Dataset Information

Diagnostic Accuracy of a Convolutional Neural Network Assessment of Solitary Pulmonary Nodules Compared With PET With CT Imaging and Dynamic Contrast-Enhanced CT Imaging Using Unenhanced and Contrast-Enhanced CT Imaging.

ABSTRACT:

Background

Solitary pulmonary nodules (SPNs) measuring 8 to 30 mm in diameter require further workup to determine the likelihood of malignancy.

Research question

What is the diagnostic performance of a lung cancer prediction convolutional neural network (LCP-CNN) in SPNs using unenhanced and contrast-enhanced CT imaging compared with the current clinical workup?

Study design and methods

This was a post hoc analysis of the Single Pulmonary Nodule Investigation: Accuracy and Cost-Effectiveness of Dynamic Contrast Enhanced Computed Tomography in the Characterisation of Solitary Pulmonary Nodules trial, a prospective multicenter study comparing the diagnostic accuracy of dynamic contrast-enhanced (DCE) CT imaging with PET imaging in SPNs. The LCP-CNN was designed and validated in an external cohort. LCP-CNN-generated risk scores were created from the noncontrast and contrast-enhanced CT scan images from the DCE CT imaging. The gold standard was histologic analysis or 2 years of follow-up. The area under the receiver operating characteristic curves (AUC) were calculated using LCP-CNN score, maximum standardized uptake value, and DCE CT scan maximum enhancement and were compared using the DeLong test.

Results

Two hundred seventy participants (mean ± SD age, 68.3 ± 8.8 years; 49% women) underwent PET with CT scan imaging and DCE CT imaging with CT scan data available centrally for LCP-CNN analysis. The accuracy of the LCP-CNN on the noncontrast images (AUC, 0.83; 95% CI, 0.79-0.88) was superior to that of DCE CT imaging (AUC, 0.76; 95% CI, 0.69-0.82; P = .03) and equal to that of PET with CT scan imaging (AUC, 0.86; 95% CI, 0.81-0.90; P = .35). The presence of contrast resulted in a small reduction in diagnostic accuracy, with the AUC falling from 0.83 (95% CI, 0.79-0.88) on the noncontrast images to 0.80 to 0.83 after contrast (P < .05 for 240 s after contrast only).

Interpretation

An LCP-CNN algorithm provides an AUC equivalent to PET with CT scan imaging in the diagnosis of solitary pulmonary nodules.

Trial registration

ClinicalTrials.gov Identifier; No.: NCT02013063.

SUBMITTER: Weir-McCall JR

PROVIDER: S-EPMC9899635 | biostudies-literature | 2023 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Diagnostic Accuracy of a Convolutional Neural Network Assessment of Solitary Pulmonary Nodules Compared With PET With CT Imaging and Dynamic Contrast-Enhanced CT Imaging Using Unenhanced and Contrast-Enhanced CT Imaging.

Weir-McCall Jonathan R JR Debruyn Elise E Harris Scott S Qureshi Nagmi R NR Rintoul Robert C RC Gleeson Fergus V FV Gilbert Fiona J FJ

Chest 20220908 2

<h4>Background</h4>Solitary pulmonary nodules (SPNs) measuring 8 to 30 mm in diameter require further workup to determine the likelihood of malignancy.<h4>Research question</h4>What is the diagnostic performance of a lung cancer prediction convolutional neural network (LCP-CNN) in SPNs using unenhanced and contrast-enhanced CT imaging compared with the current clinical workup?<h4>Study design and methods</h4>This was a post hoc analysis of the Single Pulmonary Nodule Investigation: Accuracy and ...[more]

PMID: 36087795

Similar Datasets

Project description:To compare the diagnostic performance between plain CT-based model and plain plus contrast CT-based modelin the classification of malignancy for solitary solid pulmonary nodules. Between January 2012 and July 2021, 527 patients with pathologically confirmed solitary solid pulmonary nodules were collected at dual centers with similar CT examinations and scanning parameters. Before surgery, all patients underwent both plain and contrast-enhanced chest CT scans. Two clinical characteristics, fifteen plain CT characteristics, and four enhanced characteristics were used to develop two logistic regression models: model 1 (plain CT only) and model 2 (plain + contrast CT). The diagnostic performance of the two models was assessed separately in the development and external validation cohorts using the AUC. 392 patients from Center A were included in the training cohort (median size, 20.0 [IQR, 15.0-24.0] mm; mean age, 55.8 [SD, 9.9] years; male, 53.3%). 135 patients from Center B were included in the external validation cohort (median size, 20.0 [IQR, 16.0-24.0] mm; mean age, 56.4 [SD, 9.6] years; male, 51.9%). Preoperative patients with 201 malignant (adenocarcinoma, 148 [73.6%]; squamous cell carcinoma, 35 [17.4%]; large cell carcinoma,18 [9.0%]) and 326 benign (pulmonary hamartoma, 118 [36.2%]; sclerosing pneumocytoma, 35 [10.7%]; tuberculosis, 104 [31.9%]; inflammatory pseudonodule, 69 [21.2%]) solitary solid pulmonary nodules were gathered from two independent centers. The mean sensitivity, specificity, accuracy, PPV, NPV, and AUC (95%CI) of model 1 (Plain CT only) were 0.79, 0.78, 0.79, 0.67, 0.87, and 0.88 (95%CI, 0.82-0.93), the model 2 (Plain + Contrast CT) were 0.88, 0.91, 0.90, 0.84, 0.93, 0.93 (95%CI, 0.88-0.98) in external validation cohort, respectively. A logistic regression model based on plain and contrast-enhanced CT characteristics showed exceptional performance in the evaluation of malignancy for solitary solid lung nodules. Utilizing this contrast-enhanced CT model would provide recommendations concerning follow-up or surgical intervention for preoperative patients presenting with solid lung nodules.

Project description:PurposeTo evaluate the diagnostic accuracy and specific imaging characteristics of positron emission tomography/computed tomography with 18F-fluorodeoxyglucose (PET/CT), contrast enhanced CT (CE-CT), and a combined imaging approach (CE-PET/CT) in patients with infectious/mycotic (MAA), inflammatory (IAA), and non-infected, non-inflammatory abdominal aortic aneurysm (AAA).Materials and methodsIn this single-center retrospective cohort study, all imaging data sets of 29 consecutive patients with clinically suspected MAA or IAA were anonymised with different, reshuffled identification numbers and retrospectively and independently analysed by two experienced readers, blinded to all clinical patient data. Readers determined the presence or absence and MAA, IAA and AAA and of predefined imaging characteristics (e.g. fluid collection), and measured metabolic activity and wall thickness of all aneurysms. A multidisciplinary team of specialists served as standard of reference and re-evaluated every clinical case, considering all clinical, laboratory, microbiological, histopathological and imaging results, including all follow-up examinations.ResultsDiagnostic accuracy was higher in PET/CT as compared to CE-CT in differentiating AAA from MAA and IAA: area under the receiver operating characteristic curve (AUC-ROC) 0.81 (95% confidence intervals 0.69-0.92) and 0.63 (0.52-0.74) (P = 0.027). Specific imaging characteristics were significantly associated with different types of aneurysms (P<0.05), i.e. very high metabolic activity and dorsal sparing of metabolic activity in PET/CT and wall thickening in CE-CT were indicative for IAA; fat stranding and fluid collections in CE-CT were associated with MAA; while low metabolic acitivity and absence of wall thickening in PET/CT, and less fat stranding and absence of wall thickening in CE-CT were indicative for non-infected, non-inflammatory AAA.ConclusionSpecific imaging characteristics of PET/CT and CE-CT may be helpful in differentiating between MAA, IAA, and non-infected, non-inflammatory AAA.

Project description:Purpose18F-FDG PET/CT is widely used to evaluate indeterminate pulmonary nodules (IPNs). False positive results occur, especially from active granulomatous nodules. A PET-based imaging agent with superior specificity to 18F-FDG for IPNs, is badly needed, especially in areas of endemic granulomatous nodules. Somatostatin receptors (SSTR) are expressed in many malignant cells including small cell and non-small cell lung cancers (NSCLCs). 68Ga-DOTATATE, a positron emitter labeled somatostatin analog, combined with PET/CT imaging, may improve the diagnosis of IPNs over 18F-FDG by reducing false positives. Our study purpose was to test this hypothesis in our region with high endemic granulomatous IPNs.MethodsWe prospectively performed 68Ga-DOTATATE PET/CT and 18F-FDG PET/CT scans in the same 30 patients with newly diagnosed, treatment-naïve lung cancer (N = 14) or IPNs (N = 15) and one metastatic nodule. 68Ga-DOTATATE SUVmax levels at or above 1.5 were considered likely malignant. We analyzed the scan results, correlating with ultimate diagnosis via biopsy or 2-year chest CT follow-up. We also correlated 68Ga-DOTATATE uptake with immunohistochemical (IHC) staining for SSTR subtype 2A (SSTR2A) in pathological specimens.ResultsWe analyzed 31 lesions in 30 individuals, with 14 (45%) being non-neuroendocrine lung cancers and 1 (3%) being metastatic disease. McNemar's result comparing the two radiopharmaceuticals (p = 0.65) indicates that their accuracy of diagnosis in this indication are equivalent. 68Ga-DOTATATE was more specific (94% compared to 81%) and less sensitive 73% compared to 93%) than 18F-FDG. 68Ga-DOTATATE uptake correlated with SSTR2A expression in tumor stroma determined by immunohistochemical (IHC) staining in 5 of 9 (55%) NSCLCs.Conclusion68Ga-DOTATATE and 18F-FDG PET/CT had equivalent accuracy in the diagnosis of non-neuroendocrine lung cancer and 68Ga-DOTATATE was more specific than 18F-FDG for the diagnosis of IPNs. IHC staining for SSTR2A receptor expression correlated with tumor stroma but not tumor cells.

Project description:BackgroundAlterations in the vascular microenvironment of cancer have gradually gained widespread attention in oncology research, and early qualitative diagnosis using noninvasive examination techniques is crucial in clinical practice. This study aimed to explore the value of quantitative parameters of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in the assessment of vascular architecture of solid solitary pulmonary nodules (SSPNs).MethodsDCE-MRI scan images of a total of 79 patients with SSPNs from November 2021 to December 2023 were prospectively analyzed. Quantitative permeability parameters Ktrans, Kep, and Ve values of the lesions were measured, labeled immunohistochemically with CD31 microvessel density (CD31-MVD), smooth muscle actin microvessel density (SMA-MVD), and vascular endothelial growth factor (VEGF), and quantitative values of vascular architectural parameters were obtained using the image analysis software Image-J. The diagnostic efficacy of quantitative permeability parameters for the diagnosis of malignant SSPNs was analyzed by receiver operating characteristic (ROC) curves using surgical pathology findings as the gold standard for the diagnosis of malignant SSPNs. Spearman's correlation analysis was used to evaluate the relationship between quantitative permeability parameters and vascular construct parameters.ResultsThe study included 51 female patients (64.6%) with a mean age of 50.5±11.5 years. Only 21 patients (26.6%) had a smoking history, and 30 patients (38.0%) had a normal weight. Statistically significant differences were found between benign and malignant SSPNs subgroups in Ktrans and Ve values in quantitative permeability parameters, and in SMA-MVD and microvascular perfusion index (MPI) in vascular conformation parameters (P<0.05). Ktrans and Ve values showed high diagnostic value for malignant SSPNs [area under the curve (AUC), 0.813 and 0.759]. Ktrans, Kep, and Ve were significantly positively correlated with CD31-MVD, with r-values of 0.857, 0.830, and 0.798 (P<0.05); they were also significantly positively correlated with VEGF, with r-values of 0.851, 0.859, and 0.764 (P<0.05); and Ktrans and Kep were significantly negatively correlated with MPI, with r-values were -0.779 and -0.864 (P<0.05).ConclusionsThe quantitative permeability parameters of DCE-MRI correlate with some of the vascular architectural parameters, and magnetic resonance image-based quantification of pulmonary nodal vascular architecture provides a noninvasive method for the differential diagnosis of SSPNs.

Dataset Information

Diagnostic Accuracy of a Convolutional Neural Network Assessment of Solitary Pulmonary Nodules Compared With PET With CT Imaging and Dynamic Contrast-Enhanced CT Imaging Using Unenhanced and Contrast-Enhanced CT Imaging.

Background

Research question

Study design and methods

Results

Interpretation

Trial registration

Publications

Diagnostic Accuracy of a Convolutional Neural Network Assessment of Solitary Pulmonary Nodules Compared With PET With CT Imaging and Dynamic Contrast-Enhanced CT Imaging Using Unenhanced and Contrast-Enhanced CT Imaging.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets