RESUMEN
Rationale: The optimal follow-up computed tomography (CT) interval for detecting the progression of interstitial lung abnormality (ILA) is unknown. Objectives: To identify optimal follow-up strategies and extent thresholds on CT relevant to outcomes. Methods: This retrospective study included self-referred screening participants aged 50 years or older, including nonsmokers, who had imaging findings relevant to ILA on chest CT scans. Consecutive CT scans were evaluated to determine the dates of the initial CT showing ILA and the CT showing progression. Deep learning-based ILA quantification was performed. Cox regression was used to identify risk factors for the time to ILA progression and progression to usual interstitial pneumonia (UIP). Measurements and Main Results: Of the 305 participants with a median follow-up duration of 11.3 years (interquartile range, 8.4-14.3 yr), 239 (78.4%) had ILA on at least one CT scan. In participants with serial follow-up CT studies, ILA progression was observed in 80.5% (161 of 200), and progression to UIP was observed in 17.3% (31 of 179), with median times to progression of 3.2 years (95% confidence interval [CI], 3.0-3.4 yr) and 11.8 years (95% CI, 10.8-13.0 yr), respectively. The extent of fibrosis on CT was an independent risk factor for ILA progression (hazard ratio, 1.12 [95% CI, 1.02-1.23]) and progression to UIP (hazard ratio, 1.39 [95% CI, 1.07-1.80]). Risk groups based on honeycombing and extent of fibrosis (1% in the whole lung or 5% per lung zone) showed significant differences in 10-year overall survival (P = 0.02). Conclusions: For individuals with initially detected ILA, follow-up CT at 3-year intervals may be appropriate to monitor radiologic progression; however, those at high risk of adverse outcomes on the basis of the quantified extent of fibrotic ILA and the presence of honeycombing may benefit from shortening the interval for follow-up scans.
RESUMEN
Background The solid portion size of lung cancer lesions manifesting as subsolid lesions is key in their management, but the automatic measurement of such lesions by means of a deep learning (DL) algorithm needs evaluation. Purpose To evaluate the performance of a commercially available DL algorithm for automatic measurement of the solid portion of surgically proven lung adenocarcinomas manifesting as subsolid lesions. Materials and Methods Surgically proven lung adenocarcinomas manifesting as subsolid lesions on CT images between January 2018 and December 2018 were retrospectively included. Five radiologists independently measured the maximal axial diameter of the solid portion of lesions. The DL algorithm automatically segmented and measured the maximal axial diameter of the solid portion. Reader measurements, software measurements, and invasive component size at pathologic examination were compared by using intraclass correlation coefficient (ICC) and Bland-Altman plots. Results A total of 448 patients (mean age, 63 years ± 10 [standard deviation]; 264 women) with 448 lesions were evaluated (invasive component size, 3-65 mm). The measurement agreements between each radiologist and the DL algorithm were very good (ICC range, 0.82-0.89). When a radiologist was replaced with the DL algorithm, the ICCs ranged from 0.87 to 0.90, with an ICC of 0.90 among five radiologists. The mean difference between the DL algorithm and each radiologist ranged from -3.7 to 1.5 mm. The widest 95% limit of agreement between the DL algorithm and each radiologist (-15.7 to 8.3 mm) was wider than pairwise comparisons of radiologists (-7.7 to 13.0 mm). The agreement between the DL algorithm and invasive component size at pathologic evaluation was good, with an ICC of 0.67. Measurements by the DL algorithm (mean difference, -6.0 mm) and radiologists (mean difference, -7.5 to -2.3 mm) both underestimated invasive component size. Conclusion Automatic measurements of solid portions of lung cancer manifesting as subsolid lesions by the deep learning algorithm were comparable with manual measurements and showed good agreement with invasive component size at pathologic evaluation. © RSNA, 2021 Online supplemental material is available for this article.
Asunto(s)
Adenocarcinoma/diagnóstico por imagen , Adenocarcinoma/patología , Aprendizaje Profundo , Interpretación de Imagen Asistida por Computador/métodos , Neoplasias Pulmonares/diagnóstico por imagen , Neoplasias Pulmonares/patología , Tomografía Computarizada por Rayos X , Femenino , Humanos , Masculino , Persona de Mediana Edad , Radiólogos , Estudios Retrospectivos , Programas InformáticosRESUMEN
OBJECTIVES: To develop a model for differentiating the predominant subtype-based prognostic groups of lung adenocarcinoma using CT radiomic features, and to validate its performance in comparison with radiologists' assessments. METHODS: A total of 993 patients presenting with invasive lung adenocarcinoma between March 2010 and June 2016 were identified. Predominant histologic subtypes were categorized into three groups according to their prognosis (group 0: lepidic; group 1: acinar/papillary; group 2: solid/micropapillary). Seven hundred eighteen radiomic features were extracted from segmented lung cancers on contrast-enhanced CT. A model-development set was formed from the images of 893 patients, while 100 image sets were reserved for testing. A least absolute shrinkage and selection operator method was used for feature selection. Performance of the radiomic model was evaluated using receiver operating characteristic curve analysis, and accuracy on the test set was compared with that of three radiologists with varying experiences (6, 7, and 19 years in chest CT). RESULTS: Our model differentiated the three groups with areas under the curve (AUCs) of 0.892 and 0.895 on the development and test sets, respectively. In pairwise discrimination, the AUC was highest for group 0 vs. 2 (0.984). The accuracy of the model on the test set was higher than the averaged accuracy of the three radiologists without statistical significance (73.0% vs. 61.7%, p = 0.059). For group 2, the model achieved higher PPV than the observers (85.7% vs. 35.0-48.4%). CONCLUSIONS: Predominant subtype-based prognostic groups of lung adenocarcinoma were classified by a CT-based radiomic model with comparable performance to radiologists. KEY POINTS: ⢠A CT-based radiomic model differentiated three prognosis-based subtype groups of lung adenocarcinoma with areas under the curve (AUCs) of 0.892 and 0.895 on development and test sets, respectively. ⢠The CT-based radiomic model showed near perfect discrimination between group 0 and group 2 (AUCs, 0.984-1.000). ⢠The accuracy of the CT-based radiomic model was comparable to the averaged accuracy of the three radiologists with 6, 7, and 19 years of clinical experience in chest CT (73.0% vs. 61.7%, p = 0.059), achieving a higher positive predictive value for group 2 than the observers (85.7% vs. 35.0-48.4%).
Asunto(s)
Adenocarcinoma del Pulmón/diagnóstico , Neoplasias Pulmonares/diagnóstico , Estadificación de Neoplasias/métodos , Tomografía Computarizada por Rayos X/métodos , Adulto , Anciano , Anciano de 80 o más Años , Femenino , Humanos , Masculino , Persona de Mediana Edad , Pronóstico , Curva ROCRESUMEN
RATIONALE AND OBJECTIVES: To establish a quantitative CT threshold for radiological disease progression of progressive pulmonary fibrosis (PPF) and evaluate its feasibility in patients with connective tissue disease-related interstitial lung disease (CTD-ILD). MATERIALS AND METHODS: Between April 2007 and October 2022, patients diagnosed with CTD-ILD retrospectively evaluated. CT quantification was conducted using a commercial software by summing the percentages of ground-glass opacity, consolidation, reticular opacity, and honeycombing. The quantitative threshold for radiological progression was determined based on the highest discrimination on overall survival (OS). Two thoracic radiologists independently evaluated visual radiological progression, and the senior radiologist's assessment was used as the final result. Cox regression was used to assess prognosis of PPF based on the visual assessment and quantitative threshold. RESULTS: 97 patients were included and followed up for a median of 30.3 months (range, 4.7-198.1 months). For defining radiological disease progression, the optimal quantitative CT threshold was 4%. Using this threshold, 12 patients were diagnosed with PPF, while 14 patients were diagnosed with PPF based on the visual assessment, with an agreement rate of 97.9% (95/97). Worsening respiratory symptoms (hazard ratio [HR], 12.73; P < .001), PPF based on the visual assessment (HR, 8.86; P = .002) and based on the quantitative threshold (HR, 6.72; P = .009) were independent risk factors for poor OS. CONCLUSION: The quantitative CT threshold for radiological disease progression (4%) was feasible in defining PPF in terms of its agreement with PPF grouping and prognostic performance when compared to visual assessment.
RESUMEN
In clinical diagnostics, single-stranded DNAs (ssDNA) have been prepared from the human genomic DNA for the detection of a specific gene. In this study, the human genomic DNA was degraded via ultrasonication in solution, and K-ras oncogene was detected from the DNA fragments via the fluorescence quenching of quantum dots (QDs) by intercalating dyes after hybridization of the target, to probe DNAs in a microfluidic chip. K-ras is one of the most activated common oncogenes, and many human tumors are known to be due to the mutation of this gene. QDs are nano-sized semiconductors with a wide selection of emission wavelengths and exceptional stability against photo bleaching. In this study, probe DNA-conjugated QDs were immobilized to polystyrene microbeads, and the DNA-microbead-QDs complexes were packed through a microchannel by pillars that trap the beads in the microfluidic chip. The fluorescence of the QDs could be quenched by intercalating dye (TOTO-3) after hybridization of K-ras oncogene to the probe DNA in the channel. The fluorescence intensity decrease of the QDs can be used as an indication of the K-ras oncogene. By introducing an alkaline buffer solution, the DNAs were denatured, and the fluorescence intensity of the QDs again increased, which shows the possibility of reuse of the microfluidic chip for the detection of the K-ras gene.
Asunto(s)
ADN/genética , Genes ras , Genoma Humano , Microfluídica/instrumentación , Puntos Cuánticos , Ultrasonido , Humanos , Microscopía Fluorescente , MutaciónRESUMEN
Recently quantum dots (QDs) have been extensively used in the field of biotechnology. QDs have merits of wide selection of emission wavelength and exceptional stability against photo bleaching over conventional organic fluorophores and are used in cell imaging, biomarker, and fluorescence resonance energy transfer (FRET) sensor. Magnetic beads have been used as solid support in microfluidic devices to trace bio-molecules. In this study, Polydimethylsiloxane (PDMS) based microfluidic chips were prepared for the detection of K-Ras oncogene by using QDs-DNA conjugate. K-Ras oncogene can be detected by fluorescence quenching in microfluidic chip. Carboxylated CdSe/ZnS QDs (emission wavelength: 605 nm) could bind to magnetic beads of polystyrene/divinyl benzene via EDC/NHS crosslinking reaction. The fluorescence from QDs could be quenched by intercalating dye (thiazol orange dimers: TOTO-3) after hybridization with target DNA and probe DNA in the channel of microfluidic chip. The fluorescence intensity change of QDs after hybridization in microfluidic chip has been studied.
Asunto(s)
Genes ras , Técnicas Analíticas Microfluídicas/instrumentación , Microfluídica/métodos , Neoplasias/genética , Puntos Cuánticos , ADN/química , Diseño de Equipo , Transferencia Resonante de Energía de Fluorescencia , Colorantes Fluorescentes/farmacología , Humanos , Procesamiento de Imagen Asistido por Computador , Magnetismo , Ensayo de Materiales , Neoplasias/patología , Sondas de Oligonucleótidos/química , Espectrofotometría , Factores de TiempoRESUMEN
The generative adversarial network (GAN) is a promising deep learning method for generating images. We evaluated the generation of highly realistic and high-resolution chest radiographs (CXRs) using progressive growing GAN (PGGAN). We trained two PGGAN models using normal and abnormal CXRs, solely relying on normal CXRs to demonstrate the quality of synthetic CXRs that were 1000 × 1000 pixels in size. Image Turing tests were evaluated by six radiologists in a binary fashion using two independent validation sets to judge the authenticity of each CXR, with a mean accuracy of 67.42% and 69.92% for the first and second trials, respectively. Inter-reader agreements were poor for the first (κ = 0.10) and second (κ = 0.14) Turing tests. Additionally, a convolutional neural network (CNN) was used to classify normal or abnormal CXR using only real images and/or synthetic images mixed datasets. The accuracy of the CNN model trained using a mixed dataset of synthetic and real data was 93.3%, compared to 91.0% for the model built using only the real data. PGGAN was able to generate CXRs that were identical to real CXRs, and this showed promise to overcome imbalances between classes in CNN training.
Asunto(s)
Redes Neurales de la Computación , Radiólogos , Humanos , RadiografíaRESUMEN
The concept of progressive pulmonary fibrosis (PPF) has been introduced to predict the diverse prognosis of interstitial lung disease (ILD). However, the incidence and effect of PPF on outcomes in patients with connective tissue disease-associated interstitial lung disease (CTD-ILD) need to be elucidated. This study reviewed 197 patients with CTD-ILD. Symptomatic worsening, pulmonary function decline, and radiological deterioration were investigated to assess the fulfillment of PPF diagnostic criteria. Clinical outcomes, including mortality, were compared based on the presence or absence of PPF. The median follow-up duration was 17.4 months. The mean age of the patients was 64.0 years, and 60.9% were female. Among the underlying CTDs, rheumatoid arthritis (42.1%), inflammatory myositis (19.8%), and systemic sclerosis (13.2%) were the most common. Of the 197 patients, 37 (18.8%) met the diagnostic criteria for PPF during the follow-up period. Even after adjusting for other significant risk factors, PPF was independently associated with mortality [hazard ratio (HR) 3.856; 95% confidence interval (CI) 1.387-10.715; P = 0.010] and baseline albumin was marginally significantly associated with mortality (HR 0.549; CI 0.298-1.010; P = 0.054). The median survival was also significantly shorter in the PPF group than in the non-PPF group (72.3 ± 12.9 vs. 126.8 ± 15.5 months, P < 0.001). Baseline KL-6 ≥ 1000 (U/mL) was a significant risk factor for PPF (HR 2.885; CI 1.165-7.144; P = 0.022). In addition to increased mortality, the PPF group had significantly higher rates of respiratory-related hospitalizations, pneumonia, acute exacerbations, and weight loss than the non-PPF group. PPF is a significant prognostic indicator in patients with CTD-ILD. Thus, healthcare professionals should know that patients with CTD-ILD are at risk of PPF.
Asunto(s)
Enfermedades del Tejido Conjuntivo , Enfermedades Pulmonares Intersticiales , Fibrosis Pulmonar , Humanos , Femenino , Persona de Mediana Edad , Masculino , Fibrosis Pulmonar/complicaciones , Estudios Retrospectivos , Enfermedades Pulmonares Intersticiales/complicaciones , Enfermedades Pulmonares Intersticiales/diagnóstico , Enfermedades del Tejido Conjuntivo/complicaciones , Enfermedades del Tejido Conjuntivo/diagnóstico , Enfermedades del Tejido Conjuntivo/epidemiología , PulmónRESUMEN
BACKGROUND: Computer-aided diagnosis on chest x-ray images using deep learning is a widely studied modality in medicine. Many studies are based on public datasets, such as the National Institutes of Health (NIH) dataset and the Stanford CheXpert dataset. However, these datasets are preprocessed by classical natural language processing, which may cause a certain extent of label errors. OBJECTIVE: This study aimed to investigate the robustness of deep convolutional neural networks (CNNs) for binary classification of posteroanterior chest x-ray through random incorrect labeling. METHODS: We trained and validated the CNN architecture with different noise levels of labels in 3 datasets, namely, Asan Medical Center-Seoul National University Bundang Hospital (AMC-SNUBH), NIH, and CheXpert, and tested the models with each test set. Diseases of each chest x-ray in our dataset were confirmed by a thoracic radiologist using computed tomography (CT). Receiver operating characteristic (ROC) and area under the curve (AUC) were evaluated in each test. Randomly chosen chest x-rays of public datasets were evaluated by 3 physicians and 1 thoracic radiologist. RESULTS: In comparison with the public datasets of NIH and CheXpert, where AUCs did not significantly drop to 16%, the AUC of the AMC-SNUBH dataset significantly decreased from 2% label noise. Evaluation of the public datasets by 3 physicians and 1 thoracic radiologist showed an accuracy of 65%-80%. CONCLUSIONS: The deep learning-based computer-aided diagnosis model is sensitive to label noise, and computer-aided diagnosis with inaccurate labels is not credible. Furthermore, open datasets such as NIH and CheXpert need to be distilled before being used for deep learning-based computer-aided diagnosis.