A machine learning-based strategy for predicting the mechanical strength of coral reef limestone using X-ray computed tomography

Different sedimentary zones in coral reefs lead to significant anisotropy in the pore structure of coral reef limestone(CRL),making it difficult to study mechanical behaviors.With X-ray computed tomography(CT),112 CRL samples were utilized for training the support vector machine(SVM)-,random forest(RF)-,and back propagation neural network(BPNN)-based models,respectively.Simultaneously,the machine learning model was embedded into genetic algorithm(GA)for parameter optimization to effectively predict uniaxial compressive strength(UCS)of CRL.Results indicate that the BPNN model with five hidden layers presents the best training effect in the data set of CRL.The SVM-based model shows a tendency to overfitting in the training set and poor generalization ability in the testing set.The RF-based model is suitable for training CRL samples with large data.Analysis of Pearson correlation coefficient matrix and the percentage increment method of performance metrics shows that the dry density,pore structure,and porosity of CRL are strongly correlated to UCS.However,the P-wave velocity is almost uncorrelated to the UCS,which is significantly distinct from the law for homogenous geomaterials.In addition,the pore tensor proposed in this paper can effectively reflect the pore structure of coral framework limestone(CFL)and coral boulder limestone(CBL),realizing the quantitative characterization of the heterogeneity and anisotropy of pore.The pore tensor provides a feasible idea to establish the relationship between pore structure and mechanical behavior of CRL.

Hybrid reconstruction algorithm for computed tomography based on diagonal total variation

Inspired by total variation(TV), this paper represents a new iterative algorithm based on diagonal total variation(DTV) to address the computed tomography image reconstruction problem. To improve the quality of a reconstructed image, we used DTV to sparsely represent images when iterative convergence of the reconstructed algorithm with TV-constraint had no effect during the reconstruction process. To investigate our proposed algorithm, the numerical and experimental studies were performed, and rootmean-square error(RMSE) and structure similarity(SSIM)were used to evaluate the reconstructed image quality. The results demonstrated that the proposed method could effectively reduce noise, suppress artifacts, and reconstruct highquality image from incomplete projection data.

Relation between Epicardial Adipose Tissue Thickness Assessed by Multidetector Computed Tomography and Significance of Coronary Artery Disease

Objective: To evaluate the relation between epicardial adipose tissue (EAT) thickness and also pericoronary fat assessed by Multidetector Computed Tomography (MDCT) with both calcium score and significance of coronary artery disease. Background: Epicardial adipose tissue (the visceral fat of the heart present under the visceral layer of the pericardium) has the same origin of abdominal visceral fat, which is known to be strongly related to the development of coronary artery atherosclerosis. Multidetector CT (MDCT) provides an accurate and reproducible quantification of EAT due to its high spatial and temporal resolution. Patients and Methods: The current study included 70 patients with low-intermediate probability of coronary artery disease. All patients were subjected to 256 Multidetectors CT to assess EAT thickness, the mean thickness of the pericoronary fat surrounding the three coronary arteries and coronary calcium score. Also coronary CT angiography was done and patients were then divided into 3 groups according to significance of coronary atherosclerosis: Group 1: No atherosclerosis (20 patients), Group 2: Non obstructive atherosclerosis (luminal narrowing less than 50% in diameter) (25 patients), Group3: Obstructive atherosclerosis (luminal narrowing ≥ 50%) (25 patients). Results: The mean EAT thickness and the mean pericoronary fat thickness were significantly higher in patients with obstructive coronary artery disease (CAD) with stenosis > 50% (group 3) compared to other groups with normal coronaries or non obstructive (CAD). ROC curve was used to define the best cut off value of the thickness of both EAT and pericoronary fat in predicting the obstructive CAD group which was ≥7.2 and 12.6 mm for epicardial and pericoronary fat respectively. Also there is a positive correlation between both epicardial adipose tissue and pericoronary fat thickness and the coronary calcium score. Conclusion: EAT thickness and pericoronary fat thickness can be used in predicting the significance of coronary artery disease.

Acute coronary syndrome on non-electrocardiogram-gated contrastenhanced computed tomography

It is not rare for acute coronary syndrome(ACS)patients to present with symptoms that are atypical,rather than chest pain.It is sometimes difficult to achieve a definitive diagnosis of ACS for such patients who present with atypical symptoms,normal initial biomarkers of myocardial necrosis,and normal or nondiagnostic electrocardiograms(ECGs).Although cardiac CT allows for assessments of coronary artery stenosis as well as myocardial perfusion defect in patients with suspected ACS,it requires ECG gating and is usually performed with high-performance multislice CT for highly probable ACS patients.However,several recent reports have stated that ACS is detectable by myocardial perfusion defects even on routine non-ECG-gated contrast-enhanced CT.A growing number of contrast-enhanced CT scans are now being performed in emergency departments in search of pathologies responsible for a patient’s presenting symptoms.In order to avoid inappropriate management for this life-threatening event,clinicians should be aware that myocardial perfusion defect is more commonly detectable even on routine non-ECG-gated contrast-enhanced CT performed in search of other pathologies.

Comparison of computed tomography and magnetic resonance imaging for the detection of mandibular condylar osteochondroma

Objective The purpose of this study was to compare computed tomography(CT) and magnetic resonance imaging(MRI) for the detection of mandibular condylar osteochondroma.Methods Preoperative CT and MRI of 33 patients with unilateral condylar osteochondroma were reviewed. The morphology, location, continuity with the parent bone, cartilage cap, perichondrium of tumors, and changes in soft and hard tissues adjacent to the lesions were investigated by two reviewers. Data were analyzed using Mc Nemar test. A P value < 0.05 was considered significant.Results Among the 33 condylar osteochondromas, 11 were of the diffuse type, 10 were of the sessile type, and 12 were of the pedunculated type. Continuity with the cortex and marrow of the host condyle was observed on both CT and MRI. Both modalities had identical detection rates of surface reconstruction of the temporal bone joint, condylar dislocation, and pseudarthrosis formation. However, MRI showed significantly higher detection rates of the cartilage cap and perichondrium than CT(P < 0.05). Furthermore, MRI showed ipsilateral and contralateral temporo-mandibular joint(TMJ) disc displacement in 4 cases and 6 cases, respectively, and ipsilateral and contralateral TMJ effusion in 20 cases and 14 cases, respectively.Conclusion CT can intuitively display the morphology and spatial location of condylar osteochondromas through three-dimensional reconstruction. MRI may be superior to CT in the detection of cartilage cap, perichondrium of the condylar osteochondroma, and changes in the TMJ and adjacent soft tissues.

Qualitative visual trichotomous assessment improves the value of fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography in predicting the prognosis of diffuse large B-cell lymphoma

Introduction:Fluorine-18 fluorodeoxyglucose(18F-FDG)positron emission tomography/computed tomography(PET/CT)is a powerful tool for monitoring the response of diffuse large B-cell lymphoma(DLBCL)to therapy,but the criteria to interpret PET/CT results remain under debate.We investigated the value of post-treatment PET/CT in predicting the prognosis of DLBCL patients when interpreted according to qualitative visual trichotomous assessment(QVTA)criteria compared with the Deauvil e criteria.Methods:In this retrospective study,final PET/CT scans of DLBCL patients treated with rituximab-based regimens between October 2005 and November 2010 were interpreted using the Deauvil e and QVTA criteria.Survival curves were estimated using Kaplan-Meier analysis and compared using the log-rank test.Results:A total of 253 patients were enrol ed.The interpretation according to the Deauvil e criteria revealed that 181patients had negative PET/CT scan results and 72 had positive results.The 3 year overal survival(OS)rate was significantly higher in patients with negative scan results than in those with positive results(91.6%vs.57.5%,P<0.001).The72 patients with positive scan results according to the Deauville criteria were divided into two groups by the interpretation according to the QVTA criteria:29 had indeterminate results,and 43 had positive results.The 3 year OS rate was significantly higher in patients with indeterminate scan results than in those with positive results(91.2%vs.33.5%,P<0.001)but was similar between patients with negative and indeterminate scan results(91.6%vs.91.2%,P=0.921).Conclusions:Compared with the Deauvil e criteria,using the QVTA criteria for interpreting post-treatment PET/CT scans of DLBCL patients is likely to reduce the number of false positive results.The QVTA criteria are feasible for therapeutic outcome evaluation and can be used to guide risk-adapted therapy.

Assessment of Organ Dose by Direct and Indirect Measurements for a Wide Bore X-Ray Computed Tomography Unit That Used in Radiotherapy

The aim of this study was to investigate the organ doses of patients undergoing computed tomography (CT) examination using the wide bore General Electric (GE) “Light Speed RT” unit. The head, chest and pelvic regions of the Rando-phantom were scanned with 120 kV, 200 mA, and 2.5 mm slice thickness for helical and axial modes. Thermoluminescent Dosimeter (TLD) pairs were used for the dosimetry of 10 organs. TL-counts were converted to dose by using CTDIcenter dose on CT-phantom. For the calculation of the organ doses, the ImPACT software was utilized by entering CTDIair (100 mAs) in small and large field of view (26.43 and 21.17 mGy respectively). The in-field dose ranges in helical and axial modes were 64.3 - 38 mGy and 47.6 - 19.7 mGy in head, 48.3 - 14.1 mGy and 34.1 - 10 mGy in chest, 28.4 - 10.2 mGy and 21 - 8.5 mGy in pelvic, respectively. The organ doses from software and TLD were compared and tailored as the in-field and the out-field radiation. First results showed that the organ dose was relatively higher in the helical mode on both direct and indirect measurement. The in-field organ dose differences between TLD and software were seen. In helical and axial modes, the dose differences ranged from +1 to +13.3 and -8.3 to +9.6 mGy for head exam, +1.1 to +15.3 and +0.3 to +9.1 mGy for chest, and -21.7 to +1.9 and -15.5 to +1.8 mGy for pelvic. The availability of this program for organ dose calculations by measuring CTDIair value for CT device used in the radiotherapy would be considered valuable.

Impact of random and scattered coincidences from outside of field of view on positron emission tomography/computed tomography imaging with different reconstruction protocols

Image quality in positron emission tomography(PET)is affected by random and scattered coincidences and reconstruction protocols.In this study,we investigated the effects of scattered and random coincidences from outside the field of view(FOV)on PET image quality for different reconstruction protocols.Imaging was performed on the Discovery 690 PET/CT scanner,using experimental configurations including the NEMA phantom(a body phantom,with six spheres of different sizes)with a signal background ratio of 4:1.The NEMA phantom(phantom I)was scanned separately in a one-bed position.To simulate the effect of random and scatter coincidences from outside the FOV,six cylindrical phantoms with various diameters were added to the NEMA phantom(phantom II).The 18 emission datasets with mean intervals of 15 min were acquired(3 min/scan).The emission data were reconstructed using different techniques.The image quality parameters were evaluated by both phantoms.Variations in the signal-to-noise ratio(SNR)in a 28-mm(10-mm)sphere of phantom II were 37.9%(86.5%)for ordered-subset expectation maximization(OSEM-only),36.8%(81.5%)for point spread function(PSF),32.7%(80.7%)for time of flight(TOF),and 31.5%(77.8%)for OSEM+PSF+TOF,respectively,indicating that OSEM+PSF+TOF reconstruction had the lowest noise levels and lowest coefficient of variation(COV)values.Random and scatter coincidences from outside the FOV induced lower SNR,lower contrast,and higher COV values,indicating image deterioration and significantly impacting smaller sphere sizes.Amongst reconstruction protocols,OSEM+PSF+TOF and OSEM+PSF showed higher contrast values for sphere sizes of 22,28,and 37 mm and higher contrast recovery coefficient values for smaller sphere sizes of 10 and 13 mm.

Conception the Fluid Flow Behavior within Oil Reservoir Rock by Using Computed Tomography(CT)Scan

The behavior of fluid flow has been studied during the different flow media over the past decades.In addition,the behavior of the flow of fluid through porous media has garnered much research interest.This paper sheds light on fissured rocks of oil reservoir media(as one of the porous media domain),and the effect of these fissured on fluid flow.In this article,the Finite Volume Method(FVM)has been used to visualize the behavior of single-phase fluid flow in an actual core according to the dual-porosity dual permeability model.The study was conducted in two parts,the first was the image processing for one of the real oil reservoir fractured rock images,where the image was processed and simulated by ANSYS-CFX software,and the results showed a complete visualizing of the fluid behavior during this domain.As for the other side,a simulation of a real reservoir rock belonging to the Al-Nour field in Iraq/Misan was made.The X-ray Computed Tomography(CT)scan has been used to convert the real fractured core to a dynamic domain.ANSYS-CFX program has been used and the results illustrated the pressure counter,the velocity counter,the velocity streamline,and the velocity vectors for the studied model in three dimensions.A comparison was made between the productivity index for fractured and non-fractured rock and the results explained that the presence of fracture can improve the productivity index to about 5.74%.

Ultrasonic Computed Tomography Imaging Method of Concrete Materials Based on Simulated Annealing Genetic Algorithm

Stability and accuracy of the imaging results are still unmet practical demands for ultrasonic computed tomography(CT)of concrete material.To address these issues,a CT technique based on simulated annealing genetic algorithm(SAGA)is presented in this work.Firstly,a natural weight matrix with clear physical meaning is introduced in the inverse algorithm and then a quadric broadening objective function is formed according to the propagation characteristics of ultrasound in concrete.After that,the simulated annealing(SA)searching is added to speed up the inverse process and to improve the convergence and stability of the algorithm.Finally,the optimal inverse imaging results have been achieved by variable ectopic adaptive genetic algorithm.The numerical simulation experiments have shown that the usage of the correct priori information and the excellent characteristic of SAGA in searching the global minimum value of the function have produced accurate and effective results with stable numerical values.The imaging resolution is improved and the imagining results reflecting the inner defections of the tested objects are more reliable and accurate.

Numerical analysis of the failure process of soil-rock mixtures through computed tomography and PFC3D models

Soil-rock mixture （SRM） is a unique type of geomaterial characterized by a heterogeneous composition and a complicated structure. It is intractable for the continuum-based soil and rock mechanics theories to accurately characterize and predict the SRM＇s mechanical properties. This study reports a novel numerical method incorporating microfocus computed tomography and PFC3D codes to probe the deformation and failure processes of SRM. The three-dimensional （3D） PFC models that represent the SRM＇s complex structures were built. By simulating the entire failure process in PFC3D, the SRM＇s strength, elastic modulus and crack growth were obtained. The influence of rock ratios on the SRM＇s strength, deformation and failure processes, as well as its internal mesoscale mechanism, were analyzed. By comparing simulation results with experimental data, it was verified that the 3D PFC models were in good agreement with SRM＇s real structure and the SRM＇s compression process, deformation and failure patterns; its intrinsic mesomechanism can be effectively analyzed based on such 3D PFC models.

Imaging Findings in Chest Computed Tomography: Initial Experience in a Developing Country

Background:?High resolution chest computed tomography (HRCT) is an established imaging modality that accurately assesses disease processes that are non-specific on conventional chest radiography. Chest Computed Tomography (CT) has proven invaluable in the evaluation of various diseases of the chest providing precise diagnosis that have great impact on the course of management. In this study, our aim is to evaluate the spectrum of radiological findings in all patients that had chest CTs at a tertiary health Institution in a developing Country. Method: This is a 6-year retrospective study of all patients that had Chest CT from 2009 till 2015. All patient data were retrieved and their Chest CT images reviewed and radiological findings and diagnosis documented. Results: Age range of patients was 1 - 84 years with a mean of 52.36 ± 22.9 years. Ten patients were children (12%). There were more males 53 (63.1%) than females 31 (36.9%) in the study. Referral for imaging was predominantly from internal medicine 33 (47.1%) and Surgery 27 (38.6%) departments. Most of the patients, 53 (63.1%) did not have previous chest radiographs. Consolidative changes were the most common parenchymal finding (77.4%). Few mediastinal masses were found in the study population. CT angiography constituted only 10.7% of all chest CTs done in adults. Conclusion: Few chest CT evaluations are performed compared to CTs of other parts of the body done in the same period. Chest CT is generally under-utilized even in a tertiary institution. There is a need to educate chest physicians on the role and advantages of CT use for the management of their patients.

Four Chamber Normative Data for Computed Tomography Angiography in the Axial Plane

Background: Computed Tomography Angiography (CTA) is a new and promising type of imaging to evaluate cardiac structure and function. Although it does require injection of contrast material and some radiation exposure, it is much less invasive than conventional catheter based on angiography. It has been already shown to correlate well with echocardiography images, but normative values have yet to be established. Objectives: To publish the first known data on CTA measurements for all four chambers of the heart. Methods: We conducted a retrospective chart review study of around 3000 patients over a three-year period (2006-2008) from a private cardiology practice in Phoenix, Arizona. Patients were screened for any preexisting conditions that might contribute to abnormal cardiac structures. A “normal” population of 226 patients’ CTA’s was reviewed and 29 different measurements were taken cardiac anatomy. Patients were then categorized based on sex and BMI and the cardiac measurements were compared. Results: The values obtained were similar to already established normal measurement values from echocardiography. There was good correlation between body mass index (BMI) and the sexes. Conclusion: This study serves to establish normative data for cardiac CTA while also proves to show good correlation with already established normative values for echocardiography and justifies these CTA measurements as an alternative means of cardiac evaluation. Further study regarding the variations between data for specific measurements will be addressed in future studies.

Usefulness of Chest Computed Tomography for Diagnosis of Idiopathic Pneumomediastinum with Negative Findings on Plain X-Ray

Idiopathic pneumomediastinum is rare in children. Few cases of patients with pneumomediastinum show negative findings on X-ray examination. Chest computed tomography (CT) was very useful for the diagnosis and evaluation of the extent of pneumomediastinum. We report here a case of idiopathic pneumomediastinum in a 15-year-old boy who exhibited no significant chest X-ray finding. The patient was referred to our institute for the further evaluation of pre-cordial pain and breathing difficulty. Precordial pain suddenly developed, when he was carrying a portable shrine on his shoulder (day of onset). He was admitted to another institute 3 days after onset because of worsening precordial pain. On admission, he presented with 98% saturation of hemoglobin in the peripheral blood under room air. Plain chest X-ray also revealed no abnormal findings. A half-dissolved gastrographin swallow showed no leakage of gastrographin from the pharynx and esophagus to the mediastinum, and no diverticulum within the esophagus. Plain chest CT revealed extensive emphysema around the trachea from the neck to the portion inferior to the carina of trachea. The patient was diagnosed with idiopathic pneumomediastinum because the cause was unclear. We decided to admit him to our institute under fasting conditions and rest. His symptoms improved 3 days after onset. The lesion had disap-peared 8 days after onset on chest CT. When young people experience precordial pain which increases on inspiration, we must consider pneumomediastinum in a differential diagnosis, and it is important to perform chest CT.

Evaluation of response to gemcitabine plus cisplatin-based chemotherapy using positron emission computed tomography for metastatic bladder cancer

BACKGROUND The purpose of the present study was to examine retrospectively the contribution of 18Fluorodeoxyglucose positron emission tomography computed tomography(18FDG-PET/CT)to the evaluation of response to first-line gemcitabine plus cispla-tin-based chemotherapy in patients with metastatic bladder cancer.AIM To evaluate the response to Gemcitabine plus Cisplatin-based chemotherapy using 18FDG-PET/CT imaging in patients with metastatic bladder cancer.METHODS Between July 2007 and April 2019,79 patients underwent 18FDG-PET/CT imaging with the diagnosis of Metastatic Bladder Carcinoma(M-BCa).A total of 42 pa-tients(38 male,4 female)were included in the study,and all had been admi-nistered Gemcitabine plus Cisplatin-based chemotherapy.After completion of the therapy,the patients underwent a repeat 18FDG-PET/CT scan and the results were compared with the PET/CT findings before chemotherapy according to European Organisation for the Research and treatment of cancer criteria.Mean age was 66.1 years and standard deviation was 10.7 years(range:41–84 years).RESULTS Of the patients,seven(16.6%)were in complete remission,17(40.5%)were in partial remission,six(14.3%)had a stable disease,and 12(28.6%)had a pro-gressive disease.The overall response rate was 57.1 percent.CONCLUSION 18FDG-PET/CT can be considered as a successful imaging tool in evaluating response to first-line chemotherapy for metastatic bladder cancer.Anatomical and functional data obtained from PET/CT scans may be useful in the planning of secondline and thirdline chemotherapy.

Framework for COVID-19 Segmentation and Classification Based on Deep Learning of Computed Tomography Lung Images

Corona Virus Disease 2019(COVID-19) has affected millions of people worldwide and caused more than6.3 million deaths(World Health Organization, June 2022). Increased attempts have been made to develop deep learning methods to diagnose COVID-19 based on computed tomography(CT) lung images. It is a challenge to reproduce and obtain the CT lung data, because it is not publicly available. This paper introduces a new generalized framework to segment and classify CT images and determine whether a patient is tested positive or negative for COVID-19 based on lung CT images. In this work, many different strategies are explored for the classification task.ResNet50 and VGG16 models are applied to classify CT lung images into COVID-19 positive or negative. Also,VGG16 and ReNet50 combined with U-Net, which is one of the most used architectures in deep learning for image segmentation, are employed to segment CT lung images before the classifying process to increase system performance. Moreover, the image size dependent normalization technique(ISDNT) and Wiener filter are utilized as the preprocessing techniques to enhance images and noise suppression. Additionally, transfer learning and data augmentation techniques are performed to solve the problem of COVID-19 CT lung images deficiency, therefore the over-fitting of deep models can be avoided. The proposed frameworks, which comprised of end-to-end, VGG16,ResNet50, and U-Net with VGG16 or ResNet50, are applied on the dataset that is sourced from COVID-19 lung CT images in Kaggle. The classification results show that using the preprocessed CT lung images as the input for U-Net hybrid with ResNet50 achieves the best performance. The proposed classification model achieves the 98.98%accuracy(ACC), 98.87% area under the ROC curve(AUC), 98.89% sensitivity(Se), 97.99 % precision(Pr), 97.88%F-score, and 1.8974-seconds computational time.

Radiation Dose Survey of Pediatric Chest Computed Tomography Examinations: A Local Diagnostic Reference Levels Approach to Patient Safety

Objectives: The aim of this work was to initially establish both age and weight driven pediatric diagnostic reference levels (DRLs) for chest computed tomography (CT) examinations performed at tertiary care medical institution. Another aim was to compare the presented data with internationally published ones. This initial data shall serve as basis for establishing a national DRLs values for pediatric diagnostic CT examinations. Methods: Dosimetric indexes were collected for the chest examination for 93 patients during the past 2 years in a tertiary care medical city. Results: The results are within and below the international reported levels for chest CT in several countries. Conclusion: Continuous monitoring of the radiation doses received by the patients in computed tomography is continuous and ongoing process in order to ensure compliance and to optimize clinical imaging protocols. More extensive data acquisition and analysis are required to allow better understanding of the contributing factors leading to less patient radiation dose while preserving the clinical image quality. .

优化双下肢动脉能谱CT血管造影成像方案

目的优化双下肢动脉能谱CT血管造影(CTA)成像方案。方法回顾性分析30例疑诊双下肢动脉硬化性闭塞症(ASO)患者双下肢动脉能谱CTA资料,经重建获得40~80 keV(间隔5 keV)单能量(共9种)及100 kVp混合能量图像,对比双下肢动脉在不同图像中的CT值、噪声(SD)值、信噪比(SNR)及对比度噪声比(CNR);针对50、60 keV单能量和100 kVp混合能量图像质量及血管节段的可诊断性进行主观评估,观察40、45、50、60 keV单能量和100 kVp混合能量图像的自动去骨能力。结果40~80 keV范围内,随keV升高,各动脉在图像中的CT值、SD值、SNR及CNR均逐渐降低。相比100 kVp,腘动脉(PA)及其近端动脉的CT值、CNR及SNR均在40~55 keV图像中升高(P均<0.05);50~55 keV图像中SD值升高(P均<0.05),而60 keV图像中差异无统计学意义(P>0.05)。50及60 keV图像质量主观评分及可诊断动脉节段数与100 kVp差异均无统计学意义(P均>0.05)。PA以远节段的SNR及CNR在各单能量图像及100 kVp图像中差异均无统计学意义,其CT值在40~45 keV图像中、SD在40 keV图像中均高于100 kVp(P均<0.05),但SD在45 keV与100 kVp图像中差异无统计学意义(P>0.05);50 keV图像中,PA以远节段图像质量主观评分及可诊断节段数均高于60 keV及100 kVp(P均<0.05)。40 keV图像对9例(9/30,30.00%)、45 keV图像对6例(6/30,20.00%)不能自动去骨,50及60 keV、100 kVp对30例(30/30,100%)均可自动去骨。结论行双下肢动脉能谱CTA时,对PA及其近端节段以60 keV单能量成像较佳,对其以远或双下肢全程则以50 keV单能量成像较佳。

基于增强CT影像组学预测食管鳞癌淋巴血管侵犯状态的价值

目的:探讨基于增强CT影像组学预测食管鳞癌(ESCC)淋巴血管侵犯(LVI)的价值。方法:回顾性搜集行根治性切除术并经术后病理证实的224例食管鳞癌患者,其中包括66例LVI阳性和158例LVI阴性患者。所有患者均在术前2周内进行胸部增强CT扫描。将入组的患者按照7:3的比例随机分为训练集和测试集。使用3D Slicer软件逐层勾画全肿瘤感兴趣区(ROI),采用Python软件的Pyradiomics包提取肿瘤组织的影像组学特征,建立影像组学模型用于预测食管鳞癌的LVI状态并进行验证。采用受试者工作特征(ROC)曲线的曲线下面积(AUC)、敏感度、特异度、准确度、阳性预测值和阴性预测值来评价影像组学模型的诊断效能,使用校准曲线评价影像组学模型在训练集和测试集中的拟合程度。使用决策曲线分析(DCA)评价影像组学模型的临床应用价值。结果:从全肿瘤ROI中提取了1130个组学特征,经过筛选最终保留了7个影像组学特征,并使用多因素logistic回归建立影像组学预测模型。在训练集中,影像组学模型预测LVI的AUC值为0.930,敏感度为0.851,特异度为0.919,准确度为0.899,阳性预测值为0.816,阴性预测值为0.936;在测试集中,AUC值为0.897,敏感度为0.789,特异度为0.787,准确度为0.788,阳性预测值为0.600,阴性预测值为0.902。校准曲线显示影像组学模型在训练集及测试集中的预测概率与实际概率的一致性良好。DCA曲线显示影像组学模型具有良好的临床应用价值。结论:基于增强CT构建的影像组学模型,能够在术前有效预测食管鳞癌的LVI状态。

人工智能诊断系统及CT无创血流储备分数在评估高海拔地区 冠脉临界病变中的应用

目的探讨基于冠状动脉CT血管成像(CCTA)的人工智能(AI)诊断系统及CT无创血流储备分数(CT-FFR)在评估高海拔地区冠状动脉临界病变结构及功能学中的应用价值。方法前瞻性收集2022年1月~2023年10月青海大学附属医院冠心病临界病变患者164例,按居住地海拔进行分组,其中2000~3000m为A组(n=83),3000m以上为B组(n=81),再将两组患者按冠脉狭窄程度细分为50%~60%亚组(n=84)和61%~70%亚组(n=80)。将患者冠状动脉CT血管成像数据导入AI辅助诊断及CT-FFR测量系统,以冠脉造影及冠脉传统血流储备分数(FFR)为金标准,分别评价AI及CT-FFR在高海拔地区冠脉临界病变诊断中的应用。结果以FFR为金标准,CT-FFR与FFR的一致性为83.75%。B组钙化斑块、易损斑块高于A组(P=0.037、0.020);B组冠状动脉多支病变、61%~70%狭窄程度发生率均高于A组(P<0.05);A组、B组在61%~70%亚组钙化斑块、易损斑块发生率均高于50%~60%亚组(P<0.05)。B组CT-FFR值低于A组(0.76±0.04 vs 0.88±0.05,P<0.01);A、B两组在61%~70%亚组CT-FFR值≤0.80、<0.70的发生率高于50%~60%亚组(P<0.05)。结论AI诊断系统及CT-FFR对评估高海拔地区冠状动脉临界病变的结构特征及血流动力学改变的结果与冠脉造影、FFR一致性高,具有较高的诊断敏感度和特异度。