Journal of X-Ray Science and Technology - Volume 32, issue 1

Authors: Song, Qiyi | Li, Xiang | Zhang, Mingbao | Zhang, Xiangyi | Thanh, Dang N.H.

Article Type: Research Article

Abstract: BACKGROUND: In clinical medicine, low-dose radiographic image noise reduces the quality of the detected image features and may have a negative impact on disease diagnosis. OBJECTIVE: In this study, Adaptive Projection Network (APNet) is proposed to reduce noise from low-dose medical images. METHODS: APNet is developed based on an architecture of the U-shaped network to capture multi-scale data and achieve end-to-end image denoising. To adaptively calibrate important features during information transmission, a residual block of the dual attention method throughout the encoding and decoding phases is integrated. A non-local attention module to separate the noise and …texture of the image details by using image adaptive projection during the feature fusion. RESULTS: To verify the effectiveness of APNet, experiments on lung CT images with synthetic noise are performed, and the results demonstrate that the proposed approach outperforms recent methods in both quantitative index and visual quality. In addition, the denoising experiment on the dental CT image is also carried out and it verifies that the network has a certain generalization. CONCLUSIONS: The proposed APNet is an effective method that can reduce image noise and preserve the required image details in low-dose radiographic images. Show more

Keywords: Medical image denoising, residual learning, subspace projection, X-ray, CT, low dose CT

DOI: 10.3233/XST-230181

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 1-15, 2024

Authors: Fang, Ting | Liu, Naijia | Nie, Shengdong | Jia, Shouqiang | Ye, Xiaodan

Article Type: Research Article

Abstract: BACKGROUND: Alberta stroke program early CT score (ASPECTS) is a semi-quantitative evaluation method used to evaluate early ischemic changes in patients with acute ischemic stroke, which can guide physicians in treatment decisions and prognostic judgments. OBJECTIVE: We propose a method combining deep learning and radiomics to alleviate the problem of large inter-observer variance in ASPECTS faced by physicians and assist them to improve the accuracy and comprehensiveness of the ASPECTS. METHODS: Our study used a brain region segmentation method based on an improved encoding-decoding network. Through the deep convolutional neural network, 10 regions …defined for ASPECTS will be obtained. Then, we used Pyradiomics to extract features associated with cerebral infarction and select those significantly associated with stroke to train machine learning classifiers to determine the presence of cerebral infarction in each scored brain region. RESULTS: The experimental results show that the Dice coefficient for brain region segmentation reaches 0.79. Three radioactive features are selected to identify cerebral infarction in brain regions, and the 5-fold cross-validation experiment proves that these 3 features are reliable. The classifier trained based on 3 features reaches prediction performance of AUC = 0.95. Moreover, the intraclass correlation coefficient of ASPECTS between those obtained by the automated ASPECTS method and physicians is 0.86 (95% confidence interval, 0.56-0.96). CONCLUSIONS: This study demonstrates advantages of using a deep learning network to replace the traditional template registration for brain region segmentation, which can determine the shape and location of each brain region more precisely. In addition, a new brain region classifier based on radiomics features has potential to assist physicians in clinical stroke detection and improve the consistency of ASPECTS. Show more

Keywords: Alberta stroke program early CT score, Acute ischemic stroke, Deep learning, Computer-aided method, CT angiography

DOI: 10.3233/XST-230119

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 17-30, 2024

Authors: Chempak Kumar, A. | Mubarak, D. Muhammad Noorul

Article Type: Research Article

Abstract: BACKGROUND: Esophageal cancer (EC) is aggressive cancer with a high fatality rate and a rapid rise of the incidence globally. However, early diagnosis of EC remains a challenging task for clinicians. OBJECTIVE: To help address and overcome this challenge, this study aims to develop and test a new computer-aided diagnosis (CAD) network that combines several machine learning models and optimization methods to detect EC and classify cancer stages. METHODS: The study develops a new deep learning network for the classification of the various stages of EC and the premalignant stage, Barrett’s Esophagus from endoscopic images. The …proposed model uses a multi-convolution neural network (CNN) model combined with Xception, Mobilenetv2, GoogLeNet, and Darknet53 for feature extraction. The extracted features are blended and are then applied on to wrapper based Artificial Bee Colony (ABC) optimization technique to grade the most accurate and relevant attributes. A multi-class support vector machine (SVM) classifies the selected feature set into the various stages. A study dataset involving 523 Barrett’s Esophagus images, 217 ESCC images and 288 EAC images is used to train the proposed network and test its classification performance. RESULTS: The proposed network combining Xception, mobilenetv2, GoogLeNet, and Darknet53 outperforms all the existing methods with an overall classification accuracy of 97.76% using a 3-fold cross-validation method. CONCLUSION: This study demonstrates that a new deep learning network that combines a multi-CNN model with ABC and a multi-SVM is more efficient than those with individual pre-trained networks for the EC analysis and stage classification. Show more

Keywords: Artificial bee colony optimization, barrett’s esophagus, esophageal cancer, multi-CNN models, wrapper feature selection methods

DOI: 10.3233/XST-230111

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 31-51, 2024

Authors: Gayatri, Erapaneni | Aarthy, S.L.

Article Type: Research Article

Abstract: BACKGROUND: With the rapid growth of Deep Neural Networks (DNN) and Computer-Aided Diagnosis (CAD), more significant works have been analysed for cancer related diseases. Skin cancer is the most hazardous type of cancer that cannot be diagnosed in the early stages. OBJECTIVE: The diagnosis of skin cancer is becoming a challenge to dermatologists as an abnormal lesion looks like an ordinary nevus at the initial stages. Therefore, early identification of lesions (origin of skin cancer) is essential and helpful for treating skin cancer patients effectively. The enormous development of automated skin cancer diagnosis systems significantly supports dermatologists. …METHODS: This paper performs a classification of skin cancer by utilising various deep-learning frameworks after resolving the class Imbalance problem in the ISIC-2019 dataset. A fine-tuned ResNet-50 model is used to evaluate the performance of original data, augmented data, and after by adding the focal loss. Focal loss is the best technique to solve overfitting problems by assigning weights to hard misclassified images. RESULTS: Finally, augmented data with focal loss is given a good classification performance with 98.85% accuracy, 95.52% precision, and 95.93% recall. Matthews Correlation coefficient (MCC) is the best metric to evaluate the quality of multi-class images. It has given outstanding performance by using augmented data and focal loss. Show more

Keywords: Medical images, skin cancer, ResNet-50, data augmentation, highly imbalance dataset

DOI: 10.3233/XST-230204

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 53-68, 2024

Authors: Ma, Chenchen | Su, Ting | Zhu, Jiongtao | Zhang, Xin | Zheng, Hairong | Liang, Dong | Wang, Na | Zhang, Yunxin | Ge, Yongshuai

Article Type: Research Article

Abstract: BACKGROUND: Slow kVp switching technique is an important approach to realize dual-energy CT (DECT) imaging, but its performance has not been thoroughly investigated yet. OBJECTIVE: This study aims at comparing and evaluating the DECT imaging performance of different slow kVp switching protocols, and thus helps determining the optimal system settings. METHODS: To investigate the impact of energy separation, two different beam filtration schemes are compared: the stationary beam filtration and dynamic beam filtration. Moreover, uniform tube voltage modulation and weighted tube voltage modulation are compared along with various modulation frequencies. A model-based direct decomposition algorithm is …employed to generate the water and iodine material bases. Both numerical and physical experiments are conducted to verify the slow kVp switching DECT imaging performance. RESULTS: Numerical and experimental results demonstrate that the material decomposition is less sensitive to beam filtration, voltage modulation type and modulation frequency. As a result, robust material-specific quantitative decomposition can be achieved in slow kVp switching DECT imaging. CONCLUSIONS: Quantitative DECT imaging can be implemented with slow kVp switching under a variety of system settings. Show more

Keywords: Dual-energy CT, slow kVp switching, material decomposition

DOI: 10.3233/XST-230201

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 69-85, 2024

Authors: Cortez, Jarrod | Romero, Ignacio | Ngo, Jason | Azam, Md Sayed Tanveer | Niu, Chuang | Almeida-da-Silva, Cássio Luiz Coutinho | Cabido, Leticia Ferreira | Ojcius, David M. | Chin, Wei-Chun | Wang, Ge | Li, Changqing

Article Type: Research Article

Abstract: BACKGROUND: Periodontal disease affects over 50% of the global population and is characterized by gingivitis as the initial sign. One dental health issue that may contribute to the development of periodontal disease is foreign body gingivitis (FBG), which can result from exposure to some kinds of foreign metal particles from dental products or food. OBJECTIVE: We design a novel, portable, affordable, multispectral X-ray and fluorescence optical microscopic imaging system dedicated to detecting and differentiating metal oxide particles in dental pathological tissues. A novel denoising algorithm is applied. We verify the feasibility and optimize the performance of the imaging …system with numerical simulations. METHODS: The designed imaging system has a focused X-ray tube with tunable energy spectra and thin scintillator coupled with an optical microscope as detector. A simulated soft tissue phantom is embedded with 2-micron thick metal oxide discs as the imaged object. GATE software is used to optimize the systematic parameters such as energy bandwidth and X-ray photon number. We have also applied a novel denoising method, Noise2Sim with a two-layer UNet structure, to improve the simulated image quality. RESULTS: The use of an X-ray source operating with an energy bandwidth of 5 keV, X-ray photon number of 108 , and an X-ray detector with a 0.5 micrometer pixel size in a 100 by 100-pixel array allowed for the detection of particles as small as 0.5 micrometer. With the Noise2Sim algorithm, the CNR has improved substantially. A typical example is that the Aluminum (Al) target’s CNR is improved from 6.78 to 9.72 for the case of 108 X-ray photons with the Chromium (Cr) source of 5 keV bandwidth. CONCLUSIONS: Different metal oxide particles were differentiated using Contrast-to-Noise ratio (CNR) by utilizing four different X-ray spectra. Show more

Keywords: X-ray imaging, Monte Carlo, GATE, Denoise

DOI: 10.3233/XST-230175

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 87-103, 2024

Article Type: Other

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 105-105, 2024

Authors: Jia, Wanping | Zhao, Guangyong

Article Type: Research Article

Abstract: BACKGROUND: In this research, imaging techniques such as CT and X-ray are used to locate important muscles in the shoulders and legs. Athletes who participate in sports that require running, jumping, or throwing are more likely to get injuries such as sprains, strains, tendinitis, fractures, and dislocations. One proposed automated technique has the overarching goal of enhancing recognition. OBJECTIVE: This study aims to determine how to recognize the major muscles in the shoulder and leg utilizing X-ray CT images as its primary diagnostic tool. METHODS: Using a shape model, discovering landmarks, and generating a form model …are the steps necessary to identify injuries in key shoulder and leg muscles. The method also involves identifying injuries in significant abdominal muscles. The use of adversarial deep learning, and more specifically Deep-Injury Region Identification, can improve the ability to identify damaged muscle in X-ray and CT images. RESULTS: Applying the proposed diagnostic model to 150 sets of CT images, the study results show that Jaccard similarity coefficient (JSC) rate for the procedure is 0.724, the repeatability is 0.678, and the accuracy is 94.9% respectively. CONCLUSION: The study results demonstrate feasibility of using adversarial deep learning and deep-injury region identification to automatically detect severe muscle injuries in the shoulder and leg, which can enhance the identification and diagnosis of injuries in athletes, especially for those who compete in sports that include running, jumping, and throwing. Show more

Keywords: Muscle injury, deep learning, generator-discriminator network, cross sectional area

DOI: 10.3233/XST-230135

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 107-121, 2024

Authors: Alshamrani, Khalaf | Alshamrani, Hassan A.

Article Type: Research Article

Abstract: BACKGROUND: By providing both functional and anatomical information from a single scan, digital imaging technologies like PET/CT and PET/MRI hybrids are gaining popularity in medical imaging industry. In clinical practice, the median value (SUVmed) receives less attention owing to disagreements surrounding what defines a lesion, but the SUVmax value, which is a semi-quantitative statistic used to analyse PET and PET/CT images, is commonly used to evaluate lesions. OBJECTIVE: This study aims to build an image processing technique with the purpose of automatically detecting and isolating lesions in PET/CT images, as well as measuring and assessing the SUVmed. …METHODS: The pictures are separated into their respective lesions using mathematical morphology and the crescent region, which are both part of the image processing method. In this research, a total of 18 different pictures of lesions were evaluated. RESULTS: The findings of the study reveal that the threshold is satisfied by both the SUVmax and the SUVmed for most of the lesion types. However, in six instances, the SUVmax and SUVmed values are found to be in different courts. CONCLUSION: The new information revealed by this study needs to be further investigated to determine if it has any practical value in diagnosing and monitoring lesions. However, results of this study suggest that SUVmed should receive more attention in the evaluation of lesions in PET and CT images. Show more

Keywords: PET/CT images, tumor images, standardised uptake value, lesions, diagnosing, monitoring

DOI: 10.3233/XST-230095

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 123-139, 2024

Authors: Xu, Pengcheng | Liu, Yongsheng | Wu, Shen | Cheng, Dong | Sun, Zhanfeng

Article Type: Research Article

Abstract: BACKGROUND: How to improve efficacy and reduce side effects in treating recurrent esophageal cancer by applying the second course of radiotherapy alone and its combination with chemotherapy has been attracting broad research interest. OBJECTIVE: This review paper aims to systematically evaluate efficacy and side effects of applying the second course of anterograde radiotherapy alone and its combination with chemotherapy in treating recurrent esophageal cancer. METHODS: First, the relevant research papers are retrieved from PubMed, CNKI and Wanfang databases. Next, Redman 5.3 software is used to calculate the relative risk and 95% confidence interval to evaluate the …efficacy and adverse reactions of applying the single-stage radiotherapy with and without combining single/multi dose chemotherapy to treat recurrent esophageal cancer. Then, a meta data analysis is applied to examine the effectiveness and side effects of radiation alone and re-course radiotherapy plus chemotherapy in treating esophageal cancer recurrence after the first radiotherapy. RESULTS: Fifteen papers are retrieved, which included 956 patients. Among them, 476 patients received radiotherapy combined with single drug/multi drug chemotherapy (observation) and others received only radiotherapy (control). Data analysis results show that the incidence of radiation induced lung injury and bone marrow suppression is high in the observation group. Subgroup analysis also shows the higher effective rate or one-year overall survival rate of patients treated with the second course radiotherapy combined with single drug chemotherapy. CONCLUSION: The meta-analysis result demonstrates that combining the second course of radiotherapy with single-drug chemotherapy has advantages in treating recurrent esophageal cancer with the manageable side effects. However, due to insufficient data, it is not possible to conduct the further subgroup analysis comparing the side effects of restorative radiation with the combined chemotherapy using between a single drug and multiple drugs. Show more

Keywords: Esophageal cancer, recurrence, second course radiotherapy, meta analysis, chemotherapy

DOI: 10.3233/XST-230098

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 141-155, 2024

Authors: Zhang, Liyan | Xu, Ruiyan | Zhao, Jingde

Article Type: Research Article

Abstract: BACKGROUND: Early diagnosis of breast cancer is crucial to perform effective therapy. Many medical imaging modalities including MRI, CT, and ultrasound are used to diagnose cancer. OBJECTIVE: This study aims to investigate feasibility of applying transfer learning techniques to train convoluted neural networks (CNNs) to automatically diagnose breast cancer via ultrasound images. METHODS: Transfer learning techniques helped CNNs recognise breast cancer in ultrasound images. Each model’s training and validation accuracies were assessed using the ultrasound image dataset. Ultrasound images educated and tested the models. RESULTS: MobileNet had the greatest accuracy during training and DenseNet121 …during validation. Transfer learning algorithms can detect breast cancer in ultrasound images. CONCLUSIONS: Based on the results, transfer learning models may be useful for automated breast cancer diagnosis in ultrasound images. However, only a trained medical professional should diagnose cancer, and computational approaches should only be used to help make quick decisions. Show more

Keywords: Breast cancer, ultrasound imaging, deep learning, Convoluted Neural Networks, image classification

DOI: 10.3233/XST-230085

Citation: Journal of X-Ray Science and Technology, vol. 32, no. 1, pp. 157-171, 2024