scholarly journals An Evaluation of Cellular Neural Networks for the Automatic Identification of Cephalometric Landmarks on Digital Images

2009 ◽  
Vol 2009 ◽  
pp. 1-12 ◽  
Author(s):  
Rosalia Leonardi ◽  
Daniela Giordano ◽  
Francesco Maiorana
Keyword(s):  
Neural Networks ◽  
Cellular Neural Networks ◽  
Automatic Identification ◽  
Cephalometric Analysis ◽  
X Rays ◽  
X Ray ◽  
Automatic Location ◽  
Landmark Identification ◽  
The Mean ◽  
Analogue To Digital

Several efforts have been made to completely automate cephalometric analysis by automatic landmark search. However, accuracy obtained was worse than manual identification in every study. The analogue-to-digital conversion of X-ray has been claimed to be the main problem. Therefore the aim of this investigation was to evaluate the accuracy of the Cellular Neural Networks approach for automatic location of cephalometric landmarks on softcopy of direct digital cephalometric X-rays. Forty-one, direct-digital lateral cephalometric radiographs were obtained by a Siemens Orthophos DS Ceph and were used in this study and 10 landmarks (N, A Point, Ba, Po, Pt, B Point, Pg, PM, UIE, LIE) were the object of automatic landmark identification. The mean errors and standard deviations from the best estimate of cephalometric points were calculated for each landmark. Differences in the mean errors of automatic and manual landmarking were compared with a 1-way analysis of variance. The analyses indicated that the differences were very small, and they were found at most within 0.59 mm. Furthermore, only few of these differences were statistically significant, but differences were so small to be in most instances clinically meaningless. Therefore the use of X-ray files with respect to scanned X-ray improved landmark accuracy of automatic detection. Investigations on softcopy of digital cephalometric X-rays, to search more landmarks in order to enable a complete automatic cephalometric analysis, are strongly encouraged.

scholarly journals X-ray image based pneumonia classification using convolutional neural networks

2020 ◽  
Vol 6 (20) ◽  
pp. 54-67
Author(s):  
Sarah Badr AlSumairi ◽  
Mohamed Maher Ben Ismail
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Network Architecture ◽  
Data Augmentation ◽  
Critical Factor ◽  
Classification Problem ◽  
X Rays ◽  
X Ray ◽  
Erroneous Decision ◽  
Short Period

Pneumonia is an infectious disease of the lungs. About one third to one half of pneumonia cases are caused by bacteria. Early diagnosis is a critical factor for a successful treatment process. Typically, the disease can be diagnosed by a radiologist using chest X-ray images. In fact, chest X-rays are currently the best available method for diagnosing pneumonia. However, the recognition of pneumonia symptoms is a challenging task that relies on the availability of expert radiologists. Such “human” diagnosis can be inaccurate and subjective due to lack of clarity and erroneous decision. Moreover, the error can increase more if the physician is requested to analyze tens of X-rays within a short period of time. Therefore, Computer-Aided Diagnosis (CAD) systems were introduced to support and assist physicians and make their efforts more productive. In this paper, we investigate, design, implement and assess customized Convolutional Neural Networks to overcome the image-based Pneumonia classification problem. Namely, ResNet-50 and DenseNet-161 models were inherited to design customized deep network architecture and improve the overall pneumonia classification accuracy. Moreover, data augmentation was deployed and associated with standard datasets to assess the proposed models. Besides, standard performance measures were used to validate and evaluate the proposed system.

Isoelastic Cementless Total Hip Replacement in Rheumatoid Arthritis. Long-Term Results

1992 ◽  
Vol 2 (2) ◽  
pp. 43-46
Author(s):  
U. Fusco ◽  
R. Capelli ◽  
A. Avai ◽  
M. Gerundini ◽  
L. Colombini ◽  
...  
Keyword(s):  
Rheumatoid Arthritis ◽  
Hip Replacement ◽  
The Other ◽  
Long Term Results ◽  
X Rays ◽  
X Ray ◽  
The Mean ◽  
Cementless Total Hip Replacement

Between 1980 and 1987 we have implanted 46 isoelastic cementless THR in 40 patients affected with rheumatoid arthritis. We have reviewed 38 hips clinically and by X-ray. The mean follow-up was 8,5 years. Harris hip scores ranged from 30.6 preoperatively to 73,4 post-operatively when reviewed. While on the other hand Merle D'Aubigné hip scores ranged from 7,06 pre-operatively to 15,59 post-operatively. All patients have been satisfied, and X-rays showed an improvement for both Charnely and Gruen X-ray score.

scholarly journals X-Rays from Wolf-Rayet Stars Observed by the Einstein Observatory

1982 ◽  
Vol 99 ◽  
pp. 589-595
Author(s):  
W. T. Sanders ◽  
J. P. Cassinelli ◽  
K. A. van der Hucht
Keyword(s):  
Mean Value ◽  
Spectral Energy ◽  
X Rays ◽  
X Ray ◽  
Total Luminosity ◽  
Hydrogen Emission Lines ◽  
Flux Level ◽  
The Mean ◽  
Einstein Observatory ◽  
Thin Source

Preliminary results of three X-ray surveys are presented. Out of a sample of 20 stars, X-rays were detected from four Wolf-Rayet stars and two 08f+ stars. The detected stars have about the same mean value as 0 stars for the X-ray to total luminosity ratio, LX/L = 10−7, but exhibit a much larger variation about the mean. The spectral energy distributions are also found to be like that of 0 stars in that they do not exhibit large attenuation of X-rays softer than 1 keV. This indicates that for both the 0 stars and WR stars much of the X-ray emission is coming from hot wisps or shocks in the outer regions of the winds and not from a thin source at the base of the wind. The general spectral shape and flux level place severe restrictions on models that attribute the lack of hydrogen emission lines to extremely high temperatures of the gas in the wind.

Mean-Square Exponential Stability of Stochastic Interval Cellular Neural Networks with Time-Varying Delay

2013 ◽  
Vol 760-762 ◽  
pp. 1742-1747
Author(s):  
Jin Fang Han
Keyword(s):  
Neural Networks ◽  
Exponential Stability ◽  
Cellular Neural Networks ◽  
Time Varying ◽  
Mean Square ◽  
Razumikhin Theorem ◽  
Time Varying Delay ◽  
Mean Square Exponential Stability ◽  
The Mean ◽  
Varying Delay

This paper is concerned with the mean-square exponential stability analysis problem for a class of stochastic interval cellular neural networks with time-varying delay. By using the stochastic analysis approach, employing Lyapunov function and norm inequalities, several mean-square exponential stability criteria are established in terms of the formula and Razumikhin theorem to guarantee the stochastic interval delayed cellular neural networks to be mean-square exponential stable. Some recent results reported in the literatures are generalized. A kind of equivalent description for this stochastic interval cellular neural networks with time-varying delay is also given.

The Utility of Preoperative Magnetic Resonance Imaging for Determining the Flexibility of Sagittal Imbalance

Neurosurgery ◽  
2017 ◽  
Vol 83 (3) ◽  
pp. 465-470
Author(s):  
Akshay Sharma ◽  
Sina Pourtaheri ◽  
Jason Savage ◽  
Iain Kalfas ◽  
Thomas E Mroz ◽  
...  
Keyword(s):  
Lumbar Lordosis ◽  
Surgical Planning ◽  
Pelvic Incidence ◽  
Sagittal Imbalance ◽  
X Rays ◽  
Resonance Imaging ◽  
Patient Counseling ◽  
X Ray ◽  
Preoperative Magnetic Resonance ◽  
The Mean

Abstract BACKGROUND Scoliosis X-rays are the gold standard for assessing preoperative lumbar lordosis; however, particularly for flexible lumbar deformities, it is difficult to predict from these images the extent of correction required, as standing radiographs cannot predict the thoracolumbar alignment after intraoperative positioning. OBJECTIVE To determine the utility of preoperative MRI in surgical planning for patients with flexible sagittal imbalance. METHODS We identified 138 patients with sagittal imbalance. Radiographic parameters including pelvic incidence and lumbar lordosis were obtained from images preoperatively. RESULTS The mean difference was 2.9° between the lumbar lordosis measured on supine MRI as compared to the intraoperative X-rays, as opposed to 5.53° between standing X-rays and intraoperative X-ray. In patients with flexible deformities (n = 24), the lumbar lordosis on MRI measured a discrepancy of 3.08°, as compared to a discrepancy of 11.46° when measured with standing X-ray. CONCLUSION MRI adequately determined which sagittal deformities were flexible. Furthermore, with flexible sagittal deformities, lumbar lordosis measured on MRI more accurately predicted the intraoperative lumbar lordosis than that measured on standing X-ray. The ability to preoperatively predict intraoperative lumbar lordosis with positioning helps with surgical planning and patient counseling regarding expectations and risks of surgery.

scholarly journals An Improved VGG16 Model for Pneumonia Image Classification

2021 ◽  
Vol 11 (23) ◽  
pp. 11185
Author(s):  
Zhi-Peng Jiang ◽  
Yi-Yang Liu ◽  
Zhen-En Shao ◽  
Ko-Wei Huang
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Image Recognition ◽  
Medical Image ◽  
Best Practice ◽  
Data Augmentation ◽  
X Rays ◽  
Neural Network Models ◽  
X Ray ◽  
Medical Image Recognition

Image recognition has been applied to many fields, but it is relatively rarely applied to medical images. Recent significant deep learning progress for image recognition has raised strong research interest in medical image recognition. First of all, we found the prediction result using the VGG16 model on failed pneumonia X-ray images. Thus, this paper proposes IVGG13 (Improved Visual Geometry Group-13), a modified VGG16 model for classification pneumonia X-rays images. Open-source thoracic X-ray images acquired from the Kaggle platform were employed for pneumonia recognition, but only a few data were obtained, and datasets were unbalanced after classification, either of which can result in extremely poor recognition from trained neural network models. Therefore, we applied augmentation pre-processing to compensate for low data volume and poorly balanced datasets. The original datasets without data augmentation were trained using the proposed and some well-known convolutional neural networks, such as LeNet AlexNet, GoogLeNet and VGG16. In the experimental results, the recognition rates and other evaluation criteria, such as precision, recall and f-measure, were evaluated for each model. This process was repeated for augmented and balanced datasets, with greatly improved metrics such as precision, recall and F1-measure. The proposed IVGG13 model produced superior outcomes with the F1-measure compared with the current best practice convolutional neural networks for medical image recognition, confirming data augmentation effectively improved model accuracy.

scholarly journals 1700 Using Intra-Operative Imaging as A Tool to Provide Peer Review, Scoring, Feedback and Education for Hand and Wrist Trauma Cases - A Quality Improvement Project

2021 ◽  
Vol 108 (Supplement_6) ◽  
Author(s):  
A Smith ◽  
A Thompson ◽  
P Stanier ◽  
J Rooker ◽  
I Lowdon
Keyword(s):  
Final Report ◽  
X Rays ◽  
Improvement Project ◽  
X Ray ◽  
Constructive Feedback ◽  
Wrist Trauma ◽  
Specific Feedback ◽  
The Mean ◽  
Mean Time

Abstract Aim To improve the efficiency of intraoperative hand trauma x-ray review, introduce a scoring system for quality of fixation achieved and use this as an education and feedback tool for trainees. Method A large QI project in 2019 demonstrated that intra-operative images taken using the mini C-arm were not being reviewed. In March 2020 the first QI cycle introduced an Access database to ensure that as cases were removed from the trauma board, they were added to a review list. Each x-ray was also scored in the trauma meeting (good, acceptable, poor) to assess the radiographic quality of fixation. The second QI cycle identified the named surgeon for each case, allowing surgeon specific feedback. Finally, we trained all qualified users of the mini C-arm to upload their own images after each case, reducing the mean time to upload. Results This QI project has improved the review rate of mini C-arm images from 30% to 100% and every x-ray is scored. The mean time to review images has reduced from 4 days to less than 24 hours. All trainees can access their scores for imaged hand and wrist trauma and receive written constructive feedback. Conclusions This project ensures appropriate and prompt review of all hand and wrist trauma cases allowing early identification of any concerns. All x-rays are scored, and this process acts as an educational prompt in the trauma meeting. A final report can be printed for each trainee which can be uploaded to their portfolio to document their surgical progress.

scholarly journals Transfer Learning to Detect COVID-19 Automatically from X-Ray Images Using Convolutional Neural Networks

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Mundher Mohammed Taresh ◽  
Ningbo Zhu ◽  
Talal Ahmed Ali Ali ◽  
Asaad Shakir Hameed ◽  
Modhi Lafta Mutar
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Transfer Learning ◽  
Convolutional Neural Networks ◽  
Detection Accuracy ◽  
X Rays ◽  
X Ray ◽  
Specificity And Sensitivity ◽  
X Ray Imaging ◽  
Novel Coronavirus

The novel coronavirus disease 2019 (COVID-19) is a contagious disease that has caused thousands of deaths and infected millions worldwide. Thus, various technologies that allow for the fast detection of COVID-19 infections with high accuracy can offer healthcare professionals much-needed help. This study is aimed at evaluating the effectiveness of the state-of-the-art pretrained Convolutional Neural Networks (CNNs) on the automatic diagnosis of COVID-19 from chest X-rays (CXRs). The dataset used in the experiments consists of 1200 CXR images from individuals with COVID-19, 1345 CXR images from individuals with viral pneumonia, and 1341 CXR images from healthy individuals. In this paper, the effectiveness of artificial intelligence (AI) in the rapid and precise identification of COVID-19 from CXR images has been explored based on different pretrained deep learning algorithms and fine-tuned to maximise detection accuracy to identify the best algorithms. The results showed that deep learning with X-ray imaging is useful in collecting critical biological markers associated with COVID-19 infections. VGG16 and MobileNet obtained the highest accuracy of 98.28%. However, VGG16 outperformed all other models in COVID-19 detection with an accuracy, F1 score, precision, specificity, and sensitivity of 98.72%, 97.59%, 96.43%, 98.70%, and 98.78%, respectively. The outstanding performance of these pretrained models can significantly improve the speed and accuracy of COVID-19 diagnosis. However, a larger dataset of COVID-19 X-ray images is required for a more accurate and reliable identification of COVID-19 infections when using deep transfer learning. This would be extremely beneficial in this pandemic when the disease burden and the need for preventive measures are in conflict with the currently available resources.

scholarly journals The reflexion of X-rays from the ‘anti-phase nuclei’ of AuCu 3

1943 ◽  
Vol 181 (987) ◽  
pp. 360-368 ◽  
Keyword(s):  
Mean Value ◽  
X Rays ◽  
Integral Breadth ◽  
X Ray ◽  
Structure Factors ◽  
Unit Cells ◽  
The Mean

Jones & Sykes have observed that the superlattice lines in X-ray photographs of AuCu 3 , are not always as sharp as the main lines, and that the broadening depends markedly on the indices of the line. They explain these phenomena by assuming that the crystals of AuCu 3 contain many ‘anti-phase nuclei’ in which the superlattice is organized in different ways. In the present paper it is shown that the integral breadth of a reflexion from a crystal in which all the unit cells are not the same is λ J 0 /cos θ ƒ J t dt , where J t is the mean value of the product FF* of the structure factors of two unit cells separated a distance t in the hkl direction. Detailed calculations are made of the broadening to be expected from five different ways in which the nuclei can ‘change step’. Closest agreement with the observed broadening is given by a manner of ‘ changing step ’ in which the gold atoms avoid one another.

scholarly journals Asymptotic Behavior of Switched Stochastic Delayed Cellular Neural Networks via Average Dwell Time Method

2013 ◽  
Vol 2013 ◽  
pp. 1-9 ◽  
Author(s):  
Hanfeng Kuang ◽  
Jinbo Liu ◽  
Xi Chen ◽  
Jie Mao ◽  
Linjie He
Keyword(s):  
Neural Networks ◽  
Asymptotic Behavior ◽  
Dwell Time ◽  
Sufficient Conditions ◽  
Average Dwell Time ◽  
Cellular Neural Networks ◽  
Mixed Delays ◽  
Time Varying ◽  
Mean Square ◽  
The Mean

The asymptotic behavior of a class of switched stochastic cellular neural networks (CNNs) with mixed delays (discrete time-varying delays and distributed time-varying delays) is investigated in this paper. Employing the average dwell time approach (ADT), stochastic analysis technology, and linear matrix inequalities technique (LMI), some novel sufficient conditions on the issue of asymptotic behavior (the mean-square ultimate boundedness, the existence of an attractor, and the mean-square exponential stability) are established. A numerical example is provided to illustrate the effectiveness of the proposed results.

Sign in / Sign up

Export Citation Format

Share Document