-
Frontiers in Public Health 2022Artificial intelligence has far surpassed previous related technologies in image recognition and is increasingly used in medical image analysis. We aimed to explore the... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Artificial intelligence has far surpassed previous related technologies in image recognition and is increasingly used in medical image analysis. We aimed to explore the diagnostic accuracy of the models based on deep learning or radiomics for lung cancer staging.
METHODS
Studies were systematically reviewed using literature searches from PubMed, EMBASE, Web of Science, and Wanfang Database, according to PRISMA guidelines. Studies about the diagnostic accuracy of radiomics and deep learning, including the identifications of lung cancer, tumor types, malignant lung nodules and lymph node metastase, were included. After identifying the articles, the methodological quality was assessed using the QUADAS-2 checklist. We extracted the characteristic of each study; the sensitivity, specificity, and AUROC for lung cancer diagnosis were summarized for subgroup analysis.
RESULTS
The systematic review identified 19 eligible studies, of which 14 used radiomics models and 5 used deep learning models. The pooled AUROC of 7 studies to determine whether patients had lung cancer was 0.83 (95% CI 0.78-0.88). The pooled AUROC of 9 studies to determine whether patients had NSCLC was 0.78 (95% CI 0.73-0.83). The pooled AUROC of the 6 studies that determined patients had malignant lung nodules was 0.79 (95% CI 0.77-0.82). The pooled AUROC of the other 6 studies that determined whether patients had lymph node metastases was 0.74 (95% CI 0.66-0.82).
CONCLUSION
The models based on deep learning or radiomics have the potential to improve diagnostic accuracy for lung cancer staging.
SYSTEMATIC REVIEW REGISTRATION
https://inplasy.com/inplasy-2022-3-0167/, identifier: INPLASY202230167.
Topics: Artificial Intelligence; Deep Learning; Humans; Lung; Lung Neoplasms; Neoplasm Staging
PubMed: 35923964
DOI: 10.3389/fpubh.2022.938113 -
Insights Into Imaging Sep 2023Health systems worldwide are implementing lung cancer screening programmes to identify early-stage lung cancer and maximise patient survival. Volumetry is recommended... (Review)
Review
Health systems worldwide are implementing lung cancer screening programmes to identify early-stage lung cancer and maximise patient survival. Volumetry is recommended for follow-up of pulmonary nodules and outperforms other measurement methods. However, volumetry is known to be influenced by multiple factors. The objectives of this systematic review (PROSPERO CRD42022370233) are to summarise the current knowledge regarding factors that influence volumetry tools used in the analysis of pulmonary nodules, assess for significant clinical impact, identify gaps in current knowledge and suggest future research. Five databases (Medline, Scopus, Journals@Ovid, Embase and Emcare) were searched on the 21st of September, 2022, and 137 original research studies were included, explicitly testing the potential impact of influencing factors on the outcome of volumetry tools. The summary of these studies is tabulated, and a narrative review is provided. A subset of studies (n = 16) reporting clinical significance were selected, and their results were combined, if appropriate, using meta-analysis. Factors with clinical significance include the segmentation algorithm, quality of the segmentation, slice thickness, the level of inspiration for solid nodules, and the reconstruction algorithm and kernel in subsolid nodules. Although there is a large body of evidence in this field, it is unclear how to apply the results from these studies in clinical practice as most studies do not test for clinical relevance. The meta-analysis did not improve our understanding due to the small number and heterogeneity of studies testing for clinical significance. CRITICAL RELEVANCE STATEMENT: Many studies have investigated the influencing factors of pulmonary nodule volumetry, but only 11% of these questioned their clinical relevance in their management. The heterogeneity among these studies presents a challenge in consolidating results and clinical application of the evidence. KEY POINTS: • Factors influencing the volumetry of pulmonary nodules have been extensively investigated. • Just 11% of studies test clinical significance (wrongly diagnosing growth). • Nodule size interacts with most other influencing factors (especially for smaller nodules). • Heterogeneity among studies makes comparison and consolidation of results challenging. • Future research should focus on clinical applicability, screening, and updated technology.
PubMed: 37741928
DOI: 10.1186/s13244-023-01480-z -
Diagnostics (Basel, Switzerland) Nov 2019The aim of this study was to systematically review the performance of deep learning technology in detecting and classifying pulmonary nodules on computed tomography (CT)... (Review)
Review
The Performance of Deep Learning Algorithms on Automatic Pulmonary Nodule Detection and Classification Tested on Different Datasets That Are Not Derived from LIDC-IDRI: A Systematic Review.
The aim of this study was to systematically review the performance of deep learning technology in detecting and classifying pulmonary nodules on computed tomography (CT) scans that were not from the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) database. Furthermore, we explored the difference in performance when the deep learning technology was applied to test datasets different from the training datasets. Only peer-reviewed, original research articles utilizing deep learning technology were included in this study, and only results from testing on datasets other than the LIDC-IDRI were included. We searched a total of six databases: EMBASE, PubMed, Cochrane Library, the Institute of Electrical and Electronics Engineers, Inc. (IEEE), Scopus, and Web of Science. This resulted in 1782 studies after duplicates were removed, and a total of 26 studies were included in this systematic review. Three studies explored the performance of pulmonary nodule detection only, 16 studies explored the performance of pulmonary nodule classification only, and 7 studies had reports of both pulmonary nodule detection and classification. Three different deep learning architectures were mentioned amongst the included studies: convolutional neural network (CNN), massive training artificial neural network (MTANN), and deep stacked denoising autoencoder extreme learning machine (SDAE-ELM). The studies reached a classification accuracy between 68-99.6% and a detection accuracy between 80.6-94%. Performance of deep learning technology in studies using different test and training datasets was comparable to studies using same type of test and training datasets. In conclusion, deep learning was able to achieve high levels of accuracy, sensitivity, and/or specificity in detecting and/or classifying nodules when applied to pulmonary CT scans not from the LIDC-IDRI database.
PubMed: 31795409
DOI: 10.3390/diagnostics9040207 -
Wideochirurgia I Inne Techniki... Dec 2023The diagnosis of pulmonary nodules (PNs) has traditionally relied on computed tomography (CT)-guided biopsy. To reduce radiation exposure, low-dose CT-guided PN biopsy...
INTRODUCTION
The diagnosis of pulmonary nodules (PNs) has traditionally relied on computed tomography (CT)-guided biopsy. To reduce radiation exposure, low-dose CT-guided PN biopsy has been employed.
AIM
This meta-analysis aimed at evaluating the efficacy and safety of low-dose CT-guided biopsy in the diagnosis of PNs.
MATERIAL AND METHODS
PubMed, Web of Science, and Wanfang were searched for relevant articles until June 2023. Comparing low-dose CT to normal-dose CT, we considered factors such as diagnostic yield, diagnostic accuracy, biopsy process time, dose-length product (DLP) value, the frequency of pneumothorax and pulmonary bleeding, and the frequency with which complications necessitated the placement of a chest tube.
RESULTS
This meta-analysis included data from a total of 6 investigations. There was a total of 459 patients who had a CT-guided PN biopsy performed at a low dosage, and 384 patients who had a normal-dose CT-guided PN biopsy. There were no statistically significant differences between the low-dose CT and normal-dose CT groups in terms of diagnostic accuracy (p = 0.08), diagnostic yield (p = 0.55), biopsy procedure duration (p = 0.30), pneumothorax (p = 0.61), pulmonary hemorrhage (p = 0.29), or complications requiring a chest tube (p = 0.48). Low-dose CT patients obtained a DLP that was 91% lower than those in the standard-dose CT group (p = 0.01). According to Egger's test, there is a significant possibility of publication bias in DLP (p = 0.034).
CONCLUSIONS
The diagnostic and safety results of low-dose CT-driven PN biopsy are equivalent to those of the standard one, although patients are much less exposed to radiation.
PubMed: 38239580
DOI: 10.5114/wiitm.2023.131563 -
Systematic Reviews Dec 2021Solitary pulmonary nodule (SPN) is a common finding in routine clinical practice when performing chest imaging tests. The vast majority of these nodules are benign, and...
BACKGROUND
Solitary pulmonary nodule (SPN) is a common finding in routine clinical practice when performing chest imaging tests. The vast majority of these nodules are benign, and only a small proportion are malignant. The application of predictive models of nodule malignancy in routine clinical practice would help to achieve better diagnostic management of SPN. The present systematic review was carried out with the purpose of critically assessing studies aimed at developing predictive models of solitary pulmonary nodule (SPN) malignancy from SPN incidentally detected in routine clinical practice.
METHODS
We performed a search of available scientific literature until October 2020 in Pubmed, SCOPUS and Cochrane Central databases. The inclusion criteria were observational studies carried out in low-risk population from 35 years old onwards aimed at constructing predictive models of malignancy of pulmonary solitary nodule detected incidentally in routine clinical practice. Studies had to be published in peer-reviewed journals, either in Spanish, Portuguese or English. Exclusion criteria were non-human studies, or predictive models based in high-risk populations, or models based on computational approaches. Exclusion criteria were non-human studies, or predictive models based in high-risk populations, or models based on computational approaches (such as radiomics). We used The Transparent Reporting of a multivariable Prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement, to describe the type of predictive model included in each study, and The Prediction model Risk Of Bias ASsessment Tool (PROBAST) to evaluate the quality of the selected articles.
RESULTS
A total of 186 references were retrieved, and after applying the exclusion/inclusion criteria, 15 articles remained for the final review. All studies analysed clinical and radiological variables. The most frequent independent predictors of SPN malignancy were, in order of frequency, age, diameter, spiculated edge, calcification and smoking history. Variables such as race, SPN growth rate, emphysema, fibrosis, apical scarring and exposure to asbestos, uranium and radon were not analysed by the majority of the studies. All studies were classified as high risk of bias due to inadequate study designs, selection bias, insufficient population follow-up and lack of external validation, compromising their applicability for clinical practice.
CONCLUSIONS
The studies included have been shown to have methodological weaknesses compromising the clinical applicability of the evaluated SPN malignancy predictive models and their potential influence on clinical decision-making for the SPN diagnostic management.
SYSTEMATIC REVIEW REGISTRATION
PROSPERO CRD42020161559.
Topics: Humans; Lung; Lung Neoplasms; Prognosis; Risk Factors; Solitary Pulmonary Nodule
PubMed: 34872592
DOI: 10.1186/s13643-021-01856-6 -
Rheumatology International May 2024Immunoglobulin G4-related disease (IgG4-RD) is a multisystem fibroinflammatory condition. A consistent feature of many cases is pulmonary infiltrates, or respiratory... (Review)
Review
Immunoglobulin G4-related disease (IgG4-RD) is a multisystem fibroinflammatory condition. A consistent feature of many cases is pulmonary infiltrates, or respiratory failure. This systematic literature review aims to summarise the pulmonary manifestations of IgG4-RD, including clinical outcomes and treatment. This review was registered on PROSPERO (CRD42023416410). Medline, Embase and Cochrane databases were searched for articles discussing IgG4-RD syndrome. Information was extracted on demographics, type and prevalence of pulmonary manifestations, treatment and clinical outcomes. Initially, after deduplication, 3123 articles were retrieved with 18 ultimately included. A pooled total of 724 patients with IgG4-RD were included, 68.6% male, mean age 59.4 years (SD 5.8) at disease onset. The most frequently described pulmonary manifestation was mediastinal lymphadenopathy (n = 186, 48.8%), followed by pulmonary nodules (n = 151, 39.6%) and broncho-vascular thickening (n = 85, 22.3%). Where treatment was reported, the majority of patients received glucocorticoids (n = 211, 93.4%). Other immunosuppressive therapy included cyclophosphamide (n = 31), azathioprine (n = 18), with mycophenolate mofetil (n = 6), rituximab (n = 6), methotrexate (n = 5) and other unspecified immunomodulators (50). Clinical outcomes were reported in 263 patients, where 196 patients had remission of their disease, 20 had relapse, 35 had stable disease, four had progression and eight patients died from complications of IgG4-RD. This systematic review summarises pulmonary manifestations, treatments and outcomes in patients with IgG4-RD. Pulmonary involvement in IgG4-RD is relatively common, leading to high levels of morbidity and mortality. Glucocorticoids remain the mainstay of treatment, but further work is required to explore the management of patients with pulmonary manifestations in association with IgG4-RD.
PubMed: 38769126
DOI: 10.1007/s00296-024-05611-7 -
Frontiers in Oncology 2022Establishing risk-based follow-up management strategies is crucial to the surveillance of subsolid pulmonary nodules (SSNs). However, the risk factors for SSN growth are...
BACKGROUND
Establishing risk-based follow-up management strategies is crucial to the surveillance of subsolid pulmonary nodules (SSNs). However, the risk factors for SSN growth are not currently clear. This study aimed to perform a systematic review and meta-analysis to identify clinical and CT features correlated with SSN growth.
METHODS
Relevant studies were retrieved from Web of Science, PubMed, Cochrane Library, and EMBASE. The correlations of clinical and CT features with SSN growth were pooled using a random-effects model or fixed-effects model depending on heterogeneity, which was examined by the test and test. Pooled odds ratio (OR) or pooled standardized mean differences (SMD) based on univariate analyses were calculated to assess the correlation of clinical and CT features with SSN growth. Pooled ORs based on multivariate analyses were calculated to find out independent risk factors to SSN growth. Subgroup meta-analysis was performed based on nodule consistency (pure ground-glass nodule (pGGN) and part-solid nodule (PSN). Publication bias was examined using funnel plots.
RESULTS
Nineteen original studies were included, consisting of 2444 patients and 3012 SSNs. The median/mean follow-up duration of these studies ranged from 24.2 months to 112 months. Significant correlations were observed between SSN growth and eighteen features. Male sex, history of lung cancer, nodule size > 10 mm, nodule consistency, and age > 65 years were identified as independent risk factors for SSN growth based on multivariate analyses results. Eight features, including male sex, smoking history, nodule size > 10 mm, larger nodule size, air bronchogram, higher mean CT attenuation, well-defined border, and lobulated margin were detected to be significantly correlated with pGGNs growth. Smoking history showed no significant correlation with pGGN growth based on the multivariate analysis results.
CONCLUSIONS
Eighteen clinical and CT features were identified to be correlated with SSN growth, among which male sex, history of lung cancer, nodule size > 10 mm, nodule consistency and age > 65 years were independent risk factors while history of lung cancer was not correlated with pGGN growth. These factors should be considered when making risk-based follow-up plans for SSN patients.
PubMed: 35860567
DOI: 10.3389/fonc.2022.929174 -
Academic Radiology Jan 2014To perform a meta-analysis to assess the diagnostic performance of the diffusion-weighted magnetic resonance imaging (DWI) technique in discrimination of benign and... (Meta-Analysis)
Meta-Analysis Review
RATIONALE AND OBJECTIVES
To perform a meta-analysis to assess the diagnostic performance of the diffusion-weighted magnetic resonance imaging (DWI) technique in discrimination of benign and malignant pulmonary nodules or masses.
MATERIALS AND METHODS
Data sources were studies published in PubMed, MEDLINE, EMBASE, Cochrane Library, and China National Knowledge Infrastructure databases from January 2001 to May 2013. Studies evaluating the diagnostic accuracy of DWI for benign/malignant discrimination of pulmonary nodules in English or Chinese language were considered for inclusion. Methodological quality was assessed by the quality assessment of diagnostic studies instrument. Sensitivities, specificities, predictive values, diagnostic odds ratios (DORs), and areas under the receiver operating characteristic curve (AUCs) were calculated. Potential threshold effect, heterogeneity, and publication bias were investigated. We also evaluated the clinical utility of DWI in diagnosis of lung lesions.
RESULTS
Seventeen studies comprising 855 malignant and 322 benign lesions were included in this meta-analysis. There was no significant threshold effect. Summary receiver operating characteristic curve showed that AUC was 0.909 (95% confidence interval [CI], 0.862-0.931). Pooled weighted estimates of sensitivity, specificity, positive likelihood ratio (PLR), and negative likelihood ratio (NLR) were 0.828 (95% CI, 0.801-0.853), 0.801 (95% CI, 0.753-0.843), 4.01 (95% CI, 2.78-5.80), and 0.20 (95% CI, 0.15-0.27), respectively. Heterogeneity was found to have stemmed primarily from study design (retrospective or prospective study). Subgroup analysis showed that diagnostic performance (sensitivity, 0.88; 95% CI, 0.82-0.92 and specificity, 0.89; 95% CI, 0.79-0.96) of retrospectively designed studies was significantly higher than that of prospectively designed studies. The Deeks' funnel plot indicated the absence of publication bias.
CONCLUSIONS
With respect to the accuracy and DOR, DWI is useful for differentiation between malignant and benign pulmonary nodules or masses. Diagnostic test accuracy is not the be-all and end-all of diagnostic testing. Concerning PLR and NLR, DWI may not help to alter posttest probability compared to pretest probability to sufficiently alter physician's decision making. Future analyses should be conducted in large-scale, high-quality trials to evaluate its clinical value and establish standards of DWI measurement, analysis, and cutoff values of diagnosis.
Topics: Diagnosis, Differential; Diffusion Magnetic Resonance Imaging; Female; Humans; Lung Neoplasms; Male; Observer Variation; Reproducibility of Results; Sensitivity and Specificity; Solitary Pulmonary Nodule
PubMed: 24331261
DOI: 10.1016/j.acra.2013.09.019 -
International Journal of Environmental... Feb 2022Lung cancer (LC) represents the main cause of cancer-related deaths worldwide, especially because the majority of patients present with an advanced stage of the disease... (Review)
Review
Lung cancer (LC) represents the main cause of cancer-related deaths worldwide, especially because the majority of patients present with an advanced stage of the disease at the time of diagnosis. This systematic review describes the evidence behind screening results and the current guidelines available to manage lung nodules. This review was guided by the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. The following electronic databases were searched: PubMed, EMBASE, and the Web of Science. Five studies were included in the systematic review. The study cohort included 46,364 patients, and, in this case series, LC was detected in 9028 patients. Among the patients with detected LC, 1261 died of lung cancer, 3153 died of other types of cancers and 4614 died of other causes. This systematic review validates the use of CT in LC screening follow-ups, and bids for future integration and implementation of nodule management protocols to improve LC screening, avoid missed cancers and to reduce the number of unnecessary investigations.
Topics: Early Detection of Cancer; Humans; Lung; Lung Neoplasms; Mass Screening; Research
PubMed: 35206646
DOI: 10.3390/ijerph19042460 -
Clinical Imaging Oct 2022Common CT abnormalities of pulmonary aspergillosis represent a cavity with air-meniscus sign, nodule, mass, and consolidation having an angio-invasive pattern. This...
PURPOSE
Common CT abnormalities of pulmonary aspergillosis represent a cavity with air-meniscus sign, nodule, mass, and consolidation having an angio-invasive pattern. This study aims to conduct a systematic review and an individual patient-level image analysis of CT findings of COVID-19-associated pulmonary aspergillosis (CAPA).
METHODS
A systematic literature search was conducted to identify studies reporting CT findings of CAPA as of January 7, 2021. We summarized study-level clinical and CT findings of CAPA and collected individual patient CT images by inviting corresponding authors. The CT findings were categorized into four groups: group 1, typical appearance of COVID-19; group 2, indeterminate appearance of COVID-19; group 3, atypical for COVID-19 without cavities; and group 4, atypical for COVID-19 with cavities. In group 2, cases had only minor discrepant findings including solid nodules, isolated airspace consolidation with negligible ground-glass opacities, centrilobular micronodules, bronchial abnormalities, and cavities.
RESULTS
The literature search identified 89 patients from 25 studies, and we collected CT images from 35 CAPA patients (mean age 62.4 ± 14.6 years; 21 men): group 1, thirteen patients (37.1%); group 2, eight patients (22.9%); group 3, six patients (17.1%); and group 4, eight patients (22.9%). Eight of the 14 patients (57.1%) with an atypical appearance had bronchial abnormalities, whereas only one (7.1%) had an angio-invasive fungal pattern. In the study-level analysis, cavities were reported in 12 of 54 patients (22.2%).
CONCLUSION
CAPA can frequently manifest as COVID-19 pneumonia without common CT abnormalities of pulmonary aspergillosis. If abnormalities exist on CT images, CAPA may frequently accompany bronchial abnormalities.
Topics: Aged; COVID-19; Data Analysis; Humans; Lung; Male; Middle Aged; Pulmonary Aspergillosis; Tomography, X-Ray Computed
PubMed: 35908455
DOI: 10.1016/j.clinimag.2022.07.003