Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 20 de 50
Filtrar
1.
Med Sci Monit ; 26: e920950, 2020 Jun 04.
Artigo em Inglês | MEDLINE | ID: mdl-32495750

RESUMO

BACKGROUND The aim of this study was to investigate the ability of coronary computed tomographic angiography (CCTA) characteristics of high-risk plaque (HRP) in moderate stenosis to improve differentiation of myocardial ischemia detected by stress CT perfusion (CTP) imaging. MATERIAL AND METHODS Sixty-two patients with coronary plaques and moderate stenosis confirmed by invasive coronary angiography (ICA) had stress CTP and 26 of these patients were found to have myocardial ischemia. The other 36 patients without myocardial ischemia were defined as controls. Characteristics of major plaques on CCTA images of the ischemia and non-ischemia groups were analyzed and compared. RESULTS Differences between the 2 groups were observed in plaque volume, burden and rough inner surface necrotic core volume, plaque-lipid interface and plaque length. In a multivariable analysis, plaque burden and necrotic core volume were significantly associated with myocardial ischemia: plaque burden odds ratio (OR) was 1.28 (95% confidence interval [CI], 1.12-1.48); necrotic core volume OR was 1.78 (95% CI, 1.03-1.34). Compared with other quantitative measurements, optimized thresholds for plaque burden (area under the curve was 0.852) and necrotic core volume (area under the curve was 0.730) showed significantly higher diagnostic performance for ischemia with threshold values of 60.8% and 11.25 mm³, respectively. CONCLUSIONS CCTA characteristics of major plaques may improve the discrimination of ACS patients with myocardial ischemia on stress CTP.

2.
Neuroradiology ; 62(6): 661-667, 2020 Jun.
Artigo em Inglês | MEDLINE | ID: mdl-32008047

RESUMO

PURPOSE: To recommend a new simple and explicit index termed the anteroposterior diameter of the lateral ventricle index (ALVI) for assessing brain ventricular size in neuroimaging and to compare Evans index (EI) between idiopathic normal pressure hydrocephalus (iNPH) patients and age-matched healthy elderly subjects. METHODS: Retrospective measurements of ventricular volume (VV), relative VV (RVV), the EI, and the ALVI were taken from thin-section CT scans for 23 pre-shunt-insertion iNPH patients and 62 age-matched healthy elderly volunteers. The area under the receiver operating characteristic (ROC) curve (AUC), net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were calculated to assess the effectiveness of ALVI scores for predicting VV. RESULTS: The correlations between VV or RVV and ALVI scores (VV, r = 0.957; RVV, r = 0.983) were significantly stronger than the corresponding correlations with EI scores (VV, r = 0.843; RVV, r = 0.840). The AUC for ALVI scores was significantly greater than the AUC for EI scores. Furthermore, with the inclusion of the ALVI, the NRI value was 0.14 and the IDI value was 0.14; these improvements were also statistically significant. CONCLUSION: The ALVI is a more accurate and more explicitly defined marker of VV than the EI and assesses ventricular enlargement effectively. We suggest that ventricular enlargement of the healthy elderly be defined by ALVI > 0.50.

3.
Exp Ther Med ; 18(6): 4265-4272, 2019 Dec.
Artigo em Inglês | MEDLINE | ID: mdl-31777534

RESUMO

In the present study, multi-slice CT results of patients with Behçet's disease (BD) and vascular complications were retrospectively evaluated. From January 2016 to May 2018, 45 of 361 patients with BD were diagnosed with vascular involvement. The clinical background, laboratory parameters and response to therapy of those patients were assessed. The following characteristics of vascular aneurysms were analyzed: Maximum diameter, length, wall thickness, borders, luminal changes, mural thrombus, cystic change of the vessel walls, asymmetric bulging of the right part of the aortic wall (RP type) and calcific plaques. The 45 BD patients analyzed included 37 males and 8 females with a median age of 40 years (30-49 years). Significant differences were observed among genders regarding age, ocular disorders and digestive-tract ulceration. A total of 42 aneurysms were identified with a mean diameter of 43 mm. Most aneurysmal walls (88%) were homogeneously enhanced on contrast-enhanced CT. Comparison of groups classified by aortic and larger arterial aneurysms indicated that aneurysms occurring in the aorta were more likely to form a mural thrombus, have a thicker wall (P<0.001) and unclear borders (P=0.036), to be of the RP type (P=0.003) and have a longer extension (P=0.001) compared with those in larger arteries. Unclear border of the aneurysmal wall was the only radiologic predictor correlated with an elevated erythrocyte sedimentation rate (P<0.001). In conclusion, characteristic CT imaging features of aneurysms may help to diagnose vascular involvement of BD and assess its severity, particularly in the absence of the classical clinical manifestations.

4.
J Comput Assist Tomogr ; 43(6): 937-942, 2019.
Artigo em Inglês | MEDLINE | ID: mdl-31738203

RESUMO

OBJECTIVE: The objective of this study was to evaluate the diagnostic performance of magnetic resonance imaging (MRI) in detecting the secondary breast cancer among women with a personal history of the lesion. MATERIALS AND METHODS: We retrospectively reviewed breast MRI examinations performed between 2007 and 2011. A total of 798 women with a history of breast cancer were included in the study. Cancer detection rate, positive predictive value (PPV), recall rate, sensitivity, and specificity were assed. Cancer detection rate was stratified by interval after surgery of the primary breast cancer. Also, we derived 1 comparison group from the women for comparing the performance of x-ray mammography, ultrasound, and MRI in detecting the second breast cancer. RESULTS: Of the 798 patients, 47 of the 49 secondary breast carcinomas were detected by MRI. The sensitivity and specificity of MRI in detecting the secondary lesions were 95.9% and 96.3%, respectively. The recall rate was 9.5%, and the PPV was 61.8%. Cancer detection rate of MRI examinations performed at more than 36 months after initial surgery was significantly higher than that at 36 months or less after initial surgery (13.7% vs 3.6, P < 0.001). In comparison group, the sensitivity and specificity of MRI, mammography, and ultrasound were 96.7% and 96.1%, 48.4% and 93.9%, and 77.4% and 96.1%, respectively. CONCLUSIONS: Surveillance MRI for women with a personal history of breast cancer has high sensitivity in finding the secondary malignancies with a reasonable recall rate and PPV.


Assuntos
Neoplasias da Mama/diagnóstico por imagem , Imagem por Ressonância Magnética/métodos , Mamografia/métodos , Segunda Neoplasia Primária/diagnóstico por imagem , Ultrassonografia Mamária/métodos , Adulto , Idoso , Idoso de 80 Anos ou mais , Feminino , Humanos , Pessoa de Meia-Idade , Vigilância da População , Valor Preditivo dos Testes , Estudos Retrospectivos , Sensibilidade e Especificidade
5.
Sci Rep ; 9(1): 14586, 2019 10 10.
Artigo em Inglês | MEDLINE | ID: mdl-31601919

RESUMO

Thin-section computed tomography (TSCT) imaging biomarkers are uncertain to distinguish progressive adenocarcinoma from benign lesions in pGGNs. The purpose of this study was to evaluate the usefulness of TSCT characteristics for differentiating among transient (TRA) lesions, atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IAC) presenting as pure ground-glass nodules (pGGNs). Between January 2016 and January 2018, 255 pGGNs, including 64 TRA, 22 AAH, 37 AIS, 108 MIA and 24 IAC cases, were reviewed on TSCT images. Differences in TSCT characteristics were compared among these five subtypes of pGGNs. Logistic analysis was performed to identify significant factors for predicting MIA and IAC. Progressive pGGNs were more likely to be round or oval in shape, with clear margins, air bronchograms, vascular and pleural changes, creep growth, and bubble-like lucency than were non-progressive pGGNs. The optimal cut-off values of the maximum diameter for differentiating non-progressive from progressive pGGNs and IAC from non-IAC were 6.5 mm and 11.5 mm, respectively. For the prediction of IAC vs. non-IAC and non-progressive vs. progressive adenocarcinoma, the areas under the receiver operating characteristics curves were 0.865 and 0.783 for maximum diameter and 0.784 and 0.722 for maximum CT attenuation, respectively. The optimal cut-off values of maximum CT attenuation were -532 HU and -574 HU for differentiating non-progressive from progressive pGGNs and IAC from non-IAC, respectively. Maximum diameter, maximum attenuation and morphological characteristics could help distinguish TRA lesions from MIA and IAC but not from AAH. So, CT morphologic characteristics, diameter and attenuation parameters are useful for differentiating among pGGNs of different subtypes.

6.
Thorac Cancer ; 10(10): 1893-1903, 2019 10.
Artigo em Inglês | MEDLINE | ID: mdl-31426132

RESUMO

BACKGROUND: The aim of this study was to investigate the influence of convolution kernel and iterative reconstruction on the diagnostic performance of radiomics and deep learning (DL) in lung adenocarcinomas. METHODS: A total of 183 patients with 215 lung adenocarcinomas were included in this study. All CT imaging data was reconstructed with three reconstruction algorithms (ASiR at 0%, 30%, 60% strength), each with two convolution kernels (bone and standard). A total of 171 nodules were selected as the training-validation set, whereas 44 nodules were selected as the testing set. Logistic regression and a DL framework-DenseNets were selected to tackle the task. Three logical experiments were implemented to fully explore the influence of the studied parameters on the diagnostic performance. The receiver operating characteristic curve (ROC) was used to evaluate the performance of constructed models. RESULTS: In Experiments A and B, no statistically significant results were found in the radiomic method, whereas two and six pairs were statistically significant (P < 0.05) in the DL method. In Experiment_C, significant differences in one and four models were found in the radiomics and DL methods, respectively. Moreover, models constructed with standard convolution kernel data outperformed that constructed with bone convolution kernel data in all studied ASiR levels in the DL method. In the DL method, B0 and S60 performed best in bone and standard convolution kernel, respectively. CONCLUSION: The results demonstrated that DL was more susceptible to CT parameter variability than radiomics. Standard convolution kernel images seem to be more appropriate for imaging analysis. Further investigation with a larger sample size is needed.

7.
World Neurosurg ; 129: e48-e55, 2019 Sep.
Artigo em Inglês | MEDLINE | ID: mdl-31051310

RESUMO

OBJECTIVE: To investigate potential risk factors for cerebral hyperperfusion syndrome (CHS) after surgery in patients with moyamoya disease (MMD) using phase-contrast magnetic resonance imaging (MRI). METHODS: The study included 84 adult patients (89 brain hemispheres) with MMD who underwent surgery. Preoperative phase-contrast MRI scans were performed for all patients. Peak velocity, average velocity, average flow, forward volume, and region of interest area of the bilateral internal and external carotid arteries, superficial temporal artery, and vertebral artery were calculated and analyzed. Patients were divided into CHS and non-CHS groups. Patients' age, sex, clinical symptoms, Suzuki stage, and MRI flow examination results were compared between the 2 groups. RESULTS: Nineteen of 84 patients (89 hemispheres) with MMD were in the CHS group. Patients with ischemic onset symptoms were more likely to develop CHS after surgery (P < 0.05). There were no significant differences in age, sex, and Suzuki stage between the 2 groups. For surgery ipsilateral vessels, peak velocity, average flow and forward volume of superficial temporal artery and average flow of external carotid artery and region of interest area of internal carotid artery in the CHS group were significantly lower compared with the non-CHS group (P < 0.05). For surgery contralateral vessels, forward volume of superficial temporal artery and external carotid artery in the CHS group was significantly lower compared with the non-CHS group (P < 0.05). CONCLUSIONS: Patients with MMD and ischemic onset symptoms are more likely to develop CHS after surgery. Preoperative phase-contrast MRI analysis may be helpful to predict CHS in patients with MMD after surgery.


Assuntos
Revascularização Cerebral/efeitos adversos , Doença de Moyamoya/diagnóstico por imagem , Neuroimagem/métodos , Cuidados Pré-Operatórios/métodos , Adulto , Idoso , Circulação Cerebrovascular , Feminino , Humanos , Imagem por Ressonância Magnética/métodos , Masculino , Pessoa de Meia-Idade , Doença de Moyamoya/cirurgia , Complicações Pós-Operatórias/epidemiologia , Fatores de Risco , Adulto Jovem
8.
Cancer Med ; 8(7): 3532-3543, 2019 07.
Artigo em Inglês | MEDLINE | ID: mdl-31074592

RESUMO

To develop a deep learning system based on 3D convolutional neural networks (CNNs), and to automatically predict EGFR-mutant pulmonary adenocarcinoma in CT images. A dataset of 579 nodules with EGFR mutation status labels of mutant (Mut) or wild-type (WT) was retrospectively analyzed. A deep learning system, namely 3D DenseNets, was developed to process 3D patches of nodules from CT data, and learn strong representations with supervised end-to-end training. The 3D DenseNets were trained with a training subset of 348 nodules and tuned with a development subset of 116 nodules. A strong data augmentation technique, mixup, was used for better generalization. We evaluated our model on a holdout subset of 115 nodules. An independent public dataset of 37 nodules from the cancer imaging archive (TCIA) was also used to test the generalization of our method. Conventional radiomics analysis was also performed for comparison. Our method achieved promising performance on predicting EGFR mutation status, with AUCs of 75.8% and 75.0% for our holdout test set and public test set, respectively. Moreover, strong relations were found between deep learning feature and conventional radiomics, while deep learning worked through an enhanced radiomics manner, that is, deep learned radiomics (DLR), in terms of robustness, compactness and expressiveness. The proposed deep learning system predicts EGFR-mutant of lung adenocarcinomas in CT images noninvasively and automatically, indicating its potential to help clinical decision-making by identifying eligible patients of pulmonary adenocarcinoma for EGFR-targeted therapy.

9.
Eur Radiol ; 29(10): 5423-5430, 2019 Oct.
Artigo em Inglês | MEDLINE | ID: mdl-30903336

RESUMO

OBJECTIVE: To retrospectively analyze the computed tomography (CT) features in patients with pre-invasive lesions and early-stage lung adenocarcinoma and to explore the correlation between tumor morphological changes and pathological diagnoses. MATERIALS AND METHODS: CT morphological characteristics in 2106 patients with pre-invasive (stage 0) and early-stage (stage I) lung adenocarcinoma were analyzed; lesions were confirmed by surgical pathology. Based on the morphological characteristics, the lesions were divided into eight types: I (cotton ball, ground-glass nodules), II (solid fill), III (granular), IV (dendriform), V (bubble-like lucencies), VI (alveolate or honeycomb), VII (scar-like), and VIII (notched or umbilication). The different distributions of eight morphological types in pathological types of the lesions and subtypes of invasive adenocarcinoma were analyzed by chi-squared or Fisher's exact test. Correlation between the percentage of ground-glass opacity in the lesions and pathology types were analyzed by two-tailed Pearson's test. RESULTS: A negative correlation was observed between the pathological types and proportion of ground-glass component in the lesions (p < 0.001 and r = - 0.583). Significant differences in morphological characteristics among various pathological types of pre-invasive lesions and early lung adenocarcinomas were observed (p < 0.05). Furthermore, among the different pathological subtypes of stage I invasive adenocarcinoma, the differences in their manifestation as morphological types I, II, III, and VI were statistically significant (p < 0.05). CONCLUSION: The eight types of morphological classification of pre-invasive lesions and early-stage (stage 0 or stage I) lung adenocarcinoma has different pathological bases, and morphological classification may be useful for the diagnosis and differential diagnosis of lung adenocarcinoma. KEY POINTS: • CT morphological classification of pre-invasive lesions and lung adenocarcinoma is intuitive. • CT morphological classification characterizes morphological changes of the entire lesion. • Different pathological types of lung adenocarcinoma have different morphological features.


Assuntos
Adenocarcinoma de Pulmão/diagnóstico por imagem , Neoplasias Pulmonares/diagnóstico por imagem , Adenocarcinoma de Pulmão/patologia , Adulto , Idoso , Idoso de 80 Anos ou mais , Diagnóstico Diferencial , Feminino , Humanos , Neoplasias Pulmonares/patologia , Masculino , Pessoa de Meia-Idade , Invasividade Neoplásica , Estadiamento de Neoplasias , Estudos Retrospectivos , Tomografia Computadorizada por Raios X/métodos , Adulto Jovem
10.
Eur J Radiol ; 112: 161-168, 2019 Mar.
Artigo em Inglês | MEDLINE | ID: mdl-30777206

RESUMO

The aim of the present study was to develop and validate a radiomics-based nomogram for differentiation of pre-invasive lesions from invasive lesions that appearing as ground-glass opacity nodules (GGNs) ≤10 mm (sub-centimeter) in diameter at CT. A total of 542 consecutive patients with 626 pathologically confirmed pulmonary subcentimeter GGNs were retrospectively studied from October 2011 to September 2017. All the GGNs were divided into a training set (n = 334) and a validation set (n = 292). Researchers extracted 475 radiomics features from the plain CT images; a radiomics signature was constructed with the least absolute shrinkage and selection operator (LASSO) based on multivariable regression in the training set. Based on the multivariable logistic regression model, a radiomics nomogram was developed in the training set. The performance of the nomogram was evaluated with respect to its calibration, discrimination, and clinical-utility and this was assessed in the validation set. The constructed radiomics signature, which consisted of 15 radiomics features, was significantly associated with the invasiveness of subcentimeter GGNs (P < 0.0001 for both training set and validation set). To build the nomogram model, radiomics signature and mean CT value were used. The nomogram model demonstrated good discrimination and calibration in both training set (C-index, 0.716 [95% CI, 0.632 to 0.801]) and validation set (C-index, 0.707 [95% CI, 0.625 to 0.788]). Decision curve analysis (DCA) indicated that radiomics-based nomogram was clinically useful. A radiomics-based nomogram that incorporates both radiomics signature and mean CT value is constructed in the study, which can be conveniently used to facilitate the preoperative individualized prediction of the invasiveness in patients with subcentimeter GGNs.


Assuntos
Adenocarcinoma de Pulmão/patologia , Neoplasias Pulmonares/patologia , Nomogramas , Adenocarcinoma de Pulmão/diagnóstico por imagem , Adolescente , Adulto , Idoso , Idoso de 80 Anos ou mais , Feminino , Humanos , Modelos Logísticos , Neoplasias Pulmonares/diagnóstico por imagem , Masculino , Pessoa de Meia-Idade , Invasividade Neoplásica/diagnóstico por imagem , Invasividade Neoplásica/patologia , Estudos Retrospectivos , Tomografia Computadorizada por Raios X/métodos , Adulto Jovem
11.
Diagn Interv Radiol ; 25(1): 42-49, 2019 Jan.
Artigo em Inglês | MEDLINE | ID: mdl-30644367

RESUMO

PURPOSE: We aimed to analyze multislice computed tomography (MSCT) imaging features of vasculitis in immunoglobulin G4-related disease (IgG4-RD). METHODS: In this retrospective study, we diagnosed 21 definite or possible IgG4-RD patients (71.4% male; mean age, 52.1±4.5 years) with vasculitis by MSCT and pathologic examination. The clinical background, laboratory analysis, pathologic results, CT images, and response to therapy were assessed and analyzed. RESULTS: Under enhanced MSCT, 50 vasculitic lesions were identified and were divided into five types (types A-E) according to the CT findings on the basis of luminal changes. There were more vasculitic lesions observed below the diaphragm (n=30) than above it (n=20). Aneurysms and aneurysmal dilatation were more likely to be found in the aortaventralis (n=5), aortic arch (n=3) and iliac arteries (n=3). Most of the vascular lesions were characterized by wall thickening with a normal lumen (n=15) and slight stenosis (n=22). The affected vascular walls were all thickened between 4 and 18 mm. The walls of 19 patients (90.4%) were well circumscribed. The wall thickness of the aorta, including the aortaventralis and aortic arch, was more notable than that of the other vascular sites. Fourteen patients were followed up for 2-24 months. Wall thickness decreased in all cases. The average maximum thicknesses before and after therapy were 12.2±2.7 mm and 6.1±1.8 mm, which were significantly different (P < 0.001). The lumens of two patients were found to be slightly enlarged, while those of the other cases were unchanged after therapy. CONCLUSION: IgG4-RD vascular lesions can be divided into five types, which are more likely to be present in the aorta and its main branches, and can rapidly diminish after steroid therapy. The lumen may be unchanged or slightly enlarged.


Assuntos
Doença Relacionada a Imunoglobulina G4/diagnóstico por imagem , Imunoglobulina G/metabolismo , Tomografia Computadorizada Multidetectores/métodos , Doenças Vasculares/diagnóstico por imagem , Vasculite/diagnóstico por imagem , Assistência ao Convalescente , Biomarcadores/sangue , Feminino , Humanos , Doença Relacionada a Imunoglobulina G4/complicações , Doença Relacionada a Imunoglobulina G4/tratamento farmacológico , Doença Relacionada a Imunoglobulina G4/patologia , Masculino , Pessoa de Meia-Idade , Estudos Retrospectivos , Esteroides/uso terapêutico , Doenças Vasculares/tratamento farmacológico , Doenças Vasculares/imunologia , Doenças Vasculares/patologia , Vasculite/tratamento farmacológico , Vasculite/imunologia , Vasculite/patologia
12.
Br J Radiol ; 92(1094): 20180204, 2019 Feb.
Artigo em Inglês | MEDLINE | ID: mdl-30260240

RESUMO

OBJECTIVE: To explore the CT characteristics of small lung nodules and improve the diagnosis of pulmonary ground-glass nodules less than 10 mm in size. METHODS: We retrospectively analyzed CT images of 161 pulmonary nodules (less than 10 mm in size) with spiculation, lobulation, vacuoles, and pleural indentation and compared these images with pathological results or follow-up CT images. The relationships between the ground-glass nodules (GGNs) and blood vessels were observed. The GGN-vessel relationship was divided into four types, Type I (pass-by), Type II (pass-through), Type III (distorted/dilated), Type IV (complicated). The vessels traveling through a GGN were divided into three categories, category A (arteries), category B (veins), category C (arteries and veins). RESULTS: 161 GGNs were divided into three groups (benign group, pre-invasive group, and adenocarcinoma group) according to their pathological diagnosis. Significant differences in density of nodules were observed among the three different groups (p < 0.05). Significant differences in the shape (round/round-like or not) of the nodules were observed between the benign group and the pre-invasive group and between the pre-invasive group and the adenocarcinoma group (p < 0.05). No significant differences in the presence of vacuoles were observed between the benign group and the pre-invasive group or between the pre-invasive group and the adenocarcinoma group (p >0.05), but a significant difference was observed between the benign group and the adenocarcinoma group (p < 0.05). The differences in the vascularization of the lesions among the three groups were statistically significant (p < 0.05). No significant differences or correlations were observed between vascular categories and GGN groups (p > 0.05). CONCLUSION: For subcentimeter nodules, mixed GGNs with vacuoles, well-defined border, combined with Type III or Type IV GGN-vessel relationship may strongly suggest malignant. ADVANCES IN KNOWLEDGE: Previous studies mainly focused on CT diagnosis of pulmonary nodules (≤ 3 cm in diameter), but this study focused on ground-glass nodules less than 10 mm in diameter, which had not been fully studied. For subcentimeter nodules, mixed GGNs with vacuoles, well-defined border, especially the GGN-vessel relationship manifest as Type III (distorted/dilated) or Type IV (complicated) may strongly suggest malignant.


Assuntos
Neoplasias Pulmonares/diagnóstico por imagem , Pulmão/diagnóstico por imagem , Nódulo Pulmonar Solitário/diagnóstico por imagem , Tomografia Computadorizada por Raios X , Adulto , Idoso , Idoso de 80 Anos ou mais , Diagnóstico Diferencial , Feminino , Humanos , Pulmão/patologia , Neoplasias Pulmonares/patologia , Masculino , Pessoa de Meia-Idade , Estudos Retrospectivos , Nódulo Pulmonar Solitário/patologia
13.
Cancer Res ; 78(24): 6881-6889, 2018 12 15.
Artigo em Inglês | MEDLINE | ID: mdl-30279243

RESUMO

: Identification of early-stage pulmonary adenocarcinomas before surgery, especially in cases of subcentimeter cancers, would be clinically important and could provide guidance to clinical decision making. In this study, we developed a deep learning system based on 3D convolutional neural networks and multitask learning, which automatically predicts tumor invasiveness, together with 3D nodule segmentation masks. The system processes a 3D nodule-centered patch of preprocessed CT and learns a deep representation of a given nodule without the need for any additional information. A dataset of 651 nodules with manually segmented voxel-wise masks and pathological labels of atypical adenomatous hyperplasia (AAH), adenocarcinomas in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive pulmonary adenocarcinoma (IA) was used in this study. We trained and validated our deep learning system on 523 nodules and tested its performance on 128 nodules. An observer study with 2 groups of radiologists, 2 senior and 2 junior, was also investigated. We merged AAH and AIS into one single category AAH-AIS, comprising a 3-category classification in our study. The proposed deep learning system achieved better classification performance than the radiologists; in terms of 3-class weighted average F1 score, the model achieved 63.3% while the radiologists achieved 55.6%, 56.6%, 54.3%, and 51.0%, respectively. These results suggest that deep learning methods improve the yield of discriminative results and hold promise in the CADx application domain, which could help doctors work efficiently and facilitate the application of precision medicine. SIGNIFICANCE: Machine learning tools are beginning to be implemented for clinical applications. This study represents an important milestone for this emerging technology, which could improve therapy selection for patients with lung cancer.


Assuntos
Adenocarcinoma/diagnóstico por imagem , Aprendizado Profundo , Processamento de Imagem Assistida por Computador/métodos , Imageamento Tridimensional/métodos , Neoplasias Pulmonares/diagnóstico por imagem , Tomografia Computadorizada por Raios X , Adolescente , Adulto , Idoso , Idoso de 80 Anos ou mais , Área Sob a Curva , Tomada de Decisões , Diagnóstico por Computador , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Mutação , Invasividade Neoplásica , Redes Neurais de Computação , Medicina de Precisão , Prognóstico , Estudos Retrospectivos , Adulto Jovem
14.
J Comput Assist Tomogr ; 42(4): 614-622, 2018.
Artigo em Inglês | MEDLINE | ID: mdl-29613988

RESUMO

OBJECTIVE: The objective of this study was to investigate the correlation between dual-energy computed tomography (DECT)-based iodine quantitation and fluorine-18 fluorodeoxyglucose (F-FDG) positron emission tomography (PET)/computed tomography (CT) imaging for response evaluation of lung cancers to treatment. METHODS: In this prospective study, a total of 32 pairs of DECT and F-FDG PET/CT imaging acquired consecutively from 13 patients with primary or metastatic lung cancers receiving either radiotherapy alone or chemoradiotherapy were analyzed. Imaging examinations were performed before, immediately, and no later than 6 months after treatment for response evaluation. Iodine-related parameters including the total iodine uptake (TIU) and vital volume (VIV) from DECT and metabolic metrics such as the standardized uptake value normalized to lean body mass (SULpeak), metabolic tumor volume (MTV), and the total lesion glycolysis (TLG) from F-FDG-PET/CT were generated and measured by semiautomatic approaches. Dual-energy CT and PET/CT metrics were calculated and followed up with comparison with response evaluation criteria in solid tumors (RECIST). RESULTS: Analysis of pretreatment imaging data revealed a strong correlation between DECT metrics (RECIST, TIU, and VIV) and F-FDG PET/CT metrics (MTV, TLG) with coefficients of R ranging from 0.86 to 0.90 (P < 0.01). With the delivery of treatment, all measured DECT and PET/CT metrics significantly decreased whereas the descending amplitude in RECIST was significantly smaller than that of the remaining parameters (P < 0.05). During follow-up examinations, both metrics followed a similar changing pattern. Overall, strong consistency was found between RECIST, TIU, VIV and SULpeak, MTV, TLG (R covers 0.78-0.96, P < 0.05). CONCLUSIONS: Semiautomatic iodine-related quantitation in DECT correlated well with metabolism-based measurements in F-FDG PET/CT, suggesting that DECT-based iodine quantitation might be a feasible substitute for assessment of lung cancer response to chemoradiotherapy/radiotherapy with comparison with F-FDG PET/CT.


Assuntos
Quimiorradioterapia , Fluordesoxiglucose F18/farmacocinética , Iodo/farmacocinética , Neoplasias Pulmonares/diagnóstico por imagem , Neoplasias Pulmonares/terapia , Tomografia Computadorizada com Tomografia por Emissão de Pósitrons/métodos , Tomografia Computadorizada por Raios X/métodos , Adulto , Idoso , Idoso de 80 Anos ou mais , Feminino , Humanos , Pulmão/diagnóstico por imagem , Masculino , Pessoa de Meia-Idade , Estudos Prospectivos , Imagem Radiográfica a Partir de Emissão de Duplo Fóton/métodos , Compostos Radiofarmacêuticos/farmacocinética , Resultado do Tratamento
15.
IEEE J Biomed Health Inform ; 22(1): 161-172, 2018 01.
Artigo em Inglês | MEDLINE | ID: mdl-28500014

RESUMO

OBJECTIVE: an innovative concept and method is introduced to use a 3-D anatomical graphic pattern called visual patient (VP) visually to index, represent, and render the medical diagnostic records (MDRs) of a patient, so that a doctor can quickly learn the current and historical medical status of the patient by manipulating VP. The MDRs can be imaging diagnostic reports and DICOM images, laboratory reports and clinical summaries which can have clinical information relating to medical status of human organs or body parts. METHODS: the concept and method included three steps. First, a VP data model called visual index object (VIO) and a VP graphic model called visual anatomic object (VAO) were introduced. Second, a series of processing methods of parsing and extracting key information from MDRs were used to fill the attributes of the VIO model of a patient. Third, a VP system (VPS) was designed to map VIO to VAO, to create a VP instance for each patient. RESULTS: a prototype VPS has been implemented in a simulated hospital PACS/RIS integrated environment. Two evaluation results showed that more than 70% participating radiologists would like to use the VPS in their radiological imaging tasks, and the efficiency of using VPS to review the tested patients' MDRs was 2.24 times higher than that of using PACS/RIS, while the average accuracy by using PACS/RIS was better than that by using VPS; however, this difference was only about 4%. CONCLUSION: the developed VPS can show the medical status of patient organs/sub-organs with 3-D anatomical graphic pattern and will be welcomed by radiologists with better efficiency in reviewing the patients' MDRs and with acceptable accuracy. SIGNIFICANCE: the VP introduces a new way for medical professionals to access and interact with a huge amount of patient records with better efficiency in the big data era.


Assuntos
Gráficos por Computador , Registros Eletrônicos de Saúde , Imageamento Tridimensional , Modelos Biológicos , Adulto , Humanos , Radiologistas
16.
Medicine (Baltimore) ; 96(43): e7742, 2017 Oct.
Artigo em Inglês | MEDLINE | ID: mdl-29068976

RESUMO

This study aimed to explore the value of contrast-enhanced computed tomography (CT) scans in the differential diagnosis of atypical adenomatous hyperplasia, adenocarcinoma in situ, minimal invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC), which manifested as ground glass nodules (GGNs) or mixed GGNs.Unenhanced and contrast-enhanced presurgical CT images of 136 cases of GGNs were compared. The nodules were diagnosed based on their solid portions and maximum dimensions, and the findings obtained using both contrast-enhanced and unenhanced CT images were analyzed in corroboration with the pathological diagnosis.Most (53/56) preinvasive nodules showed increased mean CT values after contrast administration. In the MIA group, after contrast administration, enlargements of the solid portions were seen in 48 nodules (48/60), and the elevation of the mean CT value was observed in 12 nodules (12/60). The vast majority of IAC nodules (29/30) showed enlargement of the solid portions after contrast administration. Moreover, for group differences, findings obtained using this approach statistically match the pathological findings.Contrast-enhanced CT scans are more useful than unenhanced CT scans for the diagnosis of GGNs, especially for pure GGNs, before surgery.


Assuntos
Adenocarcinoma/diagnóstico por imagem , Adenocarcinoma/patologia , Neoplasias Pulmonares/diagnóstico por imagem , Neoplasias Pulmonares/patologia , Tomografia Computadorizada por Raios X , Adenocarcinoma in Situ/diagnóstico por imagem , Adenocarcinoma in Situ/patologia , Adenocarcinoma de Pulmão , Meios de Contraste , Diagnóstico Diferencial , Humanos , Hiperplasia/diagnóstico por imagem , Hiperplasia/patologia , Pulmão/diagnóstico por imagem , Pulmão/patologia , Invasividade Neoplásica , Intensificação de Imagem Radiográfica , Estudos Retrospectivos , Tomografia Computadorizada por Raios X/métodos
17.
Acad Radiol ; 24(12): 1482-1490, 2017 12.
Artigo em Inglês | MEDLINE | ID: mdl-28800951

RESUMO

RATIONALE AND OBJECTIVES: The study aimed to prospectively evaluate the radiation dose reduction potential and image quality (IQ) of a high-concentration contrast media (HCCM) injection protocol in combination with a low tube current (mAs) in coronary computed tomography angiography. MATERIALS AND METHODS: Eighty-one consecutive patients (mean age: 62 years; 34 females; body mass index: 18-31) were included and randomized-assigned into two groups. All computed tomography (CT) examinations were performed in two groups with the same tube voltage (100 kV), flow rate of contrast medium (5.0 mL/s), and iodine dose (22.8 g). An automatic mAs and low concentration contrast medium (300 mgI/mL) were used in group A, whereas effective mAs was reduced by a factor 0.6 along with HCCM (400 mgI/mL) in group B. Radiation dose was assessed (CT dose index [CTDIvol] and dose length product), and vessel-based objective IQ for various regions of interest (enhancement, noise, signal-to-noise ratio, and contrast-to-noise ratio), subjective IQ, noise, and motion artifacts were analyzed overall and vessel-based with a 5-point Likert scale. RESULTS: The CT attenuation of coronary arteries and image noise in group B were significantly higher than those in group A (ranges: 507.5-548.1 Hounsfield units vs 407.5-444.5 Hounsfield units; and 20.3 ± 8.6 vs 17.7 ± 8.0) (P ≤ 0.0166). There was no significant difference between the two groups in signal-to-noise ratio, contrast-to-noise ratio, and subjective IQ of coronary arteries (29.4-31.7, 30.0-37.0, and medium score of 5 in group A vs 29.4-32.4, 27.7-36.3, and medium score of 5 in group B, respectively, P ≥ 0.1859). Both mean CTDIvol and dose length product in group B were 58% of those of group A. CONCLUSIONS: HCCM combined with low tube current allows dose reduction in coronary computed tomography angiography and does not compromise IQ.


Assuntos
Angiografia por Tomografia Computadorizada , Meios de Contraste/administração & dosagem , Angiografia Coronária , Iohexol/análogos & derivados , Iopamidol/análogos & derivados , Doses de Radiação , Artefatos , Protocolos Clínicos , Feminino , Humanos , Iohexol/administração & dosagem , Iopamidol/administração & dosagem , Masculino , Pessoa de Meia-Idade , Estudos Prospectivos
18.
Eur J Radiol ; 85(6): 1208-18, 2016 Jun.
Artigo em Inglês | MEDLINE | ID: mdl-27161072

RESUMO

OBJECTIVE: To characterize solitary pulmonary nodules (SPN) in terms of perfusion parameters using low-dose volume perfusion computed tomography (VPCT) and two different calculation methods, compare their results, look for interobserver agreement of measurements and investigate the feasibility of low-dose VPCT. MATERIALS AND METHODS: This study was approved by the local Institutional Review Board and all patients provided written informed consent. Seventy-one patients (mean age 60.8 years ±9.6) with solitary pulmonary nodules were enrolled. Low-dose VPCT was performed for 38.63s covering the involved lung (70kV, 120mAs, 22 consecutive volume measurements, 50mL iodinated contrast, flow rate 4mL/s). Mean blood flow (BF), blood volume (BV) and k-trans were determined both with the maximum slope+Patlak vs. deconvolution method. Additionally, the difference of VPCT parameters between different type lesions and normal tissue was analyzed. Interobserver agreement for all perfusion parameters was calculated using intraclass correlation coefficients (ICC). The effective radiation dose of the VPCT and the total CT scan protocol were recorded. All CT findings were histologically confirmed by surgical intervention. RESULTS: The mean lesion size was 18.6mm. Interobserver agreement measure with ICC shows high agreement between the measurements (κ=0.85). The effective radiation dose of the VPCT was 9.3mSv. The mean perfusion values for BF, BV and k-trans of 120.6mL/100g tissue/', 11.6mL/100g tissue/', and 18.5mL/100g tissue/' for the deconvolution method, and 50.1mL/100g tissue/', 11.6mL/100g tissue/' and 24.3mL/100g tissue/' for the maximum slope+Patlak method, which were significantly higher than those of normal muscle (20.7mL/100g tissue/', 2.6mL/100g tissue/', and 7.6mL/100g tissue/' for the deconvolution method and 10.9mL/100g tissue/', 3.1mL/100g tissue/' and 8.1mL/100g tissue/' for the maximum slope method). The best overall correlation between calculation methods was achieved for measurements of BF. CONCLUSION: The low-dose volume perfusion CT of the solitary pulmonary nodules can effectively reduce the radiation dose and non-invasively assess perfusion of SPN within the entire lesion volume.


Assuntos
Tomografia Computadorizada de Feixe Cônico/métodos , Nódulo Pulmonar Solitário/diagnóstico por imagem , Adulto , Idoso , Idoso de 80 Anos ou mais , Meios de Contraste , Estudos de Viabilidade , Feminino , Humanos , Iopamidol , Pulmão/irrigação sanguínea , Pulmão/diagnóstico por imagem , Masculino , Pessoa de Meia-Idade , Doses de Radiação , Intensificação de Imagem Radiográfica , Reprodutibilidade dos Testes , Sensibilidade e Especificidade
19.
Radiol Med ; 121(4): 279-90, 2016 Apr.
Artigo em Inglês | MEDLINE | ID: mdl-26612322

RESUMO

While conventional ultrasound and laboratory testing can differentiate most thyroid conditions and malignancies, spectral dual-energy computed tomography (DECT) provides molecular data potentially useful in differential diagnosis of small, complex, and partially obscured hemorrhaged nodules with recent bleeding. To demonstrate that Gemstone Spectral Imaging (GSI) analysis of DECT data differentiates challenging benign thyroid nodule conditions in patients with intranodular hemorrhages. A retrospective study was conducted of 30 intranodular hemorrhage patients from 2010 to 2013, including 18 that underwent surgery and provided thyroid tissues for prospective histological analysis. Iodine and water content were determined in patient CT scans and ex vivo tissue specimen scans by reconstruction of raw CT data at 65 keV (optimal contrast). Slope of spectral curve (λ HU), effective atomic number (Z eff), and final pathological diagnosis were recorded. Iodine content, water content, Z eff, and λ HU significantly varied by region (intranodular hemorrhage, solid thyroid nodule, and adjacent margins) in patients and tissue specimens (P < 0.05). Intranodular hemorrhage exhibited elevated water concentrations (~1100 mg/mL), suggesting a practical threshold of 1075 mg/mL for differentiating intra-plaque hemorrhage and solid nodular regions. Spectral CT provided diagnostic information in 14 thyroid adenomas and four goiters (histologically confirmed in donor specimens), and eight thyroid adenomas and four nodular goiters based on clinical diagnosis. Diagnostically useful regional characteristic of intranodular hemorrhage in the thyroid was visualized via spectral CT employing material decomposition, potentially yielding additional molecular data about complex lesion characteristics no apparent in conventional imaging or laboratory methods.


Assuntos
Hemorragia/diagnóstico por imagem , Nódulo da Glândula Tireoide/diagnóstico por imagem , Tomografia Computadorizada por Raios X , Adulto , Feminino , Hemorragia/complicações , Humanos , Masculino , Pessoa de Meia-Idade , Estudos Retrospectivos , Nódulo da Glândula Tireoide/complicações
20.
J Thorac Dis ; 7(9): 1606-15, 2015 Sep.
Artigo em Inglês | MEDLINE | ID: mdl-26543608

RESUMO

BACKGROUND: To evaluate the information gain by the application of both non-contrast and contrast enhanced computed tomography (CT) with extended mediastinal display window settings in the evaluation of pure ground glass nodules (pGGNs) and or mixed ground glass nodules (mGGNs) in the context of pre-invasive or early stage lung adenocarcinoma. METHODS: One hundred and fifty patients with ground glass nodules (GGNs) and mGGNs, with contrast enhanced CT scans within 2 weeks of thoracic surgery were included in the study. Quantitative evaluation of all nodules was performed in a conventional mediastinal window (CMW) and an extended mediastinal window (EMW) both on non-contrast images and contrast-enhanced images. RESULTS: Contrast-enhanced images with CMW demonstrated amplification of solid portion in 23 (43%), 41 (77%) with EMW out of 53 minimally invasive adenocarcinoma (MIA) nodules, and in 34 of 37 (91%) of invasive adenocarcinoma (IAC) nodules. Using the increase in size of solid portion of the nodule measured on the enhanced CT images with EMW, area under the receiver operating characteristic (ROC) curve of 0.872 and 0.899 was utilized for differentiating between the pre-invasive nodules and MIA and between MIA and IAC nodules, respectively. Statistically significant differences existed between the pre-invasive and the MIA groups, and MIA and the IAC groups in smaller nodules (P<0.01). CONCLUSIONS: Comparative quantitative analysis of the pre and post contrast images can help differentiate between atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), MIAs, and IACs. Extension of the CT mediastinal window setting improves the evaluation of small GGNs, and can augment the diagnostic accuracy when evaluating small pGGNs and mGGNs.

SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA