Journal of X-Ray Science and Technology - Volume 31, issue 4

Authors: Zhang, Kun | Lin, Jun | Lin, Fan | Wang, Zhongyi | Zhang, Haicheng | Zhang, Shijie | Mao, Ning | Qiao, Guangdong

Article Type: Research Article

Abstract: BACKGROUND: Neoadjuvant chemotherapy (NAC) has been regarded as one of the standard treatments for patients with locally advanced breast cancer. No previous study has investigated the feasibility of using a contrast-enhanced spectral mammography (CESM)-based radiomics nomogram to predict pathological complete response (pCR) after NAC. OBJECTIVE: To develop and validate a CESM-based radiomics nomogram to predict pCR after NAC in breast cancer. METHODS: A total of 118 patients were enrolled, which are divided into a training dataset including 82 patients (with 21 pCR and 61 non-pCR) and a testing dataset of 36 patients (with 9 pCR and …27 non-pCR). The tumor regions of interest (ROIs) were manually segmented by two radiologists on the low-energy and recombined images and radiomics features were extracted. Intraclass correlation coefficients (ICCs) were used to assess the intra- and inter-observer agreements of ROI features extraction. In the training set, the variance threshold, SelectKBest method, and least absolute shrinkage and selection operator regression were used to select the optimal radiomics features. Radiomics signature was calculated through a linear combination of selected features. A radiomics nomogram containing radiomics signature score (Rad-score) and clinical risk factors was developed. The receiver operating characteristic (ROC) curve and calibration curve were used to evaluate prediction performance of the radiomics nomogram, and decision curve analysis (DCA) was used to evaluate the clinical usefulness of the radiomics nomogram. RESULTS: The intra- and inter- observer ICCs were 0.769–0.815 and 0.786–0.853, respectively. Thirteen radiomics features were selected to calculate Rad-score. The radiomics nomogram containing Rad-score and clinical risk factor showed an encouraging calibration and discrimination performance with area under the ROC curves of 0.906 (95% confidence interval (CI): 0.840–0.966) in the training dataset and 0.790 (95% CI: 0.554–0.952) in the test dataset. CONCLUSIONS: The CESM-based radiomics nomogram had good prediction performance for pCR after NAC in breast cancer; therefore, it has a good clinical application prospect. Show more

Keywords: Breast cancer, neoadjuvant chemotherapy, pathological complete response, radiomics, contrast-enhanced spectral mammography

DOI: 10.3233/XST-221349

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 669-683, 2023

Authors: Chen, Yi | Liu, Yue | Wu, Dan | Wen, Yanting | Li, Lun | Jiang, Huabei

Article Type: Research Article

Abstract: BACKGROUND: Electrical conductivity directly correlates with tissue functional information such as blood and water contents, and quantitative extraction of tissue conductivity is of significant importance for disease detection and diagnosis using microwave-induced thermoacoustic tomography (TAT). OBJECTIVE: The existing quantitative TAT (qTAT) approaches capable of extracting tissue conductivity require two steps for the recovery of conductivity. Such two steps approaches depend on an accurate knowledge of the microwave energy loss distribution in tissue and offer a slow computational convergence rate. The purpose of this study is to develop a new algorithm to reconstruct tissue conductivity with higher reconstruction accuracy …and greater computational efficiency. METHODS: We propose an improved qTAT method for direct recovery of tissue conductivity from thermoacoustic data measured along the boundary with only one step without the dependence of microwave energy loss information. The feasibility of our one-step qTAT method is validated in both simulated and tissue-mimicking phantom experiments with single-target and multi-target configurations with different contrast levels. RESULTS: Compared with the previous two-step methods, our one-step qTAT method improves the accuracy of conductivity recovery with approximately one-fold reduction in the mean absolute error (MAE) and root mean square error (RMSE) with p -values greater than 0.05. In addition, the convergence rate is improved by more than two folds for the one-step method. CONCLUSIONS: The study demonstrates that new method can quantitatively reconstruct conductivity of tissue more accurately and efficiently over the existing qTAT methods, leading to potentially enhanced accuracy for disease detection and diagnosis. Show more

Keywords: Conductivity distribution, thermoacoustic tomography, quantitative reconstruction method

DOI: 10.3233/XST-221353

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 685-698, 2023

Authors: Abraham, Bejoy | Mohan, Jesna | John, Shinu Mathew | Ramachandran, Sivakumar

Article Type: Research Article

Abstract: BACKGROUND: Tuberculosis (TB) is a highly infectious disease that mainly affects the human lungs. The gold standard for TB diagnosis is Xpert Mycobacterium tuberculosis/ resistance to rifampicin (MTB/RIF) testing. X-ray, a relatively inexpensive and widely used imaging modality, can be employed as an alternative for early diagnosis of the disease. Computer-aided techniques can be used to assist radiologists in interpreting X-ray images, which can improve the ease and accuracy of diagnosis. OBJECTIVE: To develop a computer-aided technique for the diagnosis of TB from X-ray images using deep learning techniques. METHODS: This research paper presents a novel …approach for TB diagnosis from X-ray using deep learning methods. The proposed method uses an ensemble of two pre-trained neural networks, namely EfficientnetB0 and Densenet201, for feature extraction. The features extracted using two CNNs are expected to generate more accurate and representative features than a single CNN. A custom-built artificial neural network (ANN) called PatternNet with two hidden layers is utilized to classify the extracted features. RESULTS: The effectiveness of the proposed method was assessed on two publicly accessible datasets, namely the Montgomery and Shenzhen datasets. The Montgomery dataset comprises 138 X-ray images, while the Shenzhen dataset has 662 X-ray images. The method was further evaluated after combining both datasets. The method performed exceptionally well on all three datasets, achieving high Area Under the Curve (AUC) scores of 0.9978, 0.9836, and 0.9914, respectively, using a 10-fold cross-validation technique. CONCLUSION: The experiments performed in this study prove the effectiveness of features extracted using EfficientnetB0 and Densenet201 in combination with PatternNet classifier in the diagnosis of tuberculosis from X-ray images. Show more

Keywords: Tuberculosis, Artificial Neural Network, CNN, EfficientnetB0, Densenet201

DOI: 10.3233/XST-230028

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 699-711, 2023

Authors: Ma, Yinjin | Zhang, Yajuan | Chen, Lin | Jiang, Qiang | Wei, Biao

Article Type: Research Article

Abstract: BACKGROUND: Chest CT scan is an effective way to detect and diagnose COVID-19 infection. However, features of COVID-19 infection in chest CT images are very complex and heterogeneous, which make segmentation of COVID-19 lesions from CT images quite challenging. OBJECTIVE: To overcome this challenge, this study proposes and tests an end-to-end deep learning method called dual attention fusion UNet (DAF-UNet). METHODS: The proposed DAF-UNet improves the typical UNet into an advanced architecture. The dense-connected convolution is adopted to replace the convolution operation. The mixture of average-pooling and max-pooling acts as the down-sampling in the encoder. Bridge-connected …layers, including convolution, batch normalization, and leaky rectified linear unit (leaky ReLU) activation, serve as the skip connections between the encoder and decoder to bridge the semantic gap differences. A multiscale pyramid pooling module acts as the bottleneck to fit the features of COVID-19 lesion with complexity. Furthermore, dual attention feature (DAF) fusion containing channel and position attentions followed the improved UNet to learn the long-dependency contextual features of COVID-19 and further enhance the capacity of the proposed DAF-UNet. The proposed model is first pre-trained on the pseudo label dataset (generated by Inf-Net) containing many samples, then fine-tuned on the standard annotation dataset (provided by the Italian Society of Medical and Interventional Radiology) with high-quality but limited samples to improve performance of COVID-19 lesion segmentation on chest CT images. RESULTS: The Dice coefficient and Sensitivity are 0.778 and 0.798 respectively. The proposed DAF-UNet has higher scores than the popular models (Att-UNet, Dense-UNet, Inf-Net, COPLE-Net) tested using the same dataset as our model. CONCLUSION: The study demonstrates that the proposed DAF-UNet achieves superior performance for precisely segmenting COVID-19 lesions from chest CT scans compared with the state-of-the-art approaches. Thus, the DAF-UNet has promising potential for assisting COVID-19 disease screening and detection. Show more

Keywords: Coronavirus disease 2019 (COVID-19), computed tomography (CT), deep learning, dual attention, medical image segmentation

DOI: 10.3233/XST-230001

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 713-729, 2023

Authors: Cao, Keyan | Tao, Hangbo | Wang, Zhiqiong | Jin, Xi

Article Type: Research Article

Abstract: BACKGROUND: Accurate classification of benign and malignant pulmonary nodules using chest computed tomography (CT) images is important for early diagnosis and treatment of lung cancer. In terms of natural image classification, the ViT-based model has greater advantages in extracting global features than the traditional CNN model. However, due to the small image dataset and low image resolution, it is difficult to directly apply the ViT-based model to pulmonary nodule classification. OBJECTIVE: To propose and test a new ViT-based MSM-ViT model aiming to achieve good performance in classifying pulmonary nodules. METHODS: In this study, CNN structure was …used in the task of classifying pulmonary nodules to compensate for the poor generalization of ViT structure and the difficulty in extracting multi-scale features. First, sub-pixel fusion was designed to improve the ability of the model to extract tiny features. Second, multi-scale local features were extracted by combining dilated convolution with ordinary convolution. Finally, MobileViT module was used to extract global features and predict them at the spatial level. RESULTS: CT images involving 442 benign nodules and 406 malignant nodules were extracted from LIDC-IDRI data set to verify model performance, which yielded the best accuracy of 94.04% and AUC value of 0.9636 after 10 cross-validations. CONCLUSION: The proposed new model can effectively extract multi-scale local and global features. The new model performance is also comparable to the most advanced models that use 3D volume data training, but its occupation of video memory (training resources) is less than 1/10 of the conventional 3D models. Show more

Keywords: Pulmonary nodules classification, MobileViT, multi-scale nodules, computer-aided diagnosis.

DOI: 10.3233/XST-230014

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 731-744, 2023

Authors: Yuan, Zilong | Liu, Tao | Zhang, Biao | Wu, Jiaxing | He, Yaoyao | Chen, Tiao | Zhang, Zhaoxi | Li, Cuiling | Liu, Yulin

Article Type: Research Article

Abstract: OBJECTIVE: The aim of this study is to investigate the radiation dose and image quality of head CT using SPS and OBTCM techniques. METHODS: Three anthropomorphic head phantoms (1-yr-old, 5-yr-old, and adult) were used. Images were acquired using four modes (Default protocol, OBTCM, SPS, and SPS+OBTCM). Absorbed dose to the lens, anterior brain (brain_A), and posterior brain (brain_P) was measured and compared. Image noise and CNR were assessed in the selected regions of interest (ROIs). RESULTS: Compared with that in the Default protocol, the absorbed dose to the lens reduced by up to 28.33%,71.38%, and 71.12% …in OBTCM, SPS, and SPS+OBTCM, respectively. The noise level in OBTCM slightly (≤1.45HU) increased than that in Default protocol, and the SPS or SPS+OBTCM mode resulted in a quantitatively small increase (≤2.58HU) in three phantoms. There was no significant difference in CNR of different phantoms under varies scanning modes (p  > 0.05). CONCLUSIONS: During head CT examinations, the SPS mode can reduce the radiation dose while maintaining image quality. SPS+OBTCM couldn’t further effectively reduce the absorbed dose to the lens for 1-yr and 5-yr-old phantoms. Thus, SPS mode in pediatric and SPS+OBTCM mode in adult are better than other modes, and should be used in clinical practice. Show more

Keywords: Selected photon shield, organ-based tube current modulation, computed tomography (ct), lens

DOI: 10.3233/XST-230018

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 745-756, 2023

Authors: Li, Zhiyuan | Liu, Yi | Chen, Yang | Shu, Huazhong | Lu, Jing | Gui, Zhiguo

Article Type: Research Article

Abstract: BACKGROUND: In view of the underlying health risks posed by X-ray radiation, the main goal of the present research is to achieve high-quality CT images at the same time as reducing x-ray radiation. In recent years, convolutional neural network (CNN) has shown excellent performance in removing low-dose CT noise. However, previous work mainly focused on deepening and feature extraction work on CNN without considering fusion of features from frequency domain and image domain. OBJECTIVE: To address this issue, we propose to develop and test a new LDCT image denoising method based on a dual-domain fusion deep convolutional neural …network (DFCNN). METHODS: This method deals with two domains, namely, the DCT domain and the image domain. In the DCT domain, we design a new residual CBAM network to enhance the internal and external relations of different channels while reducing noise to promote richer image structure information. For the image domain, we propose a top-down multi-scale codec network as a denoising network to obtain more acceptable edges and textures while obtaining multi-scale information. Then, the feature images of the two domains are fused by a combination network. RESULTS: The proposed method was validated on the Mayo dataset and the Piglet dataset. The denoising algorithm is optimal in both subjective and objective evaluation indexes as compared to other state-of-the-art methods reported in previous studies. CONCLUSIONS: The study results demonstrate that by applying the new fusion model denoising, denoising results in both image domain and DCT domain are better than other models developed using features extracted in the single image domain. Show more

Keywords: Low-dose CT, image denoising, dual-domain, DCT

DOI: 10.3233/XST-230020

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 757-775, 2023

Authors: Poonkuzhali, P. | Helen Prabha, K.

Article Type: Research Article

Abstract: BACKGROUND: Hyperspectral brain tissue imaging has been recently utilized in medical research aiming to study brain science and obtain various biological phenomena of the different tissue types. However, processing high-dimensional data of hyperspectral images (HSI) is challenging due to the minimum availability of training samples. OBJECTIVE: To overcome this challenge, this study proposes applying a 3D-CNN (convolution neural network) model to process spatial and temporal features and thus improve performance of tumor image classification. METHODS: A 3D-CNN model is implemented as a testing method for dealing with high-dimensional problems. The HSI pre-processing is accomplished using distinct …approaches such as hyperspectral cube creation, calibration, spectral correction, and normalization. Both spectral and spatial features are extracted from HSI. The Benchmark Vivo human brain HSI dataset is used to validate the performance of the proposed classification model. RESULTS: The proposed 3D-CNN model achieves a higher accuracy of 97% for brain tissue classification, whereas the existing linear conventional support vector machine (SVM) and 2D-CNN model yield 95% and 96% classification accuracy, respectively. Moreover, the maximum F1-score obtained by the proposed 3D-CNN model is 97.3%, which is 2.5% and 11.0% higher than the F1-scores obtained by 2D-CNN model and SVM model, respectively. CONCLUSION: A 3D-CNN model is developed for brain tissue classification by using HIS dataset. The study results demonstrate the advantages of using the new 3D-CNN model, which can achieve higher brain tissue classification accuracy than conventional 2D-CNN model and SVM model. Show more

Keywords: Hyperspectral image classification, high dimensional features, brain tumor detection, machine learning, convolutional neural network

DOI: 10.3233/XST-230045

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 777-796, 2023

Authors: Cai, Xin | Hou, Xuewen | Sun, Rong | Chang, Xiao | Zhu, Honglin | Jia, Shouqiang | Nie, Shengdong

Article Type: Research Article

Abstract: BACKGROUND: As one of the significant preoperative imaging modalities in medical diagnosis, Magnetic resonance imaging (MRI) takes a long scanning time due to its special imaging principle. OBJECTIVE: We propose an innovative MRI reconstruction strategy and data consistency method based on deep learning to reconstruct high-quality brain MRIs from down-sampled data and accelerate the MR imaging process. METHODS: Sixteen healthy subjects undergoing T1-weighted spin-echo (SE) and T2-weighted fast spin-echo (FSE) sequences by a 1.5T MRI scanner were recruited. A Y-Net3+ network was used to facilitate the high-quality MRI reconstruction through context information. In addition, the existing …data consistency fidelity method was improved. The difference between the reconstructed K-space and the original K-space was shorten by the linear regression algorithm. Therefore, the redundant artifacts derived from under-sampling were avoided. The Structural Similarity (SSIM) and Peak Signal to Noise Ratio (PSNR) were applied to quantitatively evaluate image reconstruction performance of different down-sampling patterns. RESULTS: Compared with the classical Y-Net, Y-Net3+ network improved SSIM and PSNR of MRI images from 0.9164±0.0178 and 33.2216±3.2919 to 0.9387±0.0363 and 35.1785±3.3105, respectively, under compressed sensing reconstruction with acceleration factor of 4. The improved network increases signal-to-noise ratio and adds more image texture information in the reconstructed images. Furthermore, in the process of data consistency, linear regression analysis was used to reduce the difference between the reconstructed K-space and the original K-space, so that the SSIM and PSNR were increased to 0.9808±0.0081 and 40.9254±1.1911, respectively. CONCLUSIONS: The improved Y-Net combined with data consistency fidelity method elucidates its potential in reconstructing high-quality T2-weighted images from the down-sampled data by fully exploring the T1-weighted information. With the advantage of avoiding down-sampled artifacts, the improved network exhibits remarkable clinical promise for fast MRI applications. Show more

Keywords: Magnetic resonance imaging, Deep learning, Multi-contrast MRI, Image reconstruction, Data consistency

DOI: 10.3233/XST-230012

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 797-810, 2023

Authors: Zhang, Xiaomei | Wang, Zhe | Yun, Xiangyu | Li, Mohan | Hu, Jinming | Wang, Chengmin | Wei, Cunfeng

Article Type: Research Article

Abstract: BACKGROUND: Photon counting spectral CT is a significant direction in the development of CT technology and material identification is an important application of spectral CT. However, spectrum estimation in photon counting spectral CT is highly complex and may affect quantification accuracy of material identification. OBJECTIVE: To address the problem of energy spectrum estimation in photon-counting spectral CT, this study investigates empirical material decomposition algorithms to achieve accurate quantitative decomposition of the effective atomic number. METHODS: The spectrum is first calibrated using the empirical dual-energy calibration (EDEC) method and the effective atomic number is then quantitatively estimated …based on the EDEC method. The accuracy of estimating the effective atomic number of materials under different calibration conditions is investigated by designing different calibration phantoms, and accurate quantitation is achieved using suitable calibration settings. Last, the validity of this method is verified through simulations and experimental studies. RESULTS: The results demonstrate that the error in estimating the effective atomic number is reduced to within 4% for low and medium Z materials, thereby enabling accurate material identification. CONCLUSION: The empirical dual-energy correction method can solve the problem of energy spectrum estimation in photon counting spectral CT. Accurate effective atomic number estimation can be achieved with suitable calibration. Show more

Keywords: Photon counting detectors (PCDs), material decomposition, spectral CT, effective atomic number, polynomial fitting, calibration phantoms

DOI: 10.3233/XST-230054

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 811-824, 2023

Authors: Zhang, Tiancheng | Almajidi, Yasir Q. | Awad, Sameer A. | Alhachami, Firas Rahi | Gatea, Maher Abdulfadhil | Kadhum, Wesam R.

Article Type: Research Article

Abstract: BACKGROUND: Several physical factors such as photon beam energy, electron beam energy, and dose rate may affect the dosimetric properties of polymer gel dosimeters. The photon beam energy and dose rate dependence of PASSAG gel dosimeter were previously evaluated. OBJECTIVE: This study aims to assess the dosimetric properties of the optimized PASSAG gel samples in various electron beam energies. METHODS: The optimized PASSAG gel samples are first fabricated and irradiated to various electron energies (5, 7, 10 and 12 MeV). Then, the response (R 2 ) and sensitivity of gel samples are analyzed by magnetic resonance imaging …technique at a dose range of 0 to 10 Gy, scanning room temperature range of 15 to 22 °C, and post-irradiation time range of 1 to 30 days. RESULTS: The R 2 -dose response and sensitivity of gel samples do not change under the evaluated electron beam energies (the differences are less than 5%). Furthermore, a dose resolution range of 11 to 38 cGy is obtained for the gel samples irradiated to different electron beam energies. Moreover, the findings show that the R 2 -dose response and sensitivity dependence of gel samples on electron beam energy varies over different scanning room temperatures and post-irradiation times. CONCLUSION: The dosimetric assessment of the optimized PASSAG gel samples provides the promising data for this dosimeter during electron beam radiotherapy. Show more

Keywords: Electron beam radiotherapy, PASSAG, MRI, Polymer gel dosimetry

DOI: 10.3233/XST-230073

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 825-836, 2023

Authors: Kweon, Dae Cheol

Article Type: Research Article

Abstract: BACKGROUND: Due to the wide application of the cone beam computed tomography (CBCT) in clinical practice, it is important to assess radiation dose of CBCT more accurately and efficiently in different clinical applications. OBJECTIVE: This study aims to calculate effective and absorbed doses in CBCT measured in an anthropomorphic phantom using computer-based Monte Carlo (PCXMC) software, and to conduct comparative evaluations of MOSFET (metal– oxide– semiconductor field-effect transistor) and radiophotoluminescence glass dosimeters (RPLGD). METHODS: Effective and absorbed organ doses are compared with those obtained using MOSFET and RPLGD dosimetry in an anthropomorphic phantom given the same …exposure settings. Effective and absorbed organ doses from CBCT during scout and main projections are calculated using PCXMC and PCXMCRotation software, respectively. RESULTS: The mean effective dose from CBCT calculated using PCXMC software is 233.8μSv, while the doses calculated using dosimetry (MOSFET and RPLGD) are 266.67μSv and 268.78μSv, respectively. The X-ray source variation is 0.79%. The prescription limits based on the Friedman test for MOSFET and RPLGD pre-points (i.e., in an analytical analysis of diagnostic names in CBCT) are not statistically significant. The calculated correlation coefficient between MOSFET- and RPLGD-derived absorbed dose values with respect to a field of view CBCT parameter of 17×13.5 mm is r  = 0.8623. CONCLUSIONS: The study demonstrates that the PCXMC software may be used as an alternative to MOSFET and RPLGD dosimetry for effective and absorbed organ dose estimation in CBCT conducted with a large FOV in an anthropomorphic phantom. Show more

Keywords: Absorbed dose, cone beam CT, radiation dose, MOSFET, RPLGD

DOI: 10.3233/XST-230026

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 837-852, 2023

Authors: Ma, Junmin

Article Type: Research Article

Abstract: OBJECTIVE: To explore the application value of digital tungsten-molybdenum double target three-dimensional positioning indwelling guide wire and guided surgical resection biopsy in the diagnosis of breast microcalcification. METHODS: A retrospective analysis of 168 patients with negative clinical palpation and molybdenum target X-ray examination found breast abnormalities were equally divided into two groups according to different surgical positioning methods. The control and observation group underwent gross positioning biopsy and digital tungsten-molybdenum dual-target three-dimensional positioning indwelling guide wire to guide surgical resection biopsy, respectively. The results of molybdenum target X-ray examination and the success rate of one-time complete resection of …the lesions were compared between the two groups, and the corresponding relationship between the pathological diagnosis results of the lesions after surgical resection and the performance of mammography in the observation group was compared. RESULTS: There was no significant difference in age and molybdenum target X-ray examination results between the two groups (P  > 0.05). General information is comparable; the success rate of one-time complete resection of lesions in the observation group was significantly higher than that in the control group (95.2% vs 78.6%, P  = 0.024). There were 34 cases of malignant lesions in the observation group, accounting for 40.5% (34/84), including 11 cases of invasive ductal carcinoma (64.7%), 50 cases of benign lesions, accounting for 59.5% (50/84), including 16 cases of breast lobular hyperplasia (32%), 18 cases of breast cystic hyperplasia (36%). CONCLUSION: In diagnosis of breast microcalcifications, surgical resection biopsy guided using digital tungsten-molybdenum double target three-dimensional positioning indwelling guide wire achieves high success rate and has advantages of high safety and accurate diagnosis. Thus, it has potential to play a greater role in early diagnosis of breast cancer and is worthy of clinical application. Show more

Keywords: Tungsten and molybdenum, microcalcification of breast, Three-dimensional stereoscopic, guide wire, target X-ray

DOI: 10.3233/XST-230039

Citation: Journal of X-Ray Science and Technology, vol. 31, no. 4, pp. 853-864, 2023