Development and Validation of a Deep Learning-based Automatic Detection Algorithm for Active Pulmonary Tuberculosis on Chest Radiographs

Prediction of epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma patients on computed tomography (CT) images using 3-dimensional (3D) convolutional neural network

Background

Noninvasively detecting epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma patients before targeted therapy remains a challenge. This study aimed to develop a 3-dimensional (3D) convolutional neural network (CNN)-based deep learning model to predict EGFR mutation status using computed tomography (CT) images.

Methods

We retrospectively collected 660 patients from 2 large medical centers. The patients were divided into training (n=528) and external test (n=132) sets according to hospital source. The CNN model was trained in a supervised end-to-end manner, and its performance was evaluated using an external test set. To compare the performance of the CNN model, we constructed 1 clinical and 3 radiomics models. Furthermore, we constructed a comprehensive model combining the highest-performing radiomics and CNN models. The receiver operating characteristic (ROC) curves were used as primary measures of performance for each model. Delong test was used to compare performance differences between different models.

Results

Compared with the clinical [training set, area under the curve (AUC) =69.6%, 95% confidence interval (CI), 0.661-0.732; test set, AUC =68.4%, 95% CI, 0.609-0.752] and the highest-performing radiomics models (training set, AUC =84.3%, 95% CI, 0.812-0.873; test set, AUC =72.4%, 95% CI, 0.653-0.794) models, the CNN model (training set, AUC =94.3%, 95% CI, 0.920-0.961; test set, AUC =94.7%, 95% CI, 0.894-0.978) had significantly better predictive performance for predicting EGFR mutation status. In addition, compared with the comprehensive model (training set, AUC =95.7%, 95% CI, 0.942-0.971; test set, AUC =87.4%, 95% CI, 0.820-0.924), the CNN model had better stability.

Conclusions

The CNN model has excellent performance in non-invasively predicting EGFR mutation status in patients with lung adenocarcinoma and is expected to become an auxiliary tool for clinicians.

Evolving and Novel Applications of Artificial Intelligence in Thoracic Imaging

The advent of artificial intelligence (AI) is revolutionizing medicine, particularly radiology. With the development of newer models, AI applications are demonstrating improved performance and versatile utility in the clinical setting. Thoracic imaging is an area of profound interest, given the prevalence of chest imaging and the significant health implications of thoracic diseases. This review aims to highlight the promising applications of AI within thoracic imaging. It examines the role of AI, including its contributions to improving diagnostic evaluation and interpretation, enhancing workflow, and aiding in invasive procedures. Next, it further highlights the current challenges and limitations faced by AI, such as the necessity of 'big data', ethical and legal considerations, and bias in representation. Lastly, it explores the potential directions for the application of AI in thoracic radiology.

[Clinical Application of Artificial Intelligence-Based Detection Assistance Devices for Chest X-Ray Interpretation: Current Status and Practical Considerations]

Artificial intelligence (AI) technology is actively being applied for the interpretation of medical imaging, such as chest X-rays. AI-based software medical devices, which automatically detect various types of abnormal findings in chest X-ray images to assist physicians in their interpretation, are actively being commercialized and clinically implemented in Korea. Several important issues need to be considered for AI-based detection assistant tools to be applied in clinical practice: the evaluation of performance and efficacy prior to implementation; the determination of the target application, range, and method of delivering results; and monitoring after implementation and legal liability issues. Appropriate decision making regarding these devices based on the situation in each institution is necessary. Radiologists must be engaged as medical assessment experts using the software for these devices as well as in medical image interpretation to ensure the safe and efficient implementation and operation of AI-based detection assistant tools.

A multimodal integration pipeline for accurate diagnosis, pathogen identification, and prognosis prediction of pulmonary infections

Pulmonary infections pose formidable challenges in clinical settings with high mortality rates across all age groups worldwide. Accurate diagnosis and early intervention are crucial to improve patient outcomes. Artificial intelligence (AI) has the capability to mine imaging features specific to different pathogens and fuse multimodal features to reach a synergistic diagnosis, enabling more precise investigation and individualized clinical management. In this study, we successfully developed a multimodal integration (MMI) pipeline to differentiate among bacterial, fungal, and viral pneumonia and pulmonary tuberculosis based on a real-world dataset of 24,107 patients. The area under the curve (AUC) of the MMI system comprising clinical text and computed tomography (CT) image scans yielded 0.910 (95% confidence interval [CI]: 0.904-0.916) and 0.887 (95% CI: 0.867-0.909) in the internal and external testing datasets respectively, which were comparable to those of experienced physicians. Furthermore, the MMI system was utilized to rapidly differentiate between viral subtypes with a mean AUC of 0.822 (95% CI: 0.805-0.837) and bacterial subtypes with a mean AUC of 0.803 (95% CI: 0.775-0.830). Here, the MMI system harbors the potential to guide tailored medication recommendations, thus mitigating the risk of antibiotic misuse. Additionally, the integration of multimodal factors in the AI-driven system also provided an evident advantage in predicting risks of developing critical illness, contributing to more informed clinical decision-making. To revolutionize medical care, embracing multimodal AI tools in pulmonary infections will pave the way to further facilitate early intervention and precise management in the foreseeable future.

A deep learning-based algorithm for pulmonary tuberculosis detection in chest radiography

In tuberculosis (TB), chest radiography (CXR) patterns are highly variable, mimicking pneumonia and many other diseases. This study aims to evaluate the efficacy of Google teachable machine, a deep neural network-based image classification tool, to develop algorithm for predicting TB probability of CXRs. The training dataset included 348 TB CXRs and 3806 normal CXRs for training TB detection. We also collected 1150 abnormal CXRs and 627 normal CXRs for training abnormality detection. For external validation, we collected 250 CXRs from our hospital. We also compared the accuracy of the algorithm to five pulmonologists and radiological reports. In external validation, the AI algorithm showed areas under the curve (AUC) of 0.951 and 0.975 in validation dataset 1 and 2. The accuracy of the pulmonologists on validation dataset 2 showed AUC range of 0.936-0.995. When abnormal CXRs other than TB were added, AUC decreased in both human readers (0.843-0.888) and AI algorithm (0.828). When combine human readers with AI algorithm, the AUC further increased to 0.862-0.885. The TB CXR AI algorithm developed by using Google teachable machine in this study is effective, with the accuracy close to experienced clinical physicians, and may be helpful for detecting tuberculosis by CXR.

Artificial intelligence-based radiographic extent analysis to predict tuberculosis treatment outcomes: a multicenter cohort study

Predicting outcomes in pulmonary tuberculosis is challenging despite effective treatments. This study aimed to identify factors influencing treatment success and culture conversion, focusing on artificial intelligence (AI)-based chest X-ray analysis and Xpert MTB/RIF assay cycle threshold (Ct) values. In this retrospective study across six South Korean referral centers (January 1 to December 31, 2019), we included adults with rifampicin-susceptible pulmonary tuberculosis confirmed by Xpert assay from sputum samples. We analyzed patient characteristics, AI-based tuberculosis extent scores from chest X-rays, and Xpert Ct values. Of 230 patients, 206 (89.6%) achieved treatment success. The median age was 61 years, predominantly male (76.1%). AI-based radiographic tuberculosis extent scores (median 7.5) significantly correlated with treatment success (odds ratio [OR] 0.938, 95% confidence interval [CI] 0.895-0.983) and culture conversion at 8 weeks (liquid medium: OR 0.911, 95% CI 0.853-0.973; solid medium: OR 0.910, 95% CI 0.850-0.973). Sputum smear positivity was 49.6%, with a median Ct of 26.2. However, Ct values did not significantly correlate with major treatment outcomes. AI-based radiographic scoring at diagnosis is a significant predictor of treatment success and culture conversion in pulmonary tuberculosis, underscoring its potential in personalized patient management.

Development and validation of a self-attention network-based algorithm to detect mediastinal lesions on computed tomography images

Background

Diagnosis of mediastinal lesions on computed tomography (CT) images is challenging for radiologists, as numerous conditions can present as mass-like lesions at this site. This study aimed to develop a self-attention network-based algorithm to detect mediastinal lesions on CT images and to evaluate its efficacy in lesion detection.

Methods

In this study, two separate large-scale open datasets [National Institutes of Health (NIH) DeepLesion and Medical Image Computing and Computer Assisted Intervention (MICCAI) 2022 Mediastinal Lesion Analysis (MELA) Challenge] were collected to develop a self-attention network-based algorithm for mediastinal lesion detection. We enrolled 921 abnormal CT images from the NIH DeepLesion dataset into the pretraining stage and 880 abnormal CT images from the MELA Challenge dataset into the model training and validation stages in a ratio of 8:2 at the patient level. The average precision (AP) and confidence score on lesion detection were evaluated in the validation set. Sensitivity to lesion detection was compared between the faster region-based convolutional neural network (R-CNN) model and the proposed model.

Results

The proposed model achieved an 89.3% AP score in mediastinal lesion detection and could identify comparably large lesions with a high confidence score >0.8. Moreover, the proposed model achieved a performance boost of almost 2% in the competition performance metric (CPM) compared to the faster R-CNN model. In addition, the proposed model can ensure an outstanding sensitivity with a relatively low false-positive rate by setting appropriate threshold values.

Conclusions

The proposed model showed excellent performance in detecting mediastinal lesions on CT. Thus, it can drastically reduce radiologists' workload, improve their performance, and speed up the reporting time in everyday clinical practice.

Utilizing Deep Learning and Computed Tomography to Determine Pulmonary Nodule Activity in Patients With Nontuberculous Mycobacterial-Lung Disease

PURPOSE

To develop and evaluate a deep convolutional neural network (DCNN) model for the classification of acute and chronic lung nodules from nontuberculous mycobacterial-lung disease (NTM-LD) on computed tomography (CT).

MATERIALS AND METHODS

We collected a data set of 650 nodules (316 acute and 334 chronic) from the CT scans of 110 patients with NTM-LD. The data set was divided into training, validation, and test sets in a ratio of 4:1:1. Bounding boxes were used to crop the 2D CT images down to the area of interest. A DCNN model was built using 11 convolutional layers and trained on these images. The performance of the model was evaluated on the hold-out test set and compared with that of 3 radiologists who independently reviewed the images.

RESULTS

The DCNN model achieved an area under the receiver operating characteristic curve of 0.806 for differentiating acute and chronic NTM-LD nodules, corresponding to sensitivity, specificity, and accuracy of 76%, 68%, and 72%, respectively. The performance of the model was comparable to that of the 3 radiologists, who had area under the receiver operating characteristic curve, sensitivity, specificity, and accuracy of 0.693 to 0.771, 61% to 82%, 59% to 73%, and 60% to 73%, respectively.

CONCLUSIONS

This study demonstrated the feasibility of using a DCNN model for the classification of the activity of NTM-LD nodules on chest CT. The model performance was comparable to that of radiologists. This approach can potentially and efficiently improve the diagnosis and management of NTM-LD.

AI for Detection of Tuberculosis: Implications for Global Health

Tuberculosis, which primarily affects developing countries, remains a significant global health concern. Since the 2010s, the role of chest radiography has expanded in tuberculosis triage and screening beyond its traditional complementary role in the diagnosis of tuberculosis. Computer-aided diagnosis (CAD) systems for tuberculosis detection on chest radiographs have recently made substantial progress in diagnostic performance, thanks to deep learning technologies. The current performance of CAD systems for tuberculosis has approximated that of human experts, presenting a potential solution to the shortage of human readers to interpret chest radiographs in low- or middle-income, high-tuberculosis-burden countries. This article provides a critical appraisal of developmental process reporting in extant CAD software for tuberculosis, based on the Checklist for Artificial Intelligence in Medical Imaging. It also explores several considerations to scale up CAD solutions, encompassing manufacturer-independent CAD validation, economic and political aspects, and ethical concerns, as well as the potential for broadening radiography-based diagnosis to other nontuberculosis diseases. Collectively, CAD for tuberculosis will emerge as a representative deep learning application, catalyzing advances in global health and health equity. Keywords: Computer-aided Diagnosis (CAD), Conventional Radiography, Thorax, Lung, Machine Learning Supplemental material is available for this article. © RSNA, 2024.

A scoping review of educational programmes on artificial intelligence (AI) available to medical imaging staff

INTRODUCTION

Medical imaging is arguably the most technologically advanced field in healthcare, encompassing a range of technologies which continually evolve as computing power and human knowledge expand. Artificial Intelligence (AI) is the next frontier which medical imaging is pioneering. The rapid development and implementation of AI has the potential to revolutionise healthcare, however, to do so, staff must be competent and confident in its application, hence AI readiness is an important precursor to AI adoption. Research to ascertain the best way to deliver this AI-enabled healthcare training is in its infancy. The aim of this scoping review is to compare existing studies which investigate and evaluate the efficacy of AI educational interventions for medical imaging staff.

METHODS

Following the creation of a search strategy and keyword searches, screening was conducted to determine study eligibility. This consisted of a title and abstract scan, then subsequently a full-text review. Articles were included if they were empirical studies wherein an educational intervention on AI for medical imaging staff was created, delivered, and evaluated.

RESULTS

Of the initial 1309 records returned, n = 5 (∼0.4 %) of studies met the eligibility criteria of the review. The curricula and delivery in each of the five studies shared similar aims and a 'flipped classroom' delivery was the most utilised method. However, the depth of content covered in the curricula of each varied and measured outcomes differed greatly.

CONCLUSION

The findings of this review will provide insights into the evaluation of existing AI educational interventions, which will be valuable when planning AI education for healthcare staff.

IMPLICATIONS FOR PRACTICE

This review highlights the need for standardised and comprehensive AI training programs for imaging staff.

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.

A deep-learning-based framework for identifying and localizing multiple abnormalities and assessing cardiomegaly in chest X-ray

Accurate identification and localization of multiple abnormalities are crucial steps in the interpretation of chest X-rays (CXRs); however, the lack of a large CXR dataset with bounding boxes severely constrains accurate localization research based on deep learning. We created a large CXR dataset named CXR-AL14, containing 165,988 CXRs and 253,844 bounding boxes. On the basis of this dataset, a deep-learning-based framework was developed to identify and localize 14 common abnormalities and calculate the cardiothoracic ratio (CTR) simultaneously. The mean average precision values obtained by the model for 14 abnormalities reached 0.572-0.631 with an intersection-over-union threshold of 0.5, and the intraclass correlation coefficient of the CTR algorithm exceeded 0.95 on the held-out, multicentre and prospective test datasets. This framework shows an excellent performance, good generalization ability and strong clinical applicability, which is superior to senior radiologists and suitable for routine clinical settings.

Applicability of artificial intelligence-based computer-aided detection (AI-CAD) for pulmonary tuberculosis to community-based active case finding

BACKGROUND

Artificial intelligence-based computer-aided detection (AI-CAD) for tuberculosis (TB) has become commercially available and several studies have been conducted to evaluate the performance of AI-CAD for pulmonary tuberculosis (TB) in clinical settings. However, little is known about its applicability to community-based active case-finding (ACF) for TB.

METHODS

We analysed an anonymized data set obtained from a community-based ACF in Cambodia, targeting persons aged 55 years or over, persons with any TB symptoms, such as chronic cough, and persons at risk of TB, including household contacts. All of the participants in the ACF were screened by chest radiography (CXR) by Cambodian doctors, followed by Xpert test when they were eligible for sputum examination. Interpretation by an experienced chest physician and abnormality scoring by a newly developed AI-CAD were retrospectively conducted for the CXR images. With a reference of Xpert-positive TB or human interpretations, receiver operating characteristic (ROC) curves were drawn to evaluate the AI-CAD performance by area under the ROC curve (AUROC). In addition, its applicability to community-based ACFs in Cambodia was examined.

RESULTS

TB scores of the AI-CAD were significantly associated with the CXR classifications as indicated by the severity of TB disease, and its AUROC as the bacteriological reference was 0.86 (95% confidence interval 0.83-0.89). Using a threshold for triage purposes, the human reading and bacteriological examination needed fell to 21% and 15%, respectively, detecting 95% of Xpert-positive TB in ACF. For screening purposes, we could detect 98% of Xpert-positive TB cases.

CONCLUSIONS

AI-CAD is applicable to community-based ACF in high TB burden settings, where experienced human readers for CXR images are scarce. The use of AI-CAD in developing countries has the potential to expand CXR screening in community-based ACFs, with a substantial decrease in the workload on human readers and laboratory labour. Further studies are needed to generalize the results to other countries by increasing the sample size and comparing the AI-CAD performance with that of more human readers.

Assessment of inspiration and technical quality in anteroposterior thoracic radiographs using machine learning

INTRODUCTION

Chest radiographs are the most performed radiographic procedure, but suboptimal technical factors can impact clinical interpretation. A deep learning model was developed to assess technical and inspiratory adequacy of anteroposterior chest radiographs.

METHODS

Adult anteroposterior chest radiographs (n = 2375) were assessed for technical adequacy, and if otherwise technically adequate, for adequacy of inspiration. Images were labelled by an experienced radiologist with one of three ground truth labels: inadequate technique (n = 605, 25.5 %), adequate inspiration (n = 900, 37.9 %), and inadequate inspiration (n = 870, 36.6 %). A convolutional neural network was then iteratively trained to predict these labels and evaluated using recall, precision, F1 and micro-F1, and Gradient-weighted Class Activation Mapping analysis on a hold-out test set. Impact of kyphosis on model accuracy was assessed.

RESULTS

The model performed best for radiographs with adequate technique, and worst for images with inadequate technique. Recall was highest (89 %) for radiographs with both adequate technique and inspiration, with recall of 81 % for images with adequate technique and inadequate inspiration, and 60 % for images with inadequate technique, although precision was highest (85 %) for this category. Per-class F1 was 80 %, 81 % and 70 % for adequate inspiration, inadequate inspiration, and inadequate technique respectively. Weighted F1 and Micro F1 scores were 78 %. Presence or absence of kyphosis had no significant impact on model accuracy in images with adequate technique.

CONCLUSION

This study explores the promising performance of a machine learning algorithm for assessment of inspiratory adequacy and overall technical adequacy for anteroposterior chest radiograph acquisition.

IMPLICATIONS FOR PRACTICE

With further refinement, machine learning can contribute to education and quality improvement in radiology departments.

Distinguishing infectivity in patients with pulmonary tuberculosis using deep learning

Introduction

This study aimed to develop and assess a deep-learning model based on CT images for distinguishing infectivity in patients with pulmonary tuberculosis (PTB).

Methods

We labeled all 925 patients from four centers with weak and strong infectivity based on multiple sputum smears within a month for our deep-learning model named TBINet's training. We compared TBINet's performance in identifying infectious patients to that of the conventional 3D ResNet model. For model explainability, we used gradient-weighted class activation mapping (Grad-CAM) technology to identify the site of lesion activation in the CT images.

Results

The TBINet model demonstrated superior performance with an area under the curve (AUC) of 0.819 and 0.753 on the validation and external test sets, respectively, compared to existing deep learning methods. Furthermore, using Grad-CAM, we observed that CT images with higher levels of consolidation, voids, upper lobe involvement, and enlarged lymph nodes were more likely to come from patients with highly infectious forms of PTB.

Conclusion

Our study proves the feasibility of using CT images to identify the infectivity of PTB patients based on the deep learning method.

Effect of multimodal diagnostic approach using deep learning-based automated detection algorithm for active pulmonary tuberculosis

In this study, we developed a model to predict culture test results for pulmonary tuberculosis (PTB) with a customized multimodal approach and evaluated its performance in different clinical settings. Moreover, we investigated potential performance improvements by combining this approach with deep learning-based automated detection algorithms (DLADs). This retrospective observational study enrolled patients over 18 years of age who consecutively visited the level 1 emergency department and underwent chest radiograph and sputum testing. The primary endpoint was positive sputum culture for PTB. We compared the performance of the diagnostic models by replacing radiologists' interpretations of chest radiographs with screening scores calculated through DLAD. The optimal diagnostic model had an area under the receiver operating characteristic curve of 0.924 (95% CI 0.871-0.976) and an area under precision recall curve of 0.403 (95% CI 0.195-0.580) while maintaining a specificity of 81.4% when sensitivity was fixed at 90%. Multicomponent models showed improved performance for detecting PTB when chest radiography interpretation was replaced by DLAD. Multicomponent diagnostic models with DLAD customized for different clinical settings are more practical than traditional methods for detecting patients with PTB. This novel diagnostic approach may help prevent the spread of PTB and optimize healthcare resource utilization in resource-limited clinical settings.

Effects of Implementing Artificial Intelligence-Based Computer-Aided Detection for Chest Radiographs in Daily Practice on the Rate of Referral to Chest Computed Tomography in Pulmonology Outpatient Clinic

OBJECTIVE

The clinical impact of artificial intelligence-based computer-aided detection (AI-CAD) beyond diagnostic accuracy remains uncertain. We aimed to investigate the influence of the clinical implementation of AI-CAD for chest radiograph (CR) interpretation in daily practice on the rate of referral for chest computed tomography (CT).

MATERIALS AND METHODS

AI-CAD was implemented in clinical practice at the Seoul National University Hospital. CRs obtained from patients who visited the pulmonology outpatient clinics before (January-December 2019) and after (January-December 2020) implementation were included in this study. After implementation, the referring pulmonologist requested CRs with or without AI-CAD analysis. We conducted multivariable logistic regression analyses to evaluate the associations between using AI-CAD and the following study outcomes: the rate of chest CT referral, defined as request and actual acquisition of chest CT within 30 days after CR acquisition, and the CT referral rates separately for subsequent positive and negative CT results. Multivariable analyses included various covariates such as patient age and sex, time of CR acquisition (before versus after AI-CAD implementation), referring pulmonologist, nature of the CR examination (baseline versus follow-up examination), and radiology reports presence at the time of the pulmonology visit.

RESULTS

A total of 28546 CRs from 14565 patients (mean age: 67 years; 7130 males) and 25888 CRs from 12929 patients (mean age: 67 years; 6435 males) before and after AI-CAD implementation were included. The use of AI-CAD was independently associated with increased chest CT referrals (odds ratio [OR], 1.33; P = 0.008) and referrals with subsequent negative chest CT results (OR, 1.46; P = 0.005). Meanwhile, referrals with positive chest CT results were not significantly associated with AI-CAD use (OR, 1.08; P = 0.647).

CONCLUSION

The use of AI-CAD for CR interpretation in pulmonology outpatients was independently associated with an increased frequency of overall referrals for chest CT scans and referrals with subsequent negative results.

Tuberculosis Diagnosis Using Deep Transferred EfficientNet

Tuberculosis is a very deadly disease, with more than half of all tuberculosis cases dead in countries and regions with relatively poor health care resources. Fortunately, the disease is curable, and early diagnosis and medication can go a long way toward curing TB patients. Unfortunately, traditional methods of TB diagnosis rely on specialist doctors, which is lacking in areas with high TB mortality rates. Diagnostic methods based on artificial intelligence technology are one of the solutions to this problem. We propose a Deep Transferred EfficientNet with SVM (DTE-SVM), which replaces the pre-trained EfficientNet classification layer with an SVM classifier and achieves auspicious performance on a small dataset. After ten runs of 10-fold Cross-Validation, the DTE-SVM has a sensitivity of 93.89±1.96, a specificity of 95.35±1.31, a precision of 95.30±1.24, an accuracy of 94.62±1.00, and an F1-score of 94.62±1.00. In addition, our study conducted ablation studies on the effect of the SVM classifier on model performance and briefly discussed the results.

An Efficient, Lightweight, Tiny 2D-CNN Ensemble Model to Detect Cardiomegaly in Heart CT Images

Cardiomegaly is a significant global health concern, especially in developing nations. Although advanced clinical care is available for newly diagnosed patients, many in resource-limited regions face late diagnoses and consequent increased mortality. This challenge is accentuated by a scarcity of radiography equipment and radiologists. Hence, we propose the development of a computer-aided diagnostic (CAD) system, specifically a lightweight, tiny 2D-CNN ensemble model, to facilitate early detection and, potentially, reduce mortality rates. Deep learning, with its subset of convolutional neural networks (CNN), has shown potential in visual applications, especially in medical image diagnosis. However, traditional deep CNNs often face compatibility issues with object-oriented human factor technology. Our proposed model aims to bridge this gap. Using CT scan images sourced from the Mendeley data center, our tiny 2D-CNN ensemble learning model achieved an accuracy of 96.32%, offering a promising tool for efficient and accurate cardiomegaly detection.

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.

Generative Adversarial Network (Generative Artificial Intelligence) in Pediatric Radiology: A Systematic Review

Generative artificial intelligence, especially with regard to the generative adversarial network (GAN), is an important research area in radiology as evidenced by a number of literature reviews on the role of GAN in radiology published in the last few years. However, no review article about GAN in pediatric radiology has been published yet. The purpose of this paper is to systematically review applications of GAN in pediatric radiology, their performances, and methods for their performance evaluation. Electronic databases were used for a literature search on 6 April 2023. Thirty-seven papers met the selection criteria and were included. This review reveals that the GAN can be applied to magnetic resonance imaging, X-ray, computed tomography, ultrasound and positron emission tomography for image translation, segmentation, reconstruction, quality assessment, synthesis and data augmentation, and disease diagnosis. About 80% of the included studies compared their GAN model performances with those of other approaches and indicated that their GAN models outperformed the others by 0.1-158.6%. However, these study findings should be used with caution because of a number of methodological weaknesses. For future GAN studies, more robust methods will be essential for addressing these issues. Otherwise, this would affect the clinical adoption of the GAN-based applications in pediatric radiology and the potential advantages of GAN could not be realized widely.

Artificial Intelligence in Paediatric Tuberculosis

Tuberculosis (TB) continues to be a leading cause of death in children despite global efforts focused on early diagnosis and interventions to limit the spread of the disease. This challenge has been made more complex in the context of the coronavirus pandemic, which has disrupted the "End TB Strategy" and framework set out by the World Health Organization (WHO). Since the inception of artificial intelligence (AI) more than 60 years ago, the interest in AI has risen and more recently we have seen the emergence of multiple real-world applications, many of which relate to medical imaging. Nonetheless, real-world AI applications and clinical studies are limited in the niche area of paediatric imaging. This review article will focus on how AI, or more specifically deep learning, can be applied to TB diagnosis and management in children. We describe how deep learning can be utilised in chest imaging to provide computer-assisted diagnosis to augment workflow and screening efforts. We also review examples of recent AI applications for TB screening in resource constrained environments and we explore some of the challenges and the future directions of AI in paediatric TB.

Estimating the Volume of Nodules and Masses on Serial Chest Radiography Using a Deep-Learning-Based Automatic Detection Algorithm: A Preliminary Study

BACKGROUND

The purpose of this study was to assess the volume of the pulmonary nodules and masses on serial chest X-rays (CXRs) from deep-learning-based automatic detection algorithm (DLAD)-based parameters.

METHODS

In a retrospective single-institutional study, 72 patients, who obtained serial CXRs (n = 147) for pulmonary nodules or masses with corresponding chest CT images as the reference standards, were included. A pre-trained DLAD based on a convolutional neural network was developed to detect and localize nodules using 13,710 radiographs and to calculate a localization map and the derived parameters (e.g., the area and mean probability value of pulmonary nodules) for each CXR, including serial follow-ups. For validation, reference 3D CT volumes were measured semi-automatically. Volume prediction models for pulmonary nodules were established through univariable or multivariable, and linear or non-linear regression analyses with the parameters. A polynomial regression analysis was performed as a method of a non-linear regression model.

RESULTS

Of the 147 CXRs and 208 nodules of 72 patients, the mean volume of nodules or masses was measured as 9.37 ± 11.69 cm³ (mean ± standard deviation). The area and CT volume demonstrated a linear correlation of moderate strength (i.e., R = 0.58, RMSE: 9449.9 mm³ m³ in a linear regression analysis). The area and mean probability values exhibited a strong linear correlation (R = 0.73). The volume prediction performance based on a multivariable regression model was best with a mean probability and unit-adjusted area (i.e.

, RMSE

7975.6 mm³, the smallest among the other variable parameters).

CONCLUSIONS

The prediction model with the area and the mean probability based on the DLAD showed a rather accurate quantitative estimation of pulmonary nodule or mass volume and the change in serial CXRs.

Effect of Contrast Level and Image Format on a Deep Learning Algorithm for the Detection of Pneumothorax with Chest Radiography

Under the black-box nature in the deep learning model, it is uncertain how the change in contrast level and format affects the performance. We aimed to investigate the effect of contrast level and image format on the effectiveness of deep learning for diagnosing pneumothorax on chest radiographs. We collected 3316 images (1016 pneumothorax and 2300 normal images), and all images were set to the standard contrast level (100%) and stored in the Digital Imaging and Communication in Medicine and Joint Photographic Experts Group (JPEG) formats. Data were randomly separated into 80% of training and 20% of test sets, and the contrast of images in the test set was changed to 5 levels (50%, 75%, 100%, 125%, and 150%). We trained the model to detect pneumothorax using ResNet-50 with 100% level images and tested with 5-level images in the two formats. While comparing the overall performance between each contrast level in the two formats, the area under the receiver-operating characteristic curve (AUC) was significantly different (all p < 0.001) except between 125 and 150% in JPEG format (p = 0.382). When comparing the two formats at same contrast levels, AUC was significantly different (all p < 0.001) except 50% and 100% (p = 0.079 and p = 0.082, respectively). The contrast level and format of medical images could influence the performance of the deep learning model. It is required to train with various contrast levels and formats of image, and further image processing for improvement and maintenance of the performance.

Identification of Active Pulmonary Tuberculosis Among Patients With Positive Interferon-Gamma Release Assay Results: Value of a Deep Learning-based Computer-aided Detection System in Different Scenarios of Implementation

PURPOSE

To evaluate the accuracy of a deep learning-based computer-aided detection (CAD) system in identifying active pulmonary tuberculosis on chest radiographs (CRs) of patients with positive interferon-gamma release assay (IGRA) results in different scenarios of clinical implementation.

MATERIALS AND METHODS

We collected the CRs of consecutive patients with positive IGRA results. Findings of active pulmonary tuberculosis on CRs were independently evaluated by the CAD and a thoracic radiologist, followed by interpretation using the CAD. Sensitivity and specificity were evaluated in different scenarios: (a) radiologists' interpretation, (b) radiologists' CAD-assisted interpretation, and (c) CAD-based prescreening (radiologists' interpretation for positive CAD results only). We conducted a reader test to compare the accuracy of the CAD with those of 5 radiologists.

RESULTS

Among 1780 patients (men, 53.8%; median age, 56 y), 44 (2.5%) were diagnosed with active pulmonary tuberculosis. The CAD-assisted interpretation exhibited a higher sensitivity (81.8% vs. 72.7%; P =0.046) but lower specificity than the radiologists' interpretation (84.1% vs. 85.7%; P <0.001). The CAD-based prescreening exhibited a higher specificity than the radiologists' interpretation (88.8% vs. 85.7%; P <0.001) at the same sensitivity, with a workload reduction of 85.2% (1780 to 263). In the reader test, the CAD exhibited a higher sensitivity than radiologists (72.7% vs. 59.5%; P =0.005) at the same specificity (88.0%), and CAD-assisted interpretation significantly improved the sensitivity of radiologists' interpretation (72.3%; P <0.001).

CONCLUSIONS

For identifying active pulmonary tuberculosis among patients with positive IGRA results, deep learning-based CAD can enhance the sensitivity of interpretation. CAD-based prescreening may reduce the radiologists' workload at an improved specificity.

Multicentre external validation of a commercial artificial intelligence software to analyse chest radiographs in health screening environments with low disease prevalence

OBJECTIVES

To externally validate the performance of a commercial AI software program for interpreting CXRs in a large, consecutive, real-world cohort from primary healthcare centres.

METHODS

A total of 3047 CXRs were collected from two primary healthcare centres, characterised by low disease prevalence, between January and December 2018. All CXRs were labelled as normal or abnormal according to CT findings. Four radiology residents read all CXRs twice with and without AI assistance. The performances of the AI and readers with and without AI assistance were measured in terms of area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity.

RESULTS

The prevalence of clinically significant lesions was 2.2% (68 of 3047). The AUROC, sensitivity, and specificity of the AI were 0.648 (95% confidence interval [CI] 0.630-0.665), 35.3% (CI, 24.7-47.8), and 94.2% (CI, 93.3-95.0), respectively. AI detected 12 of 41 pneumonia, 3 of 5 tuberculosis, and 9 of 22 tumours. AI-undetected lesions tended to be smaller than true-positive lesions. The readers' AUROCs ranged from 0.534-0.676 without AI and 0.571-0.688 with AI (all p values < 0.05). For all readers, the mean reading time was 2.96-10.27 s longer with AI assistance (all p values < 0.05).

CONCLUSIONS

The performance of commercial AI in these high-volume, low-prevalence settings was poorer than expected, although it modestly boosted the performance of less-experienced readers. The technical prowess of AI demonstrated in experimental settings and approved by regulatory bodies may not directly translate to real-world practice, especially where the demand for AI assistance is highest.

KEY POINTS

• This study shows the limited applicability of commercial AI software for detecting abnormalities in CXRs in a health screening population. • When using AI software in a specific clinical setting that differs from the training setting, it is necessary to adjust the threshold or perform additional training with such data that reflects this environment well. • Prospective test accuracy studies, randomised controlled trials, or cohort studies are needed to examine AI software to be implemented in real clinical practice.

A deep learning model using chest X-ray for identifying TB and NTM-LD patients: a cross-sectional study

BACKGROUND

Timely differentiating between pulmonary tuberculosis (TB) and nontuberculous mycobacterial lung disease (NTM-LD), which are radiographically similar, is important because infectiousness and treatment differ. This study aimed to evaluate whether artificial intelligence could distinguish between TB or NTM-LD patients by chest X-rays (CXRs) from suspects of mycobacterial lung disease.

METHODS

A total of 1500 CXRs, including 500 each from patients with pulmonary TB, NTM-LD, and patients with clinical suspicion but negative mycobacterial culture (Imitator) from two hospitals, were retrospectively collected and evaluated in this study. We developed a deep neural network (DNN) and evaluated model performance using the area under the receiver operating characteristic curves (AUC) in both internal and external test sets. Furthermore, we conducted a reader study and tested our model under three scenarios of different mycobacteria prevalence.

RESULTS

Among the internal and external test sets, the AUCs of our DNN model were 0.83 ± 0.005 and 0.76 ± 0.006 for pulmonary TB, 0.86 ± 0.006 and 0.64 ± 0.017 for NTM-LD, and 0.77 ± 0.007 and 0.74 ± 0.005 for Imitator. The DNN model showed higher performance on the internal test set in classification accuracy (66.5 ± 2.5%) than senior (50.8 ± 3.0%, p < 0.001) and junior pulmonologists (47.5 ± 2.8%, p < 0.001). Among different prevalence scenarios, the DNN model has stable performance in terms of AUC to detect TB and mycobacterial lung disease.

CONCLUSION

DNN model had satisfactory performance and a higher accuracy than pulmonologists on classifying patients with presumptive mycobacterial lung diseases. DNN model could be a complementary first-line screening tool.

Prediction of model generalizability for unseen data: Methodology and case study in brain metastases detection in T1-Weighted contrast-enhanced 3D MRI

BACKGROUND AND PURPOSE

A medical AI system's generalizability describes the continuity of its performance acquired from varying geographic, historical, and methodologic settings. Previous literature on this topic has mostly focused on "how" to achieve high generalizability (e.g., via larger datasets, transfer learning, data augmentation, model regularization schemes), with limited success. Instead, we aim to understand "when" the generalizability is achieved: Our study presents a medical AI system that could estimate its generalizability status for unseen data on-the-fly.

MATERIALS AND METHODS

We introduce a latent space mapping (LSM) approach utilizing Fréchet distance loss to force the underlying training data distribution into a multivariate normal distribution. During the deployment, a given test data's LSM distribution is processed to detect its deviation from the forced distribution; hence, the AI system could predict its generalizability status for any previously unseen data set. If low model generalizability is detected, then the user is informed by a warning message integrated into a sample deployment workflow. While the approach is applicable for most classification deep neural networks (DNNs), we demonstrate its application to a brain metastases (BM) detector for T1-weighted contrast-enhanced (T1c) 3D MRI. The BM detection model was trained using 175 T1c studies acquired internally (from the authors' institution) and tested using (1) 42 internally acquired exams and (2) 72 externally acquired exams from the publicly distributed Brain Mets dataset provided by the Stanford University School of Medicine. Generalizability scores, false positive (FP) rates, and sensitivities of the BM detector were computed for the test datasets.

RESULTS AND CONCLUSION

The model predicted its generalizability to be low for 31% of the testing data (i.e., two of the internally and 33 of the externally acquired exams), where it produced (1) ∼13.5 false positives (FPs) at 76.1% BM detection sensitivity for the low and (2) ∼10.5 FPs at 89.2% BM detection sensitivity for the high generalizability groups respectively. These results suggest that the proposed formulation enables a model to predict its generalizability for unseen data.

Impact of localized fine tuning in the performance of segmentation and classification of lung nodules from computed tomography scans using deep learning

Background

Algorithm malfunction may occur when there is a performance mismatch between the dataset with which it was developed and the dataset on which it was deployed.

Methods

A baseline segmentation algorithm and a baseline classification algorithm were developed using public dataset of Lung Image Database Consortium to detect benign and malignant nodules, and two additional external datasets (i.e., HB and XZ) including 542 cases and 486 cases were involved for the independent validation of these two algorithms. To explore the impact of localized fine tuning on the individual segmentation and classification process, the baseline algorithms were fine tuned with CT scans of HB and XZ datasets, respectively, and the performance of the fine tuned algorithms was tested to compare with the baseline algorithms.

Results

The proposed baseline algorithms of both segmentation and classification experienced a drop when directly deployed in external HB and XZ datasets. Comparing with the baseline validation results in nodule segmentation, the fine tuned segmentation algorithm obtained better performance in Dice coefficient, Intersection over Union, and Average Surface Distance in HB dataset (0.593 vs. 0.444; 0.450 vs. 0.348; 0.283 vs. 0.304) and XZ dataset (0.601 vs. 0.486; 0.482 vs. 0.378; 0.225 vs. 0.358). Similarly, comparing with the baseline validation results in benign and malignant nodule classification, the fine tuned classification algorithm had improved area under the receiver operating characteristic curve value, accuracy, and F1 score in HB dataset (0.851 vs. 0.812; 0.813 vs. 0.769; 0.852 vs. 0.822) and XZ dataset (0.724 vs. 0.668; 0.696 vs. 0.617; 0.737 vs. 0.668).

Conclusions

The external validation performance of localized fine tuned algorithms outperformed the baseline algorithms in both segmentation process and classification process, which showed that localized fine tuning may be an effective way to enable a baseline algorithm generalize to site-specific use.

Hospital-wide survey of clinical experience with artificial intelligence applied to daily chest radiographs

PURPOSE

To assess experience with and perceptions of clinical application of artificial intelligence (AI) to chest radiographs among doctors in a single hospital.

MATERIALS AND METHODS

A hospital-wide online survey of the use of commercially available AI-based lesion detection software for chest radiographs was conducted with all clinicians and radiologists at our hospital in this prospective study. In our hospital, version 2 of the abovementioned software was utilized from March 2020 to February 2021 and could detect three types of lesions. Version 3 was utilized for chest radiographs by detecting nine types of lesions from March 2021. The participants of this survey answered questions on their own experience using AI-based software in daily practice. The questionnaires were composed of single choice, multiple choices, and scale bar questions. Answers were analyzed according to the clinicians and radiologists using paired t-test and the Wilcoxon rank-sum test.

RESULTS

One hundred twenty-three doctors answered the survey, and 74% completed all questions. The proportion of individuals who utilized AI was higher among radiologists than clinicians (82.5% vs. 45.9%, p = 0.008). AI was perceived as being the most useful in the emergency room, and pneumothorax was considered the most valuable finding. Approximately 21% of clinicians and 16% of radiologists changed their own reading results after referring to AI, and trust levels for AI were 64.9% and 66.5%, respectively. Participants thought AI helped reduce reading times and reading requests. They answered that AI helped increase diagnostic accuracy and were more positive about AI after actual usage.

CONCLUSION

Actual adaptation of AI for daily chest radiographs received overall positive feedback from clinicians and radiologists in this hospital-wide survey. Participating doctors preferred to use AI and regarded it more favorably after actual working with the AI-based software in daily clinical practice.

Conventional Versus Artificial Intelligence-Assisted Interpretation of Chest Radiographs in Patients With Acute Respiratory Symptoms in Emergency Department: A Pragmatic Randomized Clinical Trial

OBJECTIVE

It is unknown whether artificial intelligence-based computer-aided detection (AI-CAD) can enhance the accuracy of chest radiograph (CR) interpretation in real-world clinical practice. We aimed to compare the accuracy of CR interpretation assisted by AI-CAD to that of conventional interpretation in patients who presented to the emergency department (ED) with acute respiratory symptoms using a pragmatic randomized controlled trial.

MATERIALS AND METHODS

Patients who underwent CRs for acute respiratory symptoms at the ED of a tertiary referral institution were randomly assigned to intervention group (with assistance from an AI-CAD for CR interpretation) or control group (without AI assistance). Using a commercial AI-CAD system (Lunit INSIGHT CXR, version 2.0.2.0; Lunit Inc.). Other clinical practices were consistent with standard procedures. Sensitivity and false-positive rates of CR interpretation by duty trainee radiologists for identifying acute thoracic diseases were the primary and secondary outcomes, respectively. The reference standards for acute thoracic disease were established based on a review of the patient's medical record at least 30 days after the ED visit.

RESULTS

We randomly assigned 3576 participants to either the intervention group (1761 participants; mean age ± standard deviation, 65 ± 17 years; 978 males; acute thoracic disease in 472 participants) or the control group (1815 participants; 64 ± 17 years; 988 males; acute thoracic disease in 491 participants). The sensitivity (67.2% [317/472] in the intervention group vs. 66.0% [324/491] in the control group; odds ratio, 1.02 [95% confidence interval, 0.70-1.49]; P = 0.917) and false-positive rate (19.3% [249/1289] vs. 18.5% [245/1324]; odds ratio, 1.00 [95% confidence interval, 0.79-1.26]; P = 0.985) of CR interpretation by duty radiologists were not associated with the use of AI-CAD.

CONCLUSION

AI-CAD did not improve the sensitivity and false-positive rate of CR interpretation for diagnosing acute thoracic disease in patients with acute respiratory symptoms who presented to the ED.

Artificial Intelligence for Cardiothoracic Imaging: Overview of Current and Emerging Applications

Artificial intelligence algorithms can learn by assimilating information from large datasets in order to decipher complex associations, identify previously undiscovered pathophysiological states, and construct prediction models. There has been tremendous interest and increased incorporation of artificial intelligence into various industries, including healthcare. As a result, there has been an exponential rise in the number of research articles and industry participants producing models intended for a variety of applications in medical imaging, which can be challenging to navigate for radiologists. In thoracic imaging, multiple applications are being evaluated for chest radiography and computed tomography and include applications for lung nodule evaluation and cancer imaging, quantifying diffuse lung disorders, and cardiac imaging, to name a few. This review aims to provide an overview of current clinical AI models, focusing on the most common clinical applications of AI in cardiothoracic imaging.

Comparison of Chest Radiograph Captions Based on Natural Language Processing vs Completed by Radiologists

IMPORTANCE

Artificial intelligence (AI) can interpret abnormal signs in chest radiography (CXR) and generate captions, but a prospective study is needed to examine its practical value.

OBJECTIVE

To prospectively compare natural language processing (NLP)-generated CXR captions and the diagnostic findings of radiologists.

DESIGN, SETTING, AND PARTICIPANTS

A multicenter diagnostic study was conducted. The training data set included CXR images and reports retrospectively collected from February 1, 2014, to February 28, 2018. The retrospective test data set included consecutive images and reports from April 1 to July 31, 2019. The prospective test data set included consecutive images and reports from May 1 to September 30, 2021.

EXPOSURES

A bidirectional encoder representation from a transformers model was used to extract language entities and relationships from unstructured CXR reports to establish 23 labels of abnormal signs to train convolutional neural networks. The participants in the prospective test group were randomly assigned to 1 of 3 different caption generation models: a normal template, NLP-generated captions, and rule-based captions based on convolutional neural networks. For each case, a resident drafted the report based on the randomly assigned captions and an experienced radiologist finalized the report blinded to the original captions. A total of 21 residents and 19 radiologists were involved.

MAIN OUTCOMES AND MEASURES

Time to write reports based on different caption generation models.

RESULTS

The training data set consisted of 74 082 cases (39 254 [53.0%] women; mean [SD] age, 50.0 [17.1] years). In the retrospective (n = 8126; 4345 [53.5%] women; mean [SD] age, 47.9 [15.9] years) and prospective (n = 5091; 2416 [47.5%] women; mean [SD] age, 45.1 [15.6] years) test data sets, the mean (SD) area under the curve of abnormal signs was 0.87 (0.11) in the retrospective data set and 0.84 (0.09) in the prospective data set. The residents' mean (SD) reporting time using the NLP-generated model was 283 (37) seconds-significantly shorter than the normal template (347 [58] seconds; P < .001) and the rule-based model (296 [46] seconds; P < .001). The NLP-generated captions showed the highest similarity to the final reports with a mean (SD) bilingual evaluation understudy score of 0.69 (0.24)-significantly higher than the normal template (0.37 [0.09]; P < .001) and the rule-based model (0.57 [0.19]; P < .001).

CONCLUSIONS AND RELEVANCE

In this diagnostic study of NLP-generated CXR captions, prior information provided by NLP was associated with greater efficiency in the reporting process, while maintaining good consistency with the findings of radiologists.

Comparison of single- and dual-energy CT combined with artificial intelligence for the diagnosis of pulmonary nodules

AIM

To explore the efficiency of single- and dual-energy computed tomography (CT) with artificial intelligence (AI) for the diagnosis of pulmonary nodules.

MATERIALS AND METHODS

In a prospective study, 682 patients undergoing a chest CT examination using a dual-energy system were divided randomly into two groups: single-energy mode (group S, n=341) and dual-energy mode (group D, n=341). CT images were first analysed automatically with the AI pulmonary nodule-detection software. CT features including nodule number, lesion size, and nodule type were then analysed by experienced radiologists to establish a reference diagnosis. Subsequently, the accuracy, sensitivity, false-positive rate, and miss rate of AI were calculated. Additionally, image quality and radiation dose were also compared between the two groups.

RESULTS

The contrast-to-noise ratio data suggested that the image quality of group D was superior to that of group S (0.16 ± 0.10 versus 0.00 ± 0.17), and the radiation dose of group D was lower than that of group S (0.32 ± 0.10 versus 0.62 ± 0.11 mSv.cm). Compared to group S, group D exhibited a significantly higher sensitivity and lower accuracy for nodule identification, size classification, and nodule type (all p<0.05, except for 5-10 mm and calcified nodules).

CONCLUSIONS

Compared with single-energy CT, dual-energy CT may significantly improve the sensitivity of AI for the diagnosis of pulmonary nodules and is practical for the screening of pulmonary nodules in a large population. In addition, dual-energy CT examination demonstrates improved image quality and is associated with reduced exposure to ionising radiation, but its accuracy is poorer.

Distinguishing nontuberculous mycobacterial lung disease and Mycobacterium tuberculosis lung disease on X-ray images using deep transfer learning

BACKGROUND

Nontuberculous mycobacterial lung disease (NTM-LD) and Mycobacterium tuberculosis lung disease (MTB-LD) have similar clinical characteristics. Therefore, NTM-LD is sometimes incorrectly diagnosed with MTB-LD and treated incorrectly. To solve these difficulties, we aimed to distinguish the two diseases in chest X-ray images using deep learning technology, which has been used in various fields recently.

METHODS

We retrospectively collected chest X-ray images from 3314 patients infected with Mycobacterium tuberculosis (MTB) or nontuberculosis mycobacterium (NTM). After selecting the data according to the diagnostic criteria, various experiments were conducted to create the optimal deep learning model. A performance comparison was performed with the radiologist. Additionally, the model performance was verified using newly collected MTB-LD and NTM-LD patient data.

RESULTS

Among the implemented deep learning models, the ensemble model combining EfficientNet B4 and ResNet 50 performed the best in the test data. Also, the ensemble model outperformed the radiologist on all evaluation metrics. In addition, the accuracy of the ensemble model was 0.85 for MTB-LD and 0.78 for NTM-LD on an additional validation dataset consisting of newly collected patients.

CONCLUSIONS

In previous studies, it was known that it was difficult to distinguish between MTB-LD and NTM-LD in chest X-ray images, but we have successfully distinguished the two diseases using deep learning methods. This study has the potential to aid clinical decisions if the two diseases need to be differentiated.

AI-Aided Disease Prediction in Visualized Medicine

Artificial intelligence (AI) is playing a vitally important role in promoting the revolution of future technology. Healthcare is one of the promising applications in AI, which covers medical imaging, diagnosis, robotics, disease prediction, pharmacy, health management, and hospital management. Numbers of achievements that made in these fields overturn every aspect in traditional healthcare system. Therefore, to understand the state-of-art AI in healthcare, as well as the chances and obstacles in its development, the applications of AI in disease detection and outlook and the future trends of AI-aided disease prediction were discussed in this chapter.

Development and validation of a 3D-convolutional neural network model based on chest CT for differentiating active pulmonary tuberculosis from community-acquired pneumonia

PURPOSE

To develop and validate a 3D-convolutional neural network (3D-CNN) model based on chest CT for differentiating active pulmonary tuberculosis (APTB) from community-acquired pneumonia (CAP).

MATERIALS AND METHODS

Chest CT images of APTB and CAP patients diagnosed in two imaging centers (n = 432 in center A and n = 61 in center B) were collected retrospectively. The data in center A were divided into training, validation and internal test sets, and the data in center B were used as an external test set. A 3D-CNN was built using Keras deep learning framework. After the training, the 3D-CNN selected the model with the highest accuracy in the validation set as the optimal model, which was applied to the two test sets in centers A and B. In addition, the two test sets were independently diagnosed by two radiologists. The 3D-CNN optimal model was compared with the discrimination, calibration and net benefit of the two radiologists in differentiating APTB from CAP using chest CT images.

RESULTS

The accuracy of the 3D-CNN optimal model was 0.989 and 0.934 with the internal and external test set, respectively. The area-under-the-curve values with the 3D-CNN model in the two test sets were statistically higher than that of the two radiologists (all P < 0.05), and there was a high calibration degree. The decision curve analysis showed that the 3D-CNN optimal model had significantly higher net benefit for patients than the two radiologists.

CONCLUSIONS

3D-CNN has high classification performance in differentiating APTB from CAP using chest CT images. The application of 3D-CNN provides a new automatic and rapid diagnosis method for identifying patients with APTB from CAP using chest CT images.

Diagnostic Accuracy of the Artificial Intelligence Methods in Medical Imaging for Pulmonary Tuberculosis: A Systematic Review and Meta-Analysis

Tuberculosis (TB) remains one of the leading causes of death among infectious diseases worldwide. Early screening and diagnosis of pulmonary tuberculosis (PTB) is crucial in TB control, and tend to benefit from artificial intelligence. Here, we aimed to evaluate the diagnostic efficacy of a variety of artificial intelligence methods in medical imaging for PTB. We searched MEDLINE and Embase with the OVID platform to identify trials published update to November 2022 that evaluated the effectiveness of artificial-intelligence-based software in medical imaging of patients with PTB. After data extraction, the quality of studies was assessed using quality assessment of diagnostic accuracy studies 2 (QUADAS-2). Pooled sensitivity and specificity were estimated using a bivariate random-effects model. In total, 3987 references were initially identified and 61 studies were finally included, covering a wide range of 124,959 individuals. The pooled sensitivity and the specificity were 91% (95% confidence interval (CI), 89-93%) and 65% (54-75%), respectively, in clinical trials, and 94% (89-96%) and 95% (91-97%), respectively, in model-development studies. These findings have demonstrated that artificial-intelligence-based software could serve as an accurate tool to diagnose PTB in medical imaging. However, standardized reporting guidance regarding AI-specific trials and multicenter clinical trials is urgently needed to truly transform this cutting-edge technology into clinical practice.

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.

Artificial Intelligence in Lung Imaging

Recently, interest and advances in artificial intelligence (AI) including deep learning for medical images have surged. As imaging plays a major role in the assessment of pulmonary diseases, various AI algorithms have been developed for chest imaging. Some of these have been approved by governments and are now commercially available in the marketplace. In the field of chest radiology, there are various tasks and purposes that are suitable for AI: initial evaluation/triage of certain diseases, detection and diagnosis, quantitative assessment of disease severity and monitoring, and prediction for decision support. While AI is a powerful technology that can be applied to medical imaging and is expected to improve our current clinical practice, some obstacles must be addressed for the successful implementation of AI in workflows. Understanding and becoming familiar with the current status and potential clinical applications of AI in chest imaging, as well as remaining challenges, would be essential for radiologists and clinicians in the era of AI. This review introduces the potential clinical applications of AI in chest imaging and also discusses the challenges for the implementation of AI in daily clinical practice and future directions in chest imaging.

Ensemble Technique Coupled with Deep Transfer Learning Framework for Automatic Detection of Tuberculosis from Chest X-ray Radiographs

Tuberculosis (TB) is an infectious disease affecting humans' lungs and is currently ranked the 13th leading cause of death globally. Due to advancements in technology and the availability of medical datasets, automatic analysis and classification of chest X-rays (CXRs) into TB and non-TB can be a reliable alternative for early TB screening. We propose an automatic TB detection system using advanced deep learning (DL) models. A substantial part of a CXR image is dark, with no relevant information for diagnosis and potentially confusing DL models. In this work, the U-Net model extracts the region of interest from CXRs and the segmented images are fed to the DL models for feature extraction. Eight different convolutional neural networks (CNN) models are employed in our experiments, and their classification performance is compared based on three publicly available CXR datasets. The U-Net model achieves segmentation accuracy of 98.58%, intersection over union (IoU) of 93.10, and a Dice coefficient score of 96.50. Our proposed stacked ensemble algorithm performed better by achieving accuracy, sensitivity, and specificity values of 98.38%, 98.89%, and 98.70%, respectively. Experimental results confirm that segmented lung CXR images with ensemble learning produce a better result than un-segmented lung CXR images.

Over-the-Counter Breast Cancer Classification Using Machine Learning and Patient Registration Records

This study aims to determine the feasibility of machine learning (ML) and patient registration record to be utilised to develop an over-the-counter (OTC) screening model for breast cancer risk estimation. Data were retrospectively collected from women who came to the Hospital Universiti Sains Malaysia, Malaysia for breast-related problems. Eight ML models were used: k-nearest neighbour (kNN), elastic-net logistic regression, multivariate adaptive regression splines, artificial neural network, partial least square, random forest, support vector machine (SVM), and extreme gradient boosting. Features utilised for the development of the screening models were limited to information in the patient registration form. The final model was evaluated in terms of performance across a mammographic density. Additionally, the feature importance of the final model was assessed using the model agnostic approach. kNN had the highest Youden J index, precision, and PR-AUC, while SVM had the highest F2 score. The kNN model was selected as the final model. The model had a balanced performance in terms of sensitivity, specificity, and PR-AUC across the mammographic density groups. The most important feature was the age at examination. In conclusion, this study showed that ML and patient registration information are feasible to be used as the OTC screening model for breast cancer.

Deep Learning Techniques to Diagnose Lung Cancer

Medical imaging tools are essential in early-stage lung cancer diagnostics and the monitoring of lung cancer during treatment. Various medical imaging modalities, such as chest X-ray, magnetic resonance imaging, positron emission tomography, computed tomography, and molecular imaging techniques, have been extensively studied for lung cancer detection. These techniques have some limitations, including not classifying cancer images automatically, which is unsuitable for patients with other pathologies. It is urgently necessary to develop a sensitive and accurate approach to the early diagnosis of lung cancer. Deep learning is one of the fastest-growing topics in medical imaging, with rapidly emerging applications spanning medical image-based and textural data modalities. With the help of deep learning-based medical imaging tools, clinicians can detect and classify lung nodules more accurately and quickly. This paper presents the recent development of deep learning-based imaging techniques for early lung cancer detection.

Role of an Automated Deep Learning Algorithm for Reliable Screening of Abnormality in Chest Radiographs: A Prospective Multicenter Quality Improvement Study

In medical practice, chest X-rays are the most ubiquitous diagnostic imaging tests. However, the current workload in extensive health care facilities and lack of well-trained radiologists is a significant challenge in the patient care pathway. Therefore, an accurate, reliable, and fast computer-aided diagnosis (CAD) system capable of detecting abnormalities in chest X-rays is crucial in improving the radiological workflow. In this prospective multicenter quality-improvement study, we have evaluated whether artificial intelligence (AI) can be used as a chest X-ray screening tool in real clinical settings. Methods: A team of radiologists used the AI-based chest X-ray screening tool (qXR) as a part of their daily reporting routine to report consecutive chest X-rays for this prospective multicentre study. This study took place in a large radiology network in India between June 2021 and March 2022. Results: A total of 65,604 chest X-rays were processed during the study period. The overall performance of AI achieved in detecting normal and abnormal chest X-rays was good. The high negatively predicted value (NPV) of 98.9% was achieved. The AI performance in terms of area under the curve (AUC), NPV for the corresponding subabnormalities obtained were blunted CP angle (0.97, 99.5%), hilar dysmorphism (0.86, 99.9%), cardiomegaly (0.96, 99.7%), reticulonodular pattern (0.91, 99.9%), rib fracture (0.98, 99.9%), scoliosis (0.98, 99.9%), atelectasis (0.96, 99.9%), calcification (0.96, 99.7%), consolidation (0.95, 99.6%), emphysema (0.96, 99.9%), fibrosis (0.95, 99.7%), nodule (0.91, 99.8%), opacity (0.92, 99.2%), pleural effusion (0.97, 99.7%), and pneumothorax (0.99, 99.9%). Additionally, the turnaround time (TAT) decreased by about 40.63% from pre-qXR period to post-qXR period. Conclusions: The AI-based chest X-ray solution (qXR) screened chest X-rays and assisted in ruling out normal patients with high confidence, thus allowing the radiologists to focus more on assessing pathology on abnormal chest X-rays and treatment pathways.

Diagnosis of Active Pulmonary Tuberculosis and Community Acquired Pneumonia Using Convolution Neural Network Based on Transfer Learning

RATIONALE AND OBJECTIVES

A convolutional neural network (CNN) model for the diagnosis of active pulmonary tuberculosis (APTB) and community-acquired pneumonia (CAP) using chest radiographs (CRs) was constructed and verified based on transfer learning.

MATERIALS AND METHODS

CRs of 1247 APTB cases, 1488 CAP cases and 1247 normal cases were collected. All CRs were randomly divided into training set (1992 cases), validation set (1194 cases) and test set (796 cases) by stratified sampling in 5:3:2 radio. After normalization of CRs, the convolution base of pre-trained CNN (VGG16) model on ImageNet dataset was used to extract features, and the grid search was used to determine the optimal classifier module, which was added to the convolution base for transfer learning. After the training, the model with the highest accuracy of the validation set was selected as the optimal model to verify in the test set and calculate the accuracy of the model.

RESULTS

The accuracy of validation set in the 63rd epochs was the highest, which was 0.9430, and the corresponding Categorical crossentropy was 0.1742. The accuracy of the training set was 0.9428, and the Categorical crossentropy was 0.1545. When the optimal model was applied to the test set, the accuracy was 0.9447, and the Categorical crossentropy was 0.1929.

CONCLUSION

The transfer learning-based CNN model has good classification performance in the diagnosis of APTB, CAP and normal patients using CRs.

Development of Deep Learning-based Automatic Scan Range Setting Model for Lung Cancer Screening Low-dose CT Imaging

RATIONALE AND OBJECTIVES

To develop an automatic setting of a deep learning-based system for detecting low-dose computed tomography (CT) lung cancer screening scan range and compare its efficiency with the radiographer's performance.

MATERIALS AND METHODS

This retrospective study was performed using 1984 lung cancer screening low-dose CT scans obtained between November 2019 and May 2020. Among 1984 CT scans, 600 CT scans were considered suitable for an observational study to explore the relationship between the scout landmarks and the actual lung boundaries. Further, 1144 CT scans data set was used for the development of a deep learning-based algorithm. This data set was split into an 8:2 ratio divided into a training set (80%, n = 915) and a validation set (20%, n = 229). The performance of the deep learning algorithm was evaluated in the test set (n = 240) using actual lung boundaries and radiographers' scan ranges.

RESULTS

The mean differences between the upper and lower boundaries of the deep learning-based algorithm and the actual lung boundaries were 4.72 ± 3.15 mm and 16.50 ± 14.06 mm, respectively. The accuracy and over-scanning of the scan ranges generated by the system were 97.08% (233/240) and 0% (0/240) for the upper boundary, and 96.25% (231/240) and 29.58% (71/240) for the lower boundary.

CONCLUSION

The developed deep learning-based algorithm system can effectively predict lung cancer screening low-dose CT scan range with high accuracy using only the frontal scout.

Diagnosis of fetal total anomalous pulmonary venous connection based on the post-left atrium space ratio using artificial intelligence

OBJECTIVE

To explore whether the post-left atrium space (PLAS) ratio would be useful for prenatal diagnosis of total anomalous pulmonary venous connection (TAPVC) using echocardiography and artificial intelligence.

METHODS

We retrospectively included 642 frames of four-chamber view from 319 fetuses (32 with TAPVC and 287 without TAPVC) in end-systolic and end-diastolic periods with multiple apex directions. The average gestational age was 25.6±2.7 weeks. No other cardiac or extracardiac malformations were observed. The dataset was divided into a training set (n=540; 48 with TAPVC and 492 without TAPVC) and test set (n=102; 20 with TAPVC and 82 without TAPVC). The PLAS ratio was defined as the ratio of the epicardium-descending aortic distance to the center of the heart-descending aortic distance. Supervised learning was used in DeepLabv3+, FastFCN, PSPNet, and DenseASPP segmentation models. The area under the curve (AUC) was used on the test set.

RESULTS

Expert annotations showed that this ratio was not related to the period or apex direction. It was higher in the TAPVC group than in the control group detected by expert and the four models. The AUC of expert annotations, DeepLabv3+, FastFCN, PSPNet, and DenseASPP were 0.977, 0.941, 0.925, 0.856, and 0.887, respectively.

CONCLUSION

Segmentation models achieve good diagnostic accuracy for TAPVC based on the PLAS ratio. This article is protected by copyright. All rights reserved.

The Application of Artificial Intelligence in the Diagnosis and Drug Resistance Prediction of Pulmonary Tuberculosis

With the increasing incidence and mortality of pulmonary tuberculosis, in addition to tough and controversial disease management, time-wasting and resource-limited conventional approaches to the diagnosis and differential diagnosis of tuberculosis are still awkward issues, especially in countries with high tuberculosis burden and backwardness. In the meantime, the climbing proportion of drug-resistant tuberculosis poses a significant hazard to public health. Thus, auxiliary diagnostic tools with higher efficiency and accuracy are urgently required. Artificial intelligence (AI), which is not new but has recently grown in popularity, provides researchers with opportunities and technical underpinnings to develop novel, precise, rapid, and automated implements for pulmonary tuberculosis care, including but not limited to tuberculosis detection. In this review, we aimed to introduce representative AI methods, focusing on deep learning and radiomics, followed by definite descriptions of the state-of-the-art AI models developed using medical images and genetic data to detect pulmonary tuberculosis, distinguish the infection from other pulmonary diseases, and identify drug resistance of tuberculosis, with the purpose of assisting physicians in deciding the appropriate therapeutic schedule in the early stage of the disease. We also enumerated the challenges in maximizing the impact of AI in this field such as generalization and clinical utility of the deep learning models.

Machine learning in the loop for tuberculosis diagnosis support

The use of machine learning (ML) for diagnosis support has advanced in the field of health. In the present paper, the results of studying ML techniques in a tuberculosis diagnosis loop in a scenario of limited resources are presented. Data are analyzed using a tuberculosis (TB) therapy program at a health institution in a main city of a developing country using five ML models. Logistic regression, classification trees, random forest, support vector machines, and artificial neural networks are trained under physician supervision following physicians' typical daily work. The models are trained on seven main variables collected when patients arrive at the facility. Additionally, the variables applied to train the models are analyzed, and the models' advantages and limitations are discussed in the context of the automated ML techniques. The results show that artificial neural networks obtain the best results in terms of accuracy, sensitivity, and area under the receiver operating curve. These results represent an improvement over smear microscopy, which is commonly used techniques to detect TB for special cases. Findings demonstrate that ML in the TB diagnosis loop can be reinforced with available data to serve as an alternative diagnosis tool based on data processing in places where the health infrastructure is limited.

Where Is the Artificial Intelligence Applied in Dentistry? Systematic Review and Literature Analysis

This literature research had two main objectives. The first objective was to quantify how frequently artificial intelligence (AI) was utilized in dental literature from 2011 until 2021. The second objective was to distinguish the focus of such publications; in particular, dental field and topic. The main inclusion criterium was an original article or review in English focused on dental utilization of AI. All other types of publications or non-dental or non-AI-focused were excluded. The information sources were Web of Science, PubMed, Scopus, and Google Scholar, queried on 19 April 2022. The search string was “artificial intelligence” AND (dental OR dentistry OR tooth OR teeth OR dentofacial OR maxillofacial OR orofacial OR orthodontics OR endodontics OR periodontics OR prosthodontics). Following the removal of duplicates, all remaining publications were returned by searches and were screened by three independent operators to minimize the risk of bias. The analysis of 2011–2021 publications identified 4413 records, from which 1497 were finally selected and calculated according to the year of publication. The results confirmed a historically unprecedented boom in AI dental publications, with an average increase of 21.6% per year over the last decade and a 34.9% increase per year over the last 5 years. In the achievement of the second objective, qualitative assessment of dental AI publications since 2021 identified 1717 records, with 497 papers finally selected. The results of this assessment indicated the relative proportions of focal topics, as follows: radiology 26.36%, orthodontics 18.31%, general scope 17.10%, restorative 12.09%, surgery 11.87% and education 5.63%. The review confirms that the current use of artificial intelligence in dentistry is concentrated mainly around the evaluation of digital diagnostic methods, especially radiology; however, its implementation is expected to gradually penetrate all parts of the profession.