNs. CT datasets of an anthropomorphic chest phantom containing solid microspheres were obtained with a third-generation dual-source CT at standard dose, 1/8th, 1/20th and 1/70th of standard dose [CT volume dose index (CTDIvol): 0.03-2.03 mGy]. Semi-automated volumetric measurements were performed on CT datasets reconstructed with filtered back projection (FBP) and advanced modelled iterative reconstruction (ADMIRE), at strength level 3 and 5. Absolute percentage error (APE) evaluated measurement accuracy related to the effective volume. Scan repetition differences were evaluated using Bland-Altman analysis. Two-way analysis of variance (ANOVA) assessed influence of different scan parameters on APE. Proportional differences (PDs) tested the effect of dose settings and reconstruction algorithms on volumetric measurements, as compared to the standard protocol (standard dose-FBP). Bland-Altman analysis revealed small mean interscan differences of APE with narrow limits of agreement (-0.1%±4.3% to -0.3%±3.8%). Dose settings (P<0.001), reconstruction algorithms (P<0.001), nodule diameters (P<0.001) and nodule density (P=0.011) had statistically significant influence on APE. Post-hoc Bonferroni tests showed slightly higher APE when scanning with 1/70th of standard dose [mean difference: 3.4%, 95% confidence interval (CI): 2.5-4.3%; P<0.001], and for image reconstruction with ADMIRE5 (mean difference: 1.8%, 95% CI: 1.0-2.5%; P<0.001). No significant differences for scanning with 1/20th of standard dose (P=0.42), and image reconstruction with ADMIRE3 (P=0.19) were found. Scanning with 1/70th of standard dose and image reconstruction with FBP showed the widest range of PDs (-16.8% to 23.4%) compared to standard dose-FBP. Our phantom study showed no significant difference between nodule volume measurements on standard dose CT (CTDIvol: 2 mGy) and ULDCT with 1/20th of standard dose (CTDIvol: 0.10 mGy).

Sonographic Bedside Quantification of Pleural Effusion Compared to Computed Tomography Volumetry in ICU Patients.

ObjectivesTo date, the reliability of ultrasound for the quantitative assessment of pleural effusion has been limited. In the following study, an easy and cost-effective bedside ultrasound method was developed and investigated for specific use in the intensive care unit (ICU).Methods22 patients (median age: 58.5 years, range: 37–88 years, 14 men and 8 women) with a total of 31 pleural effusions were examined in the ICU. The inclusion criterion was complete visualization of the effusion on chest computed tomography (CT). The ultrasound (US) examination was performed less than 6 h after the diagnostic CT scan. The pleural effusion volume was calculated volumetrically from the CT scan data. Within 4.58 +/- 2.87 h after the CT scan, all patients were re-examined with US in the ICU. The fluid crescent's thickness was measured between each intercostal space (ICS) with the patient in a supine position and a 30° inclination of the torso. The US measurements were compared to the calculated CT volumes using regression analysis, resulting in the following formula: V=13.330 x ICS6 (V=volume of the effusion [ml]; ICS6=sonographic measurement of the thickness of the liquid crescent [mm] in the sixth ICS).ResultsA significant correlation between the sonographically measured and the CT-calculated volumes was best observed for the sixth ICS (R2=0.589; ICC=0.7469 with p<0.0001 and a 95% CI of 0.5364–0.8705).ConclusionThe sonographic assessment of pleural effusions in a supine position and a 30° inclination of the torso is feasible for the volumetric estimation of pleural effusion. This is especially true for ICU patients with severe primary diseases and orthopnea who are unable to sit upright or lie flat.

Automated mediastinal lymph node detection from CT volumes based on intensity targeted radial structure tensor analysis.

.This paper presents a local intensity structure analysis based on an intensity targeted radial structure tensor (ITRST) and the blob-like structure enhancement filter based on it (ITRST filter) for the mediastinal lymph node detection algorithm from chest computed tomography (CT) volumes. Although the filter based on radial structure tensor analysis (RST filter) based on conventional RST analysis can be utilized to detect lymph nodes, some lymph nodes adjacent to regions with extremely high or low intensities cannot be detected. Therefore, we propose the ITRST filter, which integrates the prior knowledge on detection target intensity range into the RST filter. Our lymph node detection algorithm consists of two steps: (1) obtaining candidate regions using the ITRST filter and (2) removing false positives (FPs) using the support vector machine classifier. We evaluated lymph node detection performance of the ITRST filter on 47 contrast-enhanced chest CT volumes and compared it with the RST and Hessian filters. The detection rate of the ITRST filter was 84.2% with 9.1 FPs/volume for lymph nodes whose short axis was at least 10 mm, which outperformed the RST and Hessian filters.

Automatic segmentation of airway tree based on local intensity filter and machine learning technique in 3D chest CT volume.

Airway segmentation plays an important role in analyzing chest computed tomography (CT) volumes for computerized lung cancer detection, emphysema diagnosis and pre- and intra-operative bronchoscope navigation. However, obtaining a complete 3D airway tree structure from a CT volume is quite a challenging task. Several researchers have proposed automated airway segmentation algorithms basically based on region growing and machine learning techniques. However, these methods fail to detect the peripheral bronchial branches, which results in a large amount of leakage. This paper presents a novel approach for more accurate extraction of the complex airway tree. This proposed segmentation method is composed of three steps. First, Hessian analysis is utilized to enhance the tube-like structure in CT volumes; then, an adaptive multiscale cavity enhancement filter is employed to detect the cavity-like structure with different radii. In the second step, support vector machine learning will be utilized to remove the false positive (FP) regions from the result obtained in the previous step. Finally, the graph-cut algorithm is used to refine the candidate voxels to form an integrated airway tree. A test dataset including 50 standard-dose chest CT volumes was used for evaluating our proposed method. The average extraction rate was about 79.1% with the significantly decreased FP rate. A new method of airway segmentation based on local intensity structure and machine learning technique was developed. The method was shown to be feasible for airway segmentation in a computer-aided diagnosis system for a lung and bronchoscope guidance system.

A bronchoscopic navigation system using bronchoscope center calibration for accurate registration of electromagnetic tracker and CT volume without markers.

Various bronchoscopic navigation systems are developed for diagnosis, staging, and treatment of lung and bronchus cancers. To construct electromagnetically navigated bronchoscopy systems, registration of preoperative images and an electromagnetic tracker must be performed. This paper proposes a new marker-free registration method, which uses the centerlines of the bronchial tree and the center of a bronchoscope tip where an electromagnetic sensor is attached, to align preoperative images and electromagnetic tracker systems. The chest computed tomography (CT) volume (preoperative images) was segmented to extract the bronchial centerlines. An electromagnetic sensor was fixed at the bronchoscope tip surface. A model was designed and printed using a 3D printer to calibrate the relationship between the fixed sensor and the bronchoscope tip center. For each sensor measurement that includes sensor position and orientation information, its corresponding bronchoscope tip center position was calculated. By minimizing the distance between each bronchoscope tip center position and the bronchial centerlines, the spatial alignment of the electromagnetic tracker system and the CT volume was determined. After obtaining the spatial alignment, an electromagnetic navigation bronchoscopy system was established to real-timely track or locate a bronchoscope inside the bronchial tree during bronchoscopic examinations. The electromagnetic navigation bronchoscopy system was validated on a dynamic bronchial phantom that can simulate respiratory motion with a breath rate range of 0-10 min(-1). The fiducial and target registration errors of this navigation system were evaluated. The average fiducial registration error was reduced from 8.7 to 6.6 mm. The average target registration error, which indicates all tracked or navigated bronchoscope position accuracy, was much reduced from 6.8 to 4.5 mm compared to previous registration methods. An electromagnetically navigated bronchoscopy system was constructed with accurate registration of an electromagnetic tracker and the CT volume on the basis of an improved marker-free registration approach that uses the bronchial centerlines and bronchoscope tip center information. The fiducial and target registration errors of our electromagnetic navigation system were about 6.6 and 4.5 mm in dynamic bronchial phantom validation.

Multi-shape graph cuts with neighbor prior constraints and its application to lung segmentation from a chest CT volume

This paper presents a novel graph cut algorithm that can take into account multi-shape constraints with neighbor prior constraints, and reports on a lung segmentation process from a three-dimensional computed tomography (CT) image based on this algorithm. The major contribution of this paper is the proposal of a novel segmentation algorithm that improves lung segmentation for cases in which the lung has a unique shape and pathologies such as pleural effusion by incorporating multiple shapes and prior information on neighbor structures in a graph cut framework. We demonstrate the efficacy of the proposed algorithm by comparing it to conventional one using a synthetic image and clinical thoracic CT volumes.

Accuracy of a Remote Eye Tracker for Radiologic Observer Studies: Effects of Calibration and Recording Environment

To determine the accuracy and reproducibility of a remote eye-tracking system for studies of observer gaze while displaying volumetric chest computed tomography (CT) images. Four participants performed calibrations using three different gray-scale backgrounds (black, gray, and white). Each participant then observed a three-dimensional 10-point test pattern embedded in five Digital Imaging and Communications in Medicine (DICOM) datasets (test backgrounds): a full 190-section chest CT scan, 190 copies of a single chest CT section, and three 190-section datasets of homogeneous intensity (black, gray, and white). Significant variances between participants, calibration backgrounds, and test backgrounds were observed. The least mean systematic error (deviation of recorded gaze position from target) was obtained when the calibration background and test background were black (27 pixels). Systematic error increased when displaying a test background that deviated from the calibration background intensity. Hence, the largest mean systematic error occurred when calibrating to a black background and displaying a white background (67 pixels). For complex chest CT volumes the white calibration background performed best (38 pixels). An angular analysis of the systematic error was performed and demonstrated that the systemic error primarily affects the vertical position of the estimated gaze position. Our findings indicate a potential source of systematic error during gaze recording in a dynamic environment and highlight the importance of configuring the calibration procedure according to the brightness of the display. We recommend that investigators develop routines for postcalibration accuracy measurement and report the effective accuracy for the display environment in which the data are collected.