Assessing the Accuracy of a Deep Learning Method to Risk Stratify Indeterminate Pulmonary Nodules.
Ontology highlight
ABSTRACT:
SUBMITTER: Massion PP
PROVIDER: S-EPMC7365375 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundNon-invasive biomarkers are needed to improve management of indeterminate pulmonary nodules (IPNs) suspicious for lung cancer.MethodsProtein biomarkers were quantified in serum samples from patients with 6-30 mm IPNs (n = 338). A previously derived and validated radiomic score based upon nodule shape, size, and texture was calculated from features derived from CT scans. Lung cancer prediction models incorporating biomarkers, radiomics, and clinical factors were developed. Diagnostic performance was compared to the current standard of risk estimation (Mayo). IPN risk reclassification was determined using bias-corrected clinical net reclassification index.ResultsAge, radiomic score, CYFRA 21-1, and CEA were identified as the strongest predictors of cancer. These models provided greater diagnostic accuracy compared to Mayo with AUCs of 0.76 (95 % CI 0.70-0.81) using logistic regression and 0.73 (0.67-0.79) using random forest methods. Random forest and logistic regression models demonstrated improved risk reclassification with median cNRI of 0.21 (Q1 0.20, Q3 0.23) and 0.21 (0.19, 0.23) compared to Mayo for malignancy.ConclusionsA combined biomarker, radiomic, and clinical risk factor model provided greater diagnostic accuracy of IPNs than Mayo. This model demonstrated a strong ability to reclassify malignant IPNs. Integrating a combined approach into the current diagnostic algorithm for IPNs could improve nodule management.
| S-EPMC10057862 | biostudies-literature
Project description:Objective:To demonstrate and test the validity of a novel deep-learning-based system for the automated detection of pulmonary nodules. Materials and Methods:The proposed system uses 2 3D deep learning models, 1 for each of the essential tasks of computer-aided nodule detection: candidate generation and false positive reduction. A total of 888 scans from the LIDC-IDRI dataset were used for training and evaluation. Results:Results for candidate generation on the test data indicated a detection rate of 94.77% with 30.39 false positives per scan, while the test results for false positive reduction exhibited a sensitivity of 94.21% with 1.789 false positives per scan. The overall system detection rate on the test data was 89.29% with 1.789 false positives per scan. Discussion:An extensive and rigorous validation was conducted to assess the performance of the proposed system. The system demonstrated a novel combination of 3D deep neural network architectures and demonstrates the use of deep learning for both candidate generation and false positive reduction to be evaluated with a substantial test dataset. The results strongly support the ability of deep learning pulmonary nodule detection systems to generalize to unseen data. The source code and trained model weights have been made available. Conclusion:A novel deep-neural-network-based pulmonary nodule detection system is demonstrated and validated. The results provide comparison of the proposed deep-learning-based system over other similar systems based on performance.
| S-EPMC6188511 | biostudies-literature
Project description:PURPOSE:Diagnosis and resection of indeterminate pulmonary nodules (IPNs) is a growing challenge with increased utilization of chest computed tomography. Photoacoustic (PA) -guided surgical resection with local injection of indocyanine green (ICG) may have utility for IPNs that are suspicious for lung cancer. This preclinical study explores the potential of PA imaging (PAI) to detect ICG-labeled tumors. MATERIALS AND METHODS:ICG uptake by H460 lung cancer cells was evaluated in vitro. A phantom study was performed to analyze PA signal intensity according to ICG concentration and tissue thickness/depth using chicken breast. PA signals were measured up to 48 hours after injection of ICG (mixed with 5% agar) into healthy subcutaneous tissue, subcutaneous H460 tumors and right healthy lung in nude mice. RESULTS:Intracellular ICG fluorescence was detected in H460 cells co-incubated with ICG in vitro. The concentration dependence of the PA signal was logarithmic, and PA signal decline was exponential with increasing tissue depth. The PA signal of 2 mg/mL ICG was still detectable at a depth of 22 mm in chicken breast. The PA signal from ICG mixed with agar was detectable 48 hours post injection into subcutaneous tissue and subcutaneous H460 tumors in nude mice. Similar features of PA signals from ICG-agar in mice lung were obtained. CONCLUSION:The results from this preclinical study suggests that PAI of injected ICG-agar may be beneficial for identifying deeply located tumors. These features may be valuable for IPNs.
| S-EPMC7173852 | biostudies-literature
Project description:RationalePulmonologists frequently encounter indeterminate pulmonary nodules in practice, but it is unclear what clinical factors they rely on to guide the diagnostic evaluation.ObjectivesTo assess the current approach to the management of indeterminate pulmonary nodules and to determine the extent to which the addition of a hypothetical diagnostic blood test will influence clinical decision making.MethodsSelected pulmonologists practicing in the United States were invited to participate in a conjoint exercise based on 20 randomly generated cases of varying age, smoking history, and nodule size. Some cases included the result of a hypothetical blood test. Each respondent chose from among three diagnostic options for a patient: noninvasive monitoring (i.e., serial CT or positron emission tomography scan), a minor procedure (i.e., biopsy or bronchoscopy), or a major procedure (i.e., video-assisted thorascopic surgery or thoracotomy). Multivariate logistic regression was used to assess the impact of the three risk factors and the diagnostic blood test on decision making.Measurements and main resultsFour hundred nineteen physicians participated (response rate, 10%). One hundred fifty-three physician surveys met predetermined criteria and were analyzed (4% of all invitees). A diagnostic procedure was recommended for 23% of 6-mm nodules, versus 54, 66, 77, and 84% of nodules 10, 14, 18, and 22 mm, respectively (P?<?0.001). Older age limited recommendations for invasive testing: 54% of 80-year-olds versus 61, 64, 63, and 61% of patients 71, 62, 53, and 44 years of age, respectively (P?<?0.001). In multivariate analyses, nodule size, smoking history, age, and the blood test each influenced decision making (P?<?0.001).ConclusionsThe pulmonologists who participated in this survey were more likely to proceed with invasive testing, instead of observation or additional imaging, as the size of the nodule increased. The use of a hypothetical blood test resulted in significant alterations in the decision to pursue invasive testing.
| S-EPMC5475427 | biostudies-literature
Project description:This study assessed the diagnostic performance of a deep learning (DL)-based model for differentiating malignant subcentimeter (≤10 mm) solid pulmonary nodules (SSPNs) from benign ones in computed tomography (CT) images compared against radiologists with 10 and 15 years of experience in thoracic imaging (medium-senior seniority). Overall, 200 SSPNs (100 benign and 100 malignant) were retrospectively collected. Malignancy was confirmed by pathology, and benignity was confirmed by follow-up or pathology. CT images were fed into the DL model to obtain the probability of malignancy (range, 0-100%) for each nodule. According to the diagnostic results, enrolled nodules were classified into benign, malignant, or indeterminate. The accuracy and diagnostic composition of the model were compared with those of the radiologists using the McNemar-Bowker test. Enrolled nodules were divided into 3-6-, 6-8-, and 8-10-mm subgroups. For each subgroup, the diagnostic results of the model were compared with those of the radiologists. The accuracy of the DL model, in differentiating malignant and benign SSPNs, was significantly higher than that of the radiologists (71.5% vs. 38.5%, P<0.001). The DL model reported more benign or malignant deterministic results and fewer indeterminate results. In subgroup analysis of nodule size, the DL model also yielded higher performance in comparison with that of the radiologists, providing fewer indeterminate results. The accuracy of the two methods in the 3-6-, 6-8-, and 8-10-mm subgroups was 75.5% vs. 28.3% (P<0.001), 62.0% vs. 28.2% (P<0.001), and 77.6% vs. 55.3% (P=0.001), respectively, and the indeterminate results were 3.8% vs. 66.0%, 8.5% vs. 66.2%, and 2.6% vs. 35.5% (all P<0.001), respectively. The DL-based method yielded higher performance in comparison with that of the radiologists in differentiating malignant and benign SSPNs. This DL model may reduce uncertainty in diagnosis and improve diagnostic accuracy, especially for SSPNs smaller than 8 mm.
| S-EPMC10636433 | biostudies-literature
Project description:BackgroundThe clinical significance of indeterminate pulmonary nodules (IPN) in patients with resectable pancreatic adenocarcinoma (PDAC) is unknown. The rate of detection on IPN has risen due to enhanced staging investigations to determine resectability. IPNs detected on preoperative imaging represent a clinical dilemma and complicate decision-making. Currently, there are no recommendations on the management of IPN. This review provides a comprehensive overview of the current knowledge on the natural history of IPN detected among patients with resectable PDAC.MethodsA systematic review based on a search in Medline and Embase databases was performed. All clinical studies evaluating the significance of IPN in patients with resectable PDAC were included. PRISMA guidelines were followed.ResultsFive studies met the inclusion criteria. The total patient population was 761. The prevalence of IPN reported ranged from 18 to 71%. The median follow-up duration was 17 months. The median overall survival was 19 months. Patients with pre-operative IPN which subsequently progressed to clinically recognizable pulmonary metastases, ranged from 1.5 to 16%. Four studies found that there was no significant difference in median overall survival in patients with or without IPNs.ConclusionThis is a first review on the significance of IPN in patients with resectable PDAC. The preoperative presence of IPN does not demonstrate an association with overall survival after surgery. The identification of IPN is a significant finding however it should not preclude patients with resectable PDAC from undergoing curative resection.
| S-EPMC8106596 | biostudies-literature
Project description:Background18F-fluorodeoxyglucose (FDG) PET/CT is recommended for evaluation of intermediate-risk indeterminate pulmonary nodules (IPNs). While highly sensitive, the specificity of FDG remains suboptimal for differentiating malignant from benign nodules, particularly in areas where fungal lung diseases are prevalent. Thus, a cancer-specific imaging probe is greatly needed. In this study, we tested the hypothesis that a PET radiotracer (S)-4-(3-[18F]-fluoropropyl)-L-glutamic acid (FSPG) improves the diagnostic accuracy of IPNs compared to 18F-FDG PET/CT.MethodsThis study was conducted at a major academic medical center and an affiliated VA medical center. Twenty-six patients with newly discovered IPNs 7-30mm diameter or newly diagnosed lung cancer completed serial PET/CT scans utilizing 18F-FDG and 18F-FSPG, without intervening treatment of the lesion. The scans were independently reviewed by two dual-trained diagnostic radiology and nuclear medicine physicians. Characteristics evaluated included quantitative SUVmax values of the pulmonary nodules and metastases.ResultsA total of 17 out of 26 patients had cancer and 9 had benign lesions. 18F-FSPG was negative in 6 of 9 benign lesions compared to 7 of 9 with 18F-FDG. 18F-FSPG and 18F-FDG were positive in 14 of 17 and 12 of 17 malignant lesions, respectively. 18F-FSPG detected brain and intracardiac metastases missed by 18F-FDG PET in one case, while 18F-FDG detected a metastasis to the kidney missed by 18F-FSPG.ConclusionIn this pilot study, there was no significant difference in overall diagnostic accuracy between 18F-FSPG and 18F-FDG for the evaluation of IPNs and staging of lung cancer. Additional studies will be needed to determine the clinical utility of this tracer in the management of IPNs and lung cancer.
| S-EPMC8926263 | biostudies-literature
Project description:In most of the world, diagnostic surgery remains the most frequent approach for indeterminate thyroid cytology. Although several molecular tests are available for testing in centralized commercial laboratories in the United States, there are no available kits for local laboratory testing. The aim of this study was to develop a prototype in vitro diagnostic (IVD) gene classifier for the further characterization of nodules with an indeterminate thyroid cytology.In a first stage, the expression of 18 genes was determined by quantitative polymerase chain reaction (qPCR) in a broad histopathological spectrum of 114 fresh-tissue biopsies. Expression data were used to train several classifiers by supervised machine learning approaches. Classifiers were tested in an independent set of 139 samples. In a second stage, the best classifier was chosen as a model to develop a multiplexed-qPCR IVD prototype assay, which was tested in a prospective multicenter cohort of fine-needle aspiration biopsies.In tissue biopsies, the best classifier, using only 10 genes, reached an optimal and consistent performance in the ninefold cross-validated testing set (sensitivity 93% and specificity 81%). In the multicenter cohort of fine-needle aspiration biopsy samples, the 10-gene signature, built into a multiplexed-qPCR IVD prototype, showed an area under the curve of 0.97, a positive predictive value of 78%, and a negative predictive value of 98%. By Bayes' theorem, the IVD prototype is expected to achieve a positive predictive value of 64-82% and a negative predictive value of 97-99% in patients with a cancer prevalence range of 20-40%.A new multiplexed-qPCR IVD prototype is reported that accurately classifies thyroid nodules and may provide a future solution suitable for local reference laboratory testing.
| S-EPMC5564024 | biostudies-literature
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.
| S-EPMC5300187 | biostudies-literature
Project description:BackgroundManagement recommendations for thyroid nodules rely primarily on the cytological diagnosis. However, 25% of biopsies render an indeterminate cytology for which management decision is more challenging due to heterogeneity of the specimens. This study aimed to stratify the cancer risk through subcategorization of indeterminate cytology.MethodsThe indeterminate cytological specimens (Bethesda-III or IV) of 518 thyroid nodules consecutively evaluated at our academic cancer center between October 2008 and September 2015, blinded to the histological outcome, were retrospectively reviewed. Cytological specimens were subclassified into four groups: aspirates exhibiting nuclear atypia (n?=?158; 31%); architectural atypia (n?=?222; 43%); oncocytic features (n?=?120; 23%); or other types of atypia (n?=?18; 3%). The prevalence of malignancy and odds ratio for malignancy were calculated in 323 nodules with histological confirmation.ResultsThe prevalence of malignancy was 26% overall (20% in Bethesda-III and 29% in Bethesda-IV; p?=?0.07), and 47%, 12%, 24%, and 25% for aspirates with nuclear atypia, architectural atypia, oncocytic features, or other types of atypia, respectively. The OR of nuclear atypia over architectural atypia was 6.4 (3.4-12.2; p?<?0.001), and 2.7 over oncocytic features (1.4-5.1; p?=?0.01), whereas the OR of architectural atypia over oncocytic features was 0.4 (0.2-0.9; p?=?0.03). Results were similar for Bethesda-III and IV aspirates when analyzed independently. Furthermore, cytological subcategories improved cytology-histology correlation, as they were associated with distinct profiles of histological diagnoses (p?<?0.001).ConclusionsCytological subcategories can effectively stratify the risk of malignancy of thyroid nodules with indeterminate cytology and improve cytology-histology correlation.
| S-EPMC6112164 | biostudies-literature