Detecting abnormal thyroid cartilages on CT using deep learning
تشخیص غضروف غیر طبیعی تیروئید در CT با استفاده از یادگیری عمیق-2019
Purpose: The purpose of this study was to evaluate the performance of a deep learning algorithm in detecting abnormalities of thyroid cartilage from computed tomography (CT) examination. Materials and methods: A database of 515 harmonized thyroid CT examinations was used, of which information regarding cartilage abnormality was provided for 326. The process consisted of determining image abnormality and, from these preprocessed images, finding the best learning algorithm to appropriately characterize thyroid cartilage as normal or abnormal. CT images were cropped to be centered around the cartilage in order to focus on the relevant area. New images were generated from the originals by applying simple transformations in order to augment the database. Characterizations of cartilage abnormalities were made using transfer learning, by using the architecture of a pre-trained neural network called VGG16 and adapting the final layers to a binary classification problem. Results: The best algorithm yielded an area under the receiving operator characteristic curve (AUC) of 0.72 on a sample of 82 thyroid test images. The sensitivity and specificity of the abnormality detection were 83% and 64% at the best threshold, respectively. Applying the model on another independent sample of 189 new thyroid images resulted in an AUC of 0.70. Conclusion: This study demonstrates the feasibility of using a deep learning-based abnormality detection system to evaluate thyroid cartilage from CT examinations. However, although promising results, the model is not yet able to match an expert’s diagnosis.
KEYWORDS : Thyroid cartilage | Artificial intelligence (AI) | Deep learning | Post-mortem computed tomography (CT) | Larynx
Gender differences in the diagnostic performance of machine learning coronary CT angiography-derived fractional flow reserve -results from the MACHINE registry
تفاوت های جنسیتی در عملکرد تشخیصی یادگیری دستگاه یادگیری عروق کرونر CT-نتیجه حاصل از کسری جریان کسری ناشی از آنژیوگرافی از رجیستری ماشین-2019
Purpose: This study investigated the impact of gender differences on the diagnostic performance of machine-learning based coronary CT angiography (cCTA)-derived fractional flow reserve (CT-FFRML) for the detection of lesion-specific ischemia. Method: Five centers enrolled 351 patients (73.5% male) with 525 vessels in the MACHINE (Machine leArning Based CT angiograpHy derIved FFR: a Multi-ceNtEr) registry. CT-FFRML and invasive FFR≤0.80 were considered hemodynamically significant, whereas cCTA luminal stenosis ≥50% was considered obstructive. The diagnostic performance to assess lesion-specific ischemia in both men and women was assessed on a per-vessel basis. Results: In total, 398 vessels in men and 127 vessels in women were included. Compared to invasive FFR, CT-FFRML reached a sensitivity, specificity, positive predictive value, and negative predictive value of 78% (95%CI 72–84), 79% (95%CI 73–84), 75% (95%CI 69–79), and 82% (95%CI: 76–86) in men vs. 75% (95%CI 58–88), 81 (95%CI 72–89), 61% (95%CI 50–72) and 89% (95%CI 82–94) in women, respectively. CT-FFRML showed no statistically significant difference in the area under the receiver-operating characteristic curve (AUC) in men vs. women (AUC: 0.83 [95%CI 0.79–0.87] vs. 0.83 [95%CI 0.75–0.89], p=0.89). CT-FFRML was not superior to cCTA alone [AUC: 0.83 (95%CI: 0.75–0.89) vs. 0.74 (95%CI: 0.65–0.81), p=0.12] in women, but showed a statistically significant improvement in men [0.83 (95%CI: 0.79–0.87) vs. 0.76 (95%CI: 0.71–0.80), p=0.007]. Conclusions: Machine-learning based CT-FFR performs equally in men and women with superior diagnostic performance over cCTA alone for the detection of lesion-specific ischemia.
Keywords: Coronary artery disease | Machine learning | Spiral computed tomography | Fractional flow reserve
A new method for CF morphology distribution evaluation and CFRC property prediction using cascade deep learning
یک روش جدید برای ارزیابی توزیع مورفولوژی CF و پیش بینی ویژگی CFRC با استفاده از یادگیری عمیق آبشاری-2019
This work presents a deep-learning method to characterize the carbon fiber (CF) morphology distribution in carbon fiber reinforced cement-based composites (CFRC), predict the CFRC properties, and measure the contributions of different CF morphology distribution directly using X-ray images. Firstly, the components of CFRC in slices of X-ray images were segmented and identified using a fully convolutional network (FCN). Then the CF morphology distribution evaluation were conducted based on the results of the FCN. At last, the prediction of CFRC properties was realized using a cascade deep learning algorithm and CF morphology distribution results. The results showed that the FCN provided more reasonable segmentation results for each component in CFRC than traditional methods. CF clustered areas and CF bundles increased sharply with the increase of CF content, while uniformly dispersed CF areas showed the opposite trend. The cascade deep learning provided a method to predict the CFRC properties (e.g. resistivity and bending strength) using X-ray scanning images, which could also quantificationally measure the contributions of different CF morphology distribution to properties of the CFRC. Therefore, the proposed method could be regarded as a nondestructive and effective test for CFRC property evaluation.
Keywords: Carbon fiber reinforced cement-based | composites | Carbon fiber distribution | Computed tomography | Deep learning | Radial basis function network
Diagnostic Performance of a Novel Method for Fractional Flow Reserve Computed from Noninvasive Computed Tomography Angiography (NOVEL-FLOW Study)
کارایی تشخیصی یک روش جدید برای ذخایر جریان فراکشنال محاسبه شده از آنژیوگرافی توموگرافی کامپیوتری غیر تهاجمی (مطالعه NOVEL-FLOW)-2017
Coronary computed tomography angiography (CCTA)-derived fractional flow reserve from computed tomography (CT-FFR) may provide better diagnostic performance over CCTA alone, but the complexity of its method limits the use in clinical environment. The aim of the present study is to validate a newly developed vessel-length based computational fluid dynamics scheme for the computation of FFR based on CCTA data, compare them with invasively measured FFR, and evaluate its diagnostic performance with that of CCTA. One hundred seventeen patients from 4 medical institutions who had clinically indicated invasive coronary angiography for suspected coronary artery disease (CAD) were enrolled. Invasive FFR measurement was performed in 218 vessels and these measurements were regarded as the reference standard. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of CT-FFR on a per-vessel basis were 85.8%, 86.2%, 85.5%, 79.8%, and 90.3%, respectively, for CT-FFR £0.80, and 66.1%, 75.9%, 59.5%, 55.5%, and 78.8%, respectively, for CCTA ‡50%. A higher area under the receiver operating characteristic curve for CT-FFR was observed compared with CCTA (0.93 vs 0.74, p <0.0001). The CT-FFR and FFR correlated well (r [ 0.76, p <0.001) with slight underestimation by CT-FFR (0.014 – 0.077, p [ 0.007). With a novel method of vessellength based computational fluid dynamics scheme, CT-FFR can be performed at a personal computer enhancing its applicability in clinical situation. The diagnostic accuracy of CT-FFR for the detection of functionally significant CAD was good and was superior to that of CCTA within a population of suspected CAD. 2017 Elsevier Inc. All rights reserved. (Am J Cardiol 2017;120:362e368)
Computer-aided diagnosis of periapical cyst and keratocystic odontogenic tumor on cone beam computed tomography
تشخیص کامپیوتری از کیست پرایپیکال و تومور ادنتوژنیک کراتوسیستیک بر روی توموگرافی کامپیوتری پرتو مخروطی-2017
Article history:Received 2 May 2016Revised 15 April 2017Accepted 26 May 2017Keywords:Computer aided diagnosis Dental apical lesion ClassiﬁerCone beam computed tomography Periapical cyst and keratocystic odontogenic tumorVolumetric textural features Dental image datasetBackground and objectives: In this article, we propose a decision support system for effective classiﬁ- cation of dental periapical cyst and keratocystic odontogenic tumor (KCOT) lesions obtained via cone beam computed tomography (CBCT). CBCT has been effectively used in recent years for diagnosing dental pathologies and determining their boundaries and content. Unlike other imaging techniques, CBCT pro- vides detailed and distinctive information about the pathologies by enabling a three-dimensional (3D) image of the region to be displayed.Methods: We employed 50 CBCT 3D image dataset ﬁles as the full dataset of our study. These datasets were identiﬁed by experts as periapical cyst and KCOT lesions according to the clinical, radiographic and histopathologic features. Segmentation operations were performed on the CBCT images using viewer soft- ware that we developed. Using the tools of this software, we marked the lesional volume of interest and calculated and applied the order statistics and 3D gray-level co-occurrence matrix for each CBCT dataset. A feature vector of the lesional region, including 636 different feature items, was created from those statistics. Six classiﬁers were used for the classiﬁcation experiments.Results: The Support Vector Machine (SVM) classiﬁer achieved the best classiﬁcation performance with 100% accuracy, and 100% F-score (F1) scores as a result of the experiments in which a ten-fold cross vali- dation method was used with a forward feature selection algorithm. SVM achieved the best classiﬁcation performance with 96.00% accuracy, and 96.00% F1 scores in the experiments in which a split sample val- idation method was used with a forward feature selection algorithm. SVM additionally achieved the best performance of 94.00% accuracy, and 93.88% F1 in which a leave-one-out (LOOCV) method was used with a forward feature selection algorithm.Conclusions: Based on the results, we determined that periapical cyst and KCOT lesions can be classiﬁed with a high accuracy with the models that we built using the new dataset selected for this study. The studies mentioned in this article, along with the selected 3D dataset, 3D statistics calculated from the dataset, and performance results of the different classiﬁers, comprise an important contribution to the ﬁeld of computer-aided diagnosis of dental apical lesions.© 2017 Elsevier B.V. All rights reserved.
Keywords: Computer aided diagnosis | Dental apical lesion | Classifier | Cone beam computed tomography | Periapical cyst and keratocystic odontogenic | tumor | Volumetric textural features | Dental image dataset
An audit of image quality of three dental cone beam computed tomography units
ممیزی کیفیت تصویر پرتو محاسبه سه مخروط دندانپزشکی واحد توموگرافی-2016
Critical analysis of cone beam computed tomography (CBCT) image quality is recommended as part of a quality assurance program.1,2 There are few papers3 in the literature concerning subjective image quality on CBCT imaging.This study, performed as part of an audit, reviewed all images of the jaws performed on three different1Available online 28 June 2015CBCT units over a twelve month period. Images were graded according to an agreed standardreasons for image rejection recorded.andKeywords:Quality assuranceCone beam computed tomography Dental imagingThe results demonstrated that the main reasons for image rejection were motion artefact and prob- lems with ﬁeld of view size and positioning.The need for reducing the number of rejected images in order to optimize patient dose, and ways to achieve this, are discussed© 2015 The College of Radiographers. Published by Elsevier Ltd. All rights reserved.
Keywords: Quality assurance | Cone beam computed tomography | Dental imaging
تجزیه و تحلیل ساختار پارچه نساجی بر اساس استخراج خودکار اطلاعات موضعی نخ از تصویر سه بعدی توموگرافی کامپیوتری
سال انتشار: 2010 - تعداد صفحات فایل pdf انگلیسی: 6 - تعداد صفحات فایل doc فارسی: 15
در این مقاله، یک روش جدید برای استخراج غیر مخرب اطلاعات موضعی نخ برای محاسبه تحلیل ساختار پارچه نساجی از تصویر سه بعدی (3D) به دست آمده از اشعه ایکس محاسبه شده با (CT) پیشنهاد شده است. در این مقاله، دنباله ای از نقاط روی خط وسط هر نخ نمونه پارچه به عنوان اطلاعات موضعی نخ تعریف شده است. به منظور استخراج دنباله، ابتدا، جهت های نخ توسط مرتبط کننده وکسل های تصویر 3D CT با استفاده از مدل نخ بیش از کل تصویر 3D CT، برآورد شده است که از میدان برداری جهت های نخ هستند. سپس، اطلاعات موضعی نخ توسط بازسازی بردار نخ با استفاده از استخراج زمینه جهت نخ به دست آمده است. اثربخشی این روش تجربی با استفاده از تصویر 3D CT پارچه دو لایه بحث شده است.
کلمات کلیدی: پارچه نساجی | تجزیه و تحلیل ساختار | اشعه ایکس توموگرافی کامپیوتری | تصویر سه بعدی | استخراج
|مقاله ترجمه شده|