ISSN 1004-4140
CN 11-3017/P

2022 Vol. 31, No. 5

2022, 31(5)
CT Theory
A Feasibility Study of Knee Joint Semantic Segmentation on 3D MR Images
SHEN Le, LU Qian, TANG Hu, WU Sha, YI Yi, SUN Yunda, QIU Qian, ZHANG Li, ZHENG Zhuozhao, CAI Xu
2022, 31(5): 531-542. doi: 10.15953/j.ctta.2022.091
The segmentation of knee joint is of great significance for diagnosis, guidance and treatment of knee osteoarthritis. However, manual delineation is time-consuming and labor-intensive since various anatomical structures are involved in the 3D MRI volume. Automatic segmentation of the whole knee joint requires no human effort, additionally can improve the quality of arthritis diagnosis and treatment by describing the details more accurately. Existing knee joint segmentation methods in the literature focus on only one or few structures out of many. In this paper, we study the feasibility of knee joint segmentation on MR images based on neural networks and deal with the following challenges: (1) end-to-end segmentation of 15 anatomical structures, including bone and soft tissue, of the whole knee on MR images; (2) robust segmentation of small structures such as the anterior cruciate ligament, accounting for about 0.036% of the volume data. Experiments on the knee joint MR images demonstrate that the average segmentation accuracy of our method achieves 92.92%. The Dice similarity coefficients of 9 structures were above 94%, five structures were between 87% and 90%, and the remaining one was about 76%.
Experiment and Forward Modeling Analysis of Microgravity Detection of Urban Underground Space
LU Han, SUN Yan, ZHOU Mingxia, ZHANG Chong, YIN Jiuyang, MA Dong, CUI Mingfei, JIANG Hong
2022, 31(5): 543-556. doi: 10.15953/j.ctta.2021.069
With the rapid development of national cities, the demand for urban underground space exploration development and utilization has increased. Due to the interference factors in human activities areas, traditional geophysical methods can't obtain true and accurate detection data. The microgravity method is relatively less affected by interference factors. The interference from urban buildings and human activities can be eliminated by the method of model forward correction, so as to obtain high-precision gravity collection data, and then the spatial location information of tunnel, goaf, cavity, collapse area and pipe gallery in urban underground space can be obtained through effective inversion method. In this paper, through experimental detection and analysis of the theoretically affected factors in the urban detection carried out by the ground mobile high-precision gravity measuring instrument, combined with the forward model correction research, the microgravity method shows good effect in the detection of urban underground space.
Denoising of Seismic Data Based on Block Dictionary Learning Theory
ZHOU Junjie, WU Xiangling, LI Wenjie, LI Jinghe
2022, 31(5): 557-566. doi: 10.15953/j.1004-4140.2022.31.05.03
With the increasingly complex observation environment of oil and gas exploration, the seismic data collected are often mixed with various noise signals, resulting in the effective weak signal caused by the exploration target is covered, which seriously affects the high-precision seismic data interpretation, so it is more and more important to effectively suppress the seismic data noise. In this paper, the dictionary learning strategy is used to block the complex seismic data. The dictionary atoms are obtained through the dictionary learning of the block data, and the sparse representation of the seismic data is constructed by high-precision dictionary learning. The dictionary atoms are updated through two iterations for data denoising. The dictionary learning algorithm is applied to the processing of simulated and measured seismic data with random noise. The analysis results show that the algorithm can effectively removes the random noise while retains the effective signal phase axis, improves the signal-to-noise ratio which verifies the feasibility and effectiveness of the algorithm. The research results provide a new technical means for complex noisy seismic data denoising.
Research and Application of Seismic Frequency Extension technology Based on Ghost Wave Attenuation and Non-stationary Multi-order Differential Algorithm
LI Jian, YIN Wensun, LI Qin, LIU Qingwen, WANG Xiaopei
2022, 31(5): 567-576. doi: 10.15953/j.ctta.2021.013
High-resolution seismic data can realize better well seismic calibration results,clearer structural interpretation and reservoir characterization, and also hold better identification ability of thin layers. In order to improve seismic resolution, it is necessary to carry out frequency extension processing on seismic data. The conventional seismic spectral broadening method used to be carried out in the frequency domain is susceptible to high-frequency noise and thus reduces the reliability of the data. In this paper we propose a time-domain frequency extension technology which is based on the combination of ghost wave processing and non-stationary multi-order differential resolution. The sesmic spectral can be broaded merely through multiple times of integration and differential operation, whose results are performed amplitude matching at Gaussian window to ensure the consistency of the amplitude before and after the processing. Weighted fusion is performed on the difference results, the high-order difference is assigned a smaller weight to avoid the influence from the high-frequency noise, and thus improve the anti-noise performance of the algorithm. The theoretical models and processing results of the field seismic data show that the algorithm can effectively improve the resolution of seismic data.
Deep Learning Reservoir Parameter Prediction Based on Seismic Attribute Reduction: Take Ledong Area of Yinggehai Basin as an Example
LIU Shiyou, QU Fuliang, ZHOU Fan, DENG Lifeng
2022, 31(5): 577-586. doi: 10.15953/j.ctta.2021.048
As an important indicator to describe reservoir characteristics, reservoir modeling and fluid model, the accurate estimation of reservoir physical parameters can provide a powerful reference for reservoir prediction, but the traditional inversion method of reservoir physical parameters can not give consideration to inversion accuracy and spatial continuity. To solve the above problems, this paper introduced seismic attributes as input of deep learning algorithm. Aiming at the information redundancy among seismic attributes, random forest-recursive elimination method was used to reduce the seismic attributes, thus a prediction method of reservoir physical property parameters based on seismic attribute reduction was finally established. The actual data test results showed that the prediction results of reservoir physical parameters by deep learning based on seismic attribute reduction presented good accuracy and lateral resolution, which confirmed the effectiveness of the proposed method.
Application of Comprehensive Geophysical Prospecting Method in Karst Exploration of Urban Subway
YU Tao, WANG Xiaolong, WANG Junchao
2022, 31(5): 587-596. doi: 10.15953/j.ctta.2021.073
The adverse geologic conditions such as Karst often threaten the safety of subway engineering. It is necessary to ascertain the location and scale of the Karst during the exploration stage. However, the randomness of Karst distribution and the complexity of urban environment make Karst exploration quite difficult. In this paper, we use the Opposing Coils TEM and Elastic Wave CT to prospect the Karst development of a subway project in Nanjing and analyze the application effect of comprehensive geophysical prospecting method which is based on opposing coils TEM and elastic wave CT method in Karst exploration.
CT Intelligent Solution for Real-time Inspection and Release of Baggage and Cargo
LI Xinbin, ZHANG Li, CHEN Zhiqiang, SUN Yunda, TANG Hu
2022, 31(5): 597-615. doi: 10.15953/j.ctta.2022.124
Since X-ray imaging technology is applied for the inspection of baggage and cargo, CT imaging has played an increasingly important role in the filed of security inspection. The traditional CT application model cannot meet the requirements of real-time inspection and release of baggage and cargo in transportation and logistics hub centers of customs and civil aviation, as well as the epidemic prevention and control in public places. Therefore, the CT intelligent solution, with fast checking, verified accuracy and effective interception, has been developing rapidly. It is able to effectively improve the baggage inspection efficiency and the passenger passing efficiency, and also effectively guarantee the national security and the passenger personal and property safety. In conclusion, the CT intelligent solution is of a huge potential and broad market prospect in the filed of inspection and release of baggage and cargo.
Medical Imaging and Image Processing
Application of Computer-aided Diagnosis System Based on Deep Learning in Rib Fracture Diagnosis
XIONG Shan, CHEN Bo, MAO Jie, LIU Sibin, HUANG Yuanyi, CHENG Jianmin
2022, 31(5): 617-622. doi: 10.15953/j.1004-4140.2022.31.05.08
Objective: To investigate the application value of computer-aided diagnosis (CAD) system based on deep learning (DL) in rib fracture diagnosis. Methods: The CT images of 232 patients with chest trauma were analyzed retrospectively and the films were read in three ways. CAD system reading: using CAD system to detect and record the results of rib fracture; radiologists reading: two radiologists with 6 years of CT diagnosis experience read the film independently and the diagnostic results were based on the consensus of them; radiologists reading with the assistance of CAD system: one month later, the same two radiologists reassessed the images with the aid of the CAD system using a joint reading mode. Gold standard: two senior radiologists with more than 15 years of experience in the CT diagnosis of rib fractures read the radiographs independently and the consensus of them was used as the diagnostic standard. The sensitivity, false-positive rate and the reading time of the three methods were calculated and compared. Results: A total of 712 rib fractures were found in 232 patients. The reading sensitivity of the CAD system was 81.2%, which was lower than that of the radiologists, and the reading sensitivity of the radiologists was lower than that of CAD system-assisted radiologists. The false positive rate of CAD system was 0.48±0.13 and was the highest . There was no statistical difference in the false-positive rate between radiologists and CAD system-assisted radiologists. The reading time of the CAD system was (2.45±0.92)s and was the shortest. The reading time of CAD system-assisted radiologists was less than that of radiologists and the reading time was reduced by 34.2%. Conclusion: To further improve the sensitivity and reduce the false positive rate is an important part of CAD improvement; the use of CAD system based on deep learning to assist radiologists in reading images can improve the sensitivity of rib fracture diagnosis and reduce the time of reading images without increasing the false positive rate.
The Study of Diffusion Kurtosis Imaging on Microstructural Changes of Substantianigra in Patients with Early Parkinson's Disease
DONG Jiake, DENG Min, YANG Tao, KONG Xuqiang, LIU Kangyong, CAI Sipeng, WANG Tao, YALIKUN· Munire
2022, 31(5): 623-629. doi: 10.15953/j.ctta.2022.010
Objective: To explore the diagnostic value of diffusion kurtosis imaging (DKI) the microstructure changes of substantianigra in patients with early Parkinson's disease (PD). Methods: 30 patients with early PD were collected as the PD group while 20 healthy volunteers as the control group. All the subjects underwent brain routine magnetic resonance imaging (MRI) and DKI examination. The DKI scan data were processed to obtain quantitative parameter maps, with the three paramrters as fractional anisotropy (FA), mean diffusivity (MD) and mean kurtosis (MK). The DKI parameters of substantianigra (SN) were measured respectively and the differences between the two groups were statistically analyzed. The receiver operating characteristic (ROC) curve was used to evaluate the diagnostic efficacy of DKI parameters in patients with early PD. Results: The FA value of the substantianigra of the PD group was lower than that of the healthy control group, and the difference between the two groups was statistically significant. Compared with the healthy control group, the MD value of the substantianigra in the PD group was slightly higher, and the difference between the two groups was not statistically significant. The MK value of the substantianigra in the PD group was significantly higher than that in the healthy control group, and the difference between the two groups was statistically significant. The ROC curve showed the AUC of FA value, MD value and MK value in the substantianigra in the diagnosis of early PD were 0.716, 0.613 and 0.820, respectively. The substantianigra MK value showed certain accuracy in the diagnosis of early PD patients, and the diagnostic sensitivity and specificity were higher than the FA value and MD value. Conclusions: DKI can be used to analyze the changes of the microstructure of substantianigra in early PD patients, and the MK value can better reflect the changes of the microstructure of substantianigra in early PD patients.
Study on the Influence of Head and Neck CTA Image Quality under Different Tube Voltages
WANG Yongsheng, ZHANG Pengyu, LI Fangzhong, WANG Zhenghua, LU Mengyun, CHEN Lei, HE Junlin, CHEN Wenjing
2022, 31(5): 631-638. doi: 10.15953/j.ctta.2022.026
Objective: To explore the influence on the image quality of head and neck CTA under different tube voltages. Materials and methods: A total of 67 patients with clinically suspected head and neck vascular diseases in Tinglin hospital from September 2021 to January 2022 were collected as the research objects, including 37 males and 30 females, with an average age of 68.94. According to the applied tube voltage, the patients were randomly assigned into groups A, B and C. There were 23 cases of 120 kV in group A, 24 cases of 100 kV in group B and 20 cases of 80 kV in group C. Single factor ANOVA test was used to compare the CT value, SD value, SNR, CNR of aortic arch, internal carotid artery, M1 segment of middle cerebral artery and sternocleidomastoid muscle among the three groups, and furthermore the subjective score of image quality, TDLP and ED. Kappa test was used to evaluate the consistency of subjective scores of the two physicians. Results: There were statistical differences in CT and SD values of aortic arch, internal carotid artery and M1 segment of middle cerebral artery among the three groups; The CT and SD values of aortic arch, internal carotid artery and M1 segment of middle cerebral artery increased with the decrease of tube voltage, with an increase range of 15%~55%. There was no statistical difference in SNR and CNR among the three groups; There was statistical difference in subjective score of image quality among the three groups; Group B > group A > group C, and group B showed the highest score; the kappa value of the subjective score of the two phsicianss=0. 80, which was highly consistent; There were statistical differences in TDLP and ED among the three groups, TDLP and ED decreased with the decrease of tube voltage, with the decrease ranges of 25% and 39%. Conclusion: According to the requirements of CT radiation dose diagnosis reference level, 100 kV is the best tube voltage condition for head and neck CTA.
Spectral CT
A Preliminary Study of Spectral CT in Predicting Pfirrmann Grading of Lumbar Intervertebral Disc
LI Rui, DU Chunling, DENG Kai, SU Yuwen, ZHANG Man, WANG Lulu, LI Wen
2022, 31(5): 639-646. doi: 10.15953/j.1004-4140.2022.31.05.11
Objective: To investigate the predictive value of material separation technology of spectral CT in Pfirrmann grading of lumbar intervertebral disc degeneration. Methods: Retrospective analysis was performed on 30 patients with lumbar disc herniation in our hospital from October 2020 to February 2021. Spectral CT scan and MRI scan were performed respectively. Grade 1-3 of Pfirrmann grading was classified into low grade group, while grade 4-5 was classified into high grade group. Water (calcium) concentration, water (HAP) concentration, calcium (water) concentration, HAP (water) concentration and Eff-Z of intervertebral disc were measured by spectral post-processing analysis software in the same ROI. Independent sample t est was used to compare the differences among parameters, furthermore, the ROC curve was drawn. The area under the curve was used to evaluate the diagnostic efficiency and select the optimal diagnostic threshold. Results: The concentrations of water (calcium) and water (HAP) in low-grade intervertebral discs were higher than those in high-grade intervertebral discs, while the concentrations of water (calcium), water (HAP) and Eff-Z in low-grade intervertebral discs, were lower than those in high-grade intervertebral discs, which held statistical significance. The ROC curve showed that water (calcium) concentration and water (HAP) concentration were less effective in diagnosing the difference between low-grade and high-grade discs. Calcium concentration holds certain diagnostic efficacy. HAP and Eff-Z hold high diagnostic efficacy. while Eff-Z shows better diagnostic efficacy with AUC of 0.97. Taking 7.69 as the standard, the sensitivity and specificity of differentiating low-grade and high-grade intervertebral discs are respectively 96.25% and 96.00%. Conclusion: Spectral CT with multi-parameters quantitative analysis holds certain value in distinguishing low grade and high grade intervertebral discs.
The Characteristics and Its Potential Application Value of Pancreatic Ductal Adenocarcinoma with Single-source Dual Energy Spectral CT Imaging
ZHANG Jing, ZHANG Ting, LI Honglei, CHEN Xiaoyan, LI Weixia, TANG Rongbiao, PAN Zilai, CHEN Kemin, BU Yulian, FANG Weihuan, LIN Xiaozhu, YAN Fuhua
2022, 31(5): 647-654. doi: 10.15953/j.ctta.2022.049
Objective: To study the quantitative multi-parameters of CT energy spectrum imaging of pancreatic ductal adenocarcinoma by using enhanced CT energy spectrum imaging, and to explore the parameter characteristics and potential application value of CT energy spectrum imaging of pancreatic ductal adenocarcinoma. Methods: Retrospective analysis was performed on 61 patients with pancreatic ductal adenocarcinoma confirmed by pathology, all of whom underwent enhanced CT scan of advanced arterial and portal vein phase. Single energy CT value, Effective atomic number (Effective-Z), iodine (water) base value and water (iodine) base value of pancreatic cancer lesions and pancreatic parenchyma were recorded. Paired t test (normal distribution) or Wilcoxon signed-rank sum test (non-normal distribution) were used to compare the differences of the above CT energy spectrum imaging parameters between the late arterial period and portal vein period, pancreatic cancer and pancreatic parenchyma, and to plot the corresponding energy spectrum curves. Results: The single energy CT value of pancreatic ductal adenocarcinoma in advanced arterial stage was significantly lower than that in peripheral normal pancreas. The single energy CT value of pancreatic ductal adenocarcinoma in portal vein stage was higher than that in arterial stage and lower than that in peripheral normal pancreas. The lower the low energy segment is, the more significant the difference is. The higher the energy segment is, the smaller the difference is. There were significant differences in single energy CT value, paired base substance iodine (water) base value and Effective-Z corresponding homogenization concentration between pancreatic ductal adenocarcinoma and pancreatic parenchyma in arteriovenous phase, while there was no significant difference in water (iodine) value. Conclusion: The dual-phase enhanced energy spectrum parameters of CT imaging of pancreatic ductal adenocarcinoma have certain characteristics. The single energy CT value, iodine (water) base value and effective atomic number of pancreatic cancer in arteriovenous phase were lower than those in corresponding pancreatic parenchyma, and those in arterial late phase were lower than those in portal vein phase. The lower the energy segment, the more obvious the difference. The comprehensive application of multi-band and energy spectrum multi parameter imaging is helpful to improve the CT density resolution of pancreatic ductal adenocarcinoma, and has the potential value of providing image information for early diagnosis of small pancreatic cancer.
Medical CT
Diagnostic Research of CT Combined with Serum CA125 and HE4 in Ovarian Epithelial Malignant Tumor
WANG Desheng, KAN Fanggong, MA Zhoupeng
2022, 31(5): 655-661. doi: 10.15953/j.1004-4140.2022.31.05.13
Objective: To explore the diagnostic value of CT combined with serum CA125 and HE4 in ovarian epithelial malignant tumor. Methods: 156 cases of epithelial ovarian tumors (benign 72 cases, malignant 84 cases) were studied. Detections of CT and serum tumor markers CA125 and HE4 were carried out before operation, the diagnostic results were compared with pathology, for the evaluation of the diagnostic value of CT, CA125, HE4 and combined diagnosis in epithelial ovarian tumors. Results: The positive rates of serum CA125 and HE4 of malignant tumor group were higher than those of benign tumor group; The diagnostic sensitivity of CA125 for malignant tumor was higher than HE4, and the specificity was lower than HE4; the diagnostic accuracy of CA125 combined with HE4 for malignant tumor was higher than CT; the sensitivity, specificity and accuracy of CT combined with serum CA125 and HE4 for malignant tumor was 95.2%, 88.9% and 92.3% respectively, and were higher than CT or tumor markers alone. Conclusion: CT combined with serum CA125 and HE4 shows important value in the diagnosis of ovarian epithelial malignant tumor, which is conducive to early diagnosis and precise staging, and is worthy of promoting in application.
The CT and MRI Findings and Differential Diagnosis of Sebaceous Carcinoma of Eyelid and Basal Cell Carcinoma
WU Shuang, CAO Wei, ZHANG Tao, XU Xu
2022, 31(5): 662-668. doi: 10.15953/j.1004-4140.2022.31.05.14
Objective: To improve the ability of differential diagnosis of sebaceous carcinoma of eyelid and basal cell carcinoma by investigating their CT and MRI findings. Methods: The clinical data and imaging findings of 14 patients with sebaceous carcinoma of eyelid and 7 patients with basal cell carcinoma confirmed by operation were retrospectively analyzed and compared. Results: Among the 14 patients with SC, 6 cases underwent CT examination, 7 cases underwent the MRI examination, and 1 case underwent both CT and MRI. Among the 7 patients with BCC, 1 case underwent CT examination, and 5 cases underwent the MRI plain scan, and 1 case underwent both CT and MRI. 92.9% of SC cases were female (13/14). 50% of SC were located in the upper eyelid, and most of them were in the shape of ring strip and nodule (12/14). Half of the lesions had unclear boundaries (7/14). The arc sign (5/14) and gas sign (4/7) often appeared in the lesions. One case invaded the adjacent orbital tissue. The proportion of male and female patients in BCC cases was similar. The lesions of BCC were all located in the lower eyelid (7/7). Nearly half of the shape of the lesions was nodular (3/7). Most of them had clear boundaries (6/7). Calcification was found in one case, and no invasion of adjacent tissue was found. Conclusion: There are some differences in epidemiology, location and imaging features between sebaceous carcinoma of eyelid and basal cell carcinoma. Mastering the key points of differentiation between sebaceous carcinoma of eyelid and basal cell carcinoma can improve the accuracy of preoperative qualitative diagnosis.
Evaluation of the Accuracy of Infer Read Software in Measuring the Volume of Pure Ground Glass Nodules in the Lung
LIANG Yuan, ZHANG Huihui, WU Jianlin
2022, 31(5): 669-678. doi: 10.15953/j.ctta.2021.009
Objective: To discuss the accuracy of the automatic measurement of the volume of pure ground glass nodules (pGGN) by the artificial intelligence pulmonary nodule detection software and the influencing factors of the measurement error. Methods: 170 pGGNs from 90 patients who underwent routine chest CT scan from January 1 to 31, 2021 in our hospital were selected in this retrospective study. The original CT scan data (including 1 mm thin-slice images) was sent to the AI server of Inference Technology to automatically measure the volume of lung nodules and record the measurement data. Two senior chest imaging diagnosticians manually carried out pGGNs layer-by-layer measurement and added the volume value, then took the average of three measurements as the "gold standard" data to compare with the AI measurement results, and then analyzed the influence of pGGN location, size, proximity and other factors on the AI measurement error. SPSS 26.0 was used for statistical analysis. Results: Among the total 170 pGGNs in the 90 patients in this study, 49 (28.82%) were in the right upper lobe, 21 (12.35%) were in the right middle lobe, 27 (15.88%) were in the right lower lobe, and left upper lobe 49 patients (28.82%) were involved, and 24 patients were in the lower lobe of the left lung (14.12%). Among the adjacent relationships of pGGN, 82 (48.24%) were completely located in the lung parenchyma without adjacency, 29 (17.06%) were close to the blood vessel, and 59 (34.70%) were close to the pleura. (1) There was no statistically significant difference in the volume values of pGGNs between two observers and the same observer at different time points. (2) For the measurement of the same pGGN, there was no statistically significant difference between the results of automatic measurement by AI and manual measurement and the correlation between the two was quite high (r=0.981), and the agreement was also very high (ICC value is 0.987). (3) The size, location, and adjacent relationship of pGGN lesions were not statistically significant for the error of AI volume measurement. Conclusions: The InferRead lung nodule measurement software shows high accuracy in the measurement of lung pGGN three-dimensional volume, which can be applied in clinical lung nodule diagnosis and related research. (2) For the measurement of the same pGGN volume, there was no difference between the results of automatic measurement by AI and manual measurement, and the correlation between them was quite high (r = 0.981), and the consistency was high (ICC value is 0.987). (3) The volume size, occurrence position and adjacent relationship of pGGN have no statistical significance on the error of AI volume measurement. Conclusion: The InferRead lung nodule detection software shows high accuracy in measuring the three-dimensional volume of lung pGGN, and can be applied to clinical diagnosis and related research of lung nodules.
Application of CT in the Prognostic Analysis of Ureteral Carcinoma
DAI Maoliang, DONG Jingyan, PAN Yulong, WU Bing, GAO Jin
2022, 31(5): 679-685. doi: 10.15953/j.ctta.2021.010
Objective: To discuss the application value of CT in the prognostic analysis of ureteral carcinoma. Methods: The CT and medical records of 101 patients with ureteral carcinoma who underwent radical nephroureterectomy from January 2014 to December 2016 were collected and analyzed, and then the Cox regression method was performed. Results: Cox uni-variate analysis revealled that the single risk factors affecting the survival rate of patients with ureteral cancer in CT signs included multiple lesions, necrotic cystic change, maximum diameter ≥3 cm, peritumoral fat infiltration and depth of infiltration, invasion of adjacent organs and tissues, and lymph node metastasis. Cox multi-variate analysis,revealled that the multiple risk factors affecting the long-term survival rate of patients with ureteral cancer in CT signs included peritumoral fat infiltration and depth of infiltration, invasion of adjacent organs and tissues, lymph node metastasis, tumor necrosis and cyst, and multiple lesions. Conclusion: In CT signs, peritumoral fat infiltration and depth of infiltration, invasion of adjacent organs and tissues, lymph node metastasis, tumor necrosis and cystic degeneration and multiple lesions are the main risk factors affecting the survival of patients with ureteral cancer. CT holds great application value in the prognostic analysis of primary ureteral cancer.
Research Progress on CT Radiomics of Esophageal Cancer
ZHOU Jianwen, FENG Feng
2022, 31(5): 687-696. doi: 10.15953/j.ctta.2021.006
Esophageal cancer which results in a relatively higher death rate is one of the most common cancers throughout the world. CT radiomics is a study to extract radiomics characteristics which are generalized from a large amount of CT images. These characteristics then underwent high-throughput quantitative analysis so that more heterogeneity information of the caner can be obtained. CT radiomics has been gradually used in forecast clinical stages and pathological differentiation of esophageal cancer in recent years, and applied in assessment of treatment effect and prognosis evaluation as well. This paper focuses on the application of CT radiomics in esophageal cancer and its progress.