Radiomics Analysis to Predict Lymphovascular Invasion of Gastric Cancer Based on Iodine-Based Material Decomposition Images and Virtual Monoenergetic Images.

OBJECTIVE: This study aimed to investigate the utility of virtual monoenergetic images (VMIs) and iodine-based material decomposition images (IMDIs) in the assessment of lymphovascular invasion (LVI) in gastric cancer (GC) patients. METHODS: A total of 103 GC patients who underwent dual-energy spectral computed tomography preoperatively were enrolled. The LVI status was confirmed by pathological analysis. The radiomics features obtained from the 70 keV VMI and IMDI were used to build radiomics models. Independent clinical factors for LVI were identified and used to build the clinical model. Then, combined models were constructed by fusing clinical factors and radiomics signatures. The predictive performance of these models was evaluated. RESULTS: The computed tomography-reported N stage was an independent predictor of LVI, and the areas under the curve (AUCs) of the clinical model in the training group and testing group were 0.750 and 0.765, respectively. The radiomics models using the VMI signature and IMDI signature and combining these 2 signatures outperformed the clinical model, with AUCs of 0.835, 0.855, and 0.924 in the training set and 0.838, 0.825, and 0.899 in the testing set, respectively. The model combined with the computed tomography-reported N stage and the 2 radiomics signatures achieved the best performance in the training (AUC, 0.925) and testing (AUC, 0.961) sets, with a good degree of calibration and clinical utility for LVI prediction. CONCLUSIONS: The preoperative assessment of LVI in GC is improved by radiomics features based on VMI and IMDI. The combination of clinical, VMI-, and IMDI-based radiomics features effectively predicts LVI and provides support for clinical treatment decisions.

Gd-EOB-DTPA-enhanced MRI radiomics to predict vessels encapsulating tumor clusters (VETC) and patient prognosis in hepatocellular carcinoma.

OBJECTIVES: The study was to develop a Gd-EOB-DTPA-enhanced MRI radiomics model for preoperative prediction of VETC and patient prognosis in hepatocellular cancer (HCC). METHODS: The study included 182 (training cohort: 128; validation cohort: 54) HCC patients who underwent preoperative Gd-EOB-DTPA-enhanced MRI. Volumes of interest including intratumoral and peritumoral regions were manually delineated in the hepatobiliary phase images, from which 1316 radiomics features were extracted. The least absolute shrinkage and selection operator (LASSO) and multivariable logistic regression were used to select the useful features. Clinical, intratumoral, peritumoral, combined radiomics, and clinical radiomics models were established using machine learning algorithms. The Kaplan-Meier survival analysis was used to assess early recurrence and progression-free survival (PFS) in the VETC + and VETC- patients. RESULTS: In the validation cohort, the area under the curves (AUCs) of radiomics models were higher than that of the clinical model using random forest (all p < 0.05). The peritumoral radiomics model (AUC = 0.972;95% confidence interval [CI]:0.887-0.998) had significantly higher AUC than intratumoral model (AUC = 0.919; 95% CI: 0.811-0.976) (p = 0.044). There were no significant differences in AUC between intratumoral or peritumoral radiomics model (PR) and combined radiomics model (p > 0.05). Early recurrence and PFS were significantly different between the PR-predicted VETC + and VETC- HCC patients (p < 0.05). PR-predicted VETC was independent predictor of early recurrence (hazard ratio [HR]: 2.08[1.31-3.28]; p = 0.002) and PFS (HR: 1.95[1.20-3.17]; p = 0.007). CONCLUSIONS: The intratumoral or peritumoral radiomics model may be useful in predicting VETC and patient prognosis preoperatively. The peritumoral radiomics model may yield an incremental value over intratumoral model. KEY POINTS: • Radiomics models are useful for predicting vessels encapsulating tumor clusters (VETC) and patient prognosis preoperatively. • Peritumoral radiomics model may yield an incremental value over intratumoral model in prediction of VETC. • Peritumoral radiomics-model-predicted VETC was an independent predictor of early recurrence and progression-free survival.

The added value of radiomics from dual-energy spectral CT derived iodine-based material decomposition images in predicting histological grade of gastric cancer.

BACKGROUND: The histological differentiation grades of gastric cancer (GC) are closely related to treatment choices and prognostic evaluation. Radiomics from dual-energy spectral CT (DESCT) derived iodine-based material decomposition (IMD) images may have the potential to reflect histological grades. METHODS: A total of 103 patients with pathologically proven GC (low-grade in 40 patients and high-grade in 63 patients) who underwent preoperative DESCT were enrolled in our study. Radiomic features were extracted from conventional polychromatic (CP) images and IMD images, respectively. Three radiomic predictive models (model-CP, model-IMD, and model-CP-IMD) based on solely CP selected features, IMD selected features and CP coupled with IMD selected features were constructed. The clinicopathological data of the enrolled patients were analyzed. Then, we built a combined model (model-Combine) developed with CP-IMD and clinical features. The performance of these models was evaluated and compared. RESULTS: Model-CP-IMD achieved better AUC results than both model-CP and model-IMD in both cohorts. Model-Combine, which combined CP-IMD radiomic features, pT stage, and pN stage, yielded the highest AUC values of 0.910 and 0.912 in the training and testing cohorts, respectively. Model-CP-IMD and model-Combine outperformed model-CP according to decision curve analysis. CONCLUSION: DESCT-based radiomics models showed reliable diagnostic performance in predicting GC histologic differentiation grade. The radiomic features extracted from IMD images showed great promise in terms of enhancing diagnostic performance.

Pioneering Iodine-125-Labeled Nanoscale Covalent Organic Frameworks for Brachytherapy.

Brachytherapy has been clinically used for the treatment of malignant solid tumors. However, the classic therapeutic radioactive 125I seed must be surgically implanted directly into tumors. To avoid the surgery and prevent irrational radioactive distribution, radioiodine-loaded nanomaterials are ever-developing for brachytherapy. Hence, it is still a notable challenge to obtain an advanced material that simultaneously incorporates features of high radiolabeling rate, short labeling time, good radiolabeling stability, and long tumor retention time. Covalent organic frameworks (COFs), which are crystalline polymers with ordered pores, are widely applied in guest delivery of drugs based on their high porosity and modifiable skeleton. Herein, we developed a functionalized nanoscale PEG-COF-Ag material, which could rapidly capture radioiodine reaching a 94% radiolabeling yield in 30 s. In addition, more than 95% 125I was maintained after 24 h in PBS (phosphate-buffered saline) as well as in serum and over 90% for nearly 1 week. PEG-COF-Ag-125I (125I-COF) demonstrated excellent cancer cell killing performance in vitro, and further experiments in vivo revealed a long tumor retention time and effective tumor treatment during the radiotherapy. The results indicate that radioiodine-labeled PEG-COF-Ag could be potentially applied in brachytherapy with a promising therapeutic effect.

In Vivo Uranium Sequestration using a Nanoscale Metal-Organic Framework.

An agent for actinide sequestration with fast uranium uptake kinetics and efficient in vivo uranium removal using a nanoscale metal-organic framework (nano-MOF) is proposed. UiO-66 nanoparticles post-synthetically functionalized with carboxyl groups, UiO-66-(COOH)4 -180, exhibit the fastest uranium uptake kinetics reported with more than 65 % of uranyl in fetal bovine serum (FBS) removed within 5 min. Moreover, the in vivo bio-distribution studies show that the material partially accumulates in kidneys and femurs where uranium mainly deposits facilitating the in vivo sequestration of uranium. The results of the in vivo uranium decorporation assays with mice show that UiO-66-(COOH)4 -180 could successfully reduce the amounts of uranyl deposited in kidneys and femurs by up to 55.4 % and 36.5 %, respectively, and is significantly more efficient than the commercial actinide decorporation agent, ZnNa3 -DTPA.

3-Hydroxy-2-Pyrrolidinone as a Potential Bidentate Ligand for in Vivo Chelation of Uranyl with Low Cytotoxicity and Moderate Decorporation Efficacy: A Solution Thermodynamics, Structural Chemistry, and in Vivo Uranyl Removal Survey.

Uranium poses a threat for severe renal and bone damage in vivo. With the rapid development of nuclear industry, it is more urgent than ever to search for potential in vivo uranium chelators. In this work, 3-hydroxy-2-pyrrolidinone (HPD) is investigated as a new potential uranium decorporation ligand. The potentiometric titration measurements were carried out, and the stability constants were determined to be log ß110 = 10.5(7), log ß120 = 20.7(9), and log ß130 = 28.2(4). The species distribution diagram shows that nearly all uranyl is complexed by HPD at pH 7.4 under the defined condition. A single crystal of uranyl and HPD complexes, [(UO2)3O(H2O)3(C4H6NO2)3]·NO3·12H2O (uranyl-HPD), was obtained via an evaporation method. The overall structure of uranyl-HPD is a trimer that consists of three uranyl units and three HPD ligands. The uranyl unit is equatorially coordinated by three oxygen atoms from two HPD agents, one coordinated water molecule, and one µ3-O atom that is shared by three uranyl units. The results of the cytotoxicity assay indicate that the ligand is less toxic than the chelators used clinically (i.e., DTPA-ZnNa3 and 3-hydroxy-1,2-dimethyl-4(1 H)-pyridone (DFP)). The results of the uranium removal assay using the NRK-52E cell show that it could reduce as much as 58% of the uranium content at the cellular level. Furthermore, the in vivo uranium decorporation assays demonstrate that HPD can remove 52% of uranium deposited in the kidney but shows poor uranium removal efficacy in the bone.

Spectral Computed Tomography Imaging in the Differential Diagnosis of Lung Cancer and Inflammatory Myofibroblastic Tumor.

OBJECTIVE: The aim of this study was to explore the value of spectral computed tomography (CT) imaging in differentiating lung cancer from inflammatory myofibroblastic tumor (IMT). METHODS: One hundred twelve patients with 96 lung cancers and 16 IMTs underwent spectral CT during arterial phase (AP) and venous phase (VP). The normalized iodine concentration in AP (NICAP) and VP (NICVP), slope of spectral Hounsfield unit curve in AP (λAP) and VP (λVP), and normalized iodine concentration difference between AP and VP (ICD) were calculated. The 2-sample t test compared quantitative parameters. Two readers qualitatively assessed lesion types according to imaging features. Receiver operating characteristic curves were generated to calculate sensitivity and specificity. Sensitivity and specificity of the qualitative and quantitative studies were compared. RESULTS: The patients with IMT had significantly higher NICAP, NICVP, λAP, λVP, and ICD than did the patients with lung cancer (P < 0.05). The threshold NICVP of 0.425 would yield the highest sensitivity and specificity of 92.7% and 81.3%, respectively, for differentiating lung cancer from IMT. The logistic regression model produced from combining quantitative parameters NICAP, NICVP, λAP, and λVP provided a sensitivity and specificity of 100% and 81.3%, respectively, for differentiating lung cancer from IMT. CONCLUSIONS: Spectral CT imaging with the quantitative analysis may help to increase the accuracy of differentiating lung cancer from IMT.

3,2-Hydroxypyridinone-Grafted Chitosan Oligosaccharide Nanoparticles as Efficient Decorporation Agents for Simultaneous Removal of Uranium and Radiation-Induced Reactive Oxygen Species in Vivo.

Most of the key radionuclides in the nuclear fuel cycle, such as actinides, possess a combination of heavy metal chemotoxicity and radiotoxicity and therefore represent a severe threat to the ecological environment and public safety. The radiotoxicity originates from direct radiation-induced organ damage and indirect damage, mostly through radiation-induced reactive oxygen species (ROS). Although effective chelating agents that can accelerate the excretion of actinides, such as uranium, have been developed in the past several decades, very few of them can reduce radiation-induced damage from internal contamination. In fact, the strategy of simultaneous removal of actinides and their induced-ROS in vivo has scarcely been considered. Here, we report a 3,2-hydroxypyridinone-grafted chitosan oligosaccharide nanoparticle (COS-HOPO) as a new type of decorporation agent that is effective for the removal of both uranium and ROS in vivo. The cytotoxicity and decorporation assays indicate that the marriage of chitosan oligosaccharide (COS) and hydroxypyridinone (HOPO) gives rise to a remarkable decrease in toxicity and promotion of the uranium removal capability from both kidneys and femurs. The decorporation efficacy can reach up to 43% in rat proximal tubular epithelial cells (NRK-52E), 44% in kidneys, and 32% in femurs. Moreover, the ROS levels of the cells treated with COS-HOPO are significantly lower than those of the control group, implying a promising radiation protection effect. The detoxification mechanism of COS-HOPO is closely related to both chelating U(VI)- and scavenging U(VI)-induced intracellular ROS.

Decreased stage migration rate of early gastric cancer with a new reconstruction algorithm using dual-energy CT images: a preliminary study.

OBJECTIVES: To evaluate the potential value of advanced monoenergetic images (AMEIs) on early gastric cancer (EGC) using dual-energy CT (DECT). METHODS: 31 EGC patients (19 men, 12 women; age range, 38-81 years; mean age, 57.19 years) were retrospectively enrolled in this study. Conventionally reconstructed polyenergetic images (PEIs) at 120 kV and virtual monoenergetic images (MEIs) and AMEIs at six different kiloelectron volt (keV) levels (from 40 to 90 keV) were evaluated from the 100 and Sn 140 kV dual energy image data, respectively. The visibility and stage migration of EGC for all three image data sets were evaluated and statistically analyzed. The objective and subjective image qualities were also evaluated. RESULTS: AMEIs at 40 keV showed the best visibility (80.7 %) and the lowest stage migration (35.5 %) for EGC. The stage migration for AMEIs at 40 keV was significantly lower than that for PEIs (p = 0.026). AMEIs at 40 keV had statistically higher CNR in the arterial and portal phases, gastric-specific diagnostic performance and visual sharpness compared with other AMEIs, MEIs and PEIs (all p < 0.05). CONCLUSIONS: AMEIs at 40 keV with MPR increase the CNR of EGC and thus potentially lower the stage migration of EGC. KEY POINTS: • AMEIs benefits from the recombination of low-keV images and medium energies. • AMEIs could receive better CNR results than MEIs. • AMEIs at 40 keV potentially lower the stage migration of EGC.

Gastric Cancer: Preoperative TNM Staging With Individually Adjusted Computed Tomography Scanning Phase.

OBJECTIVE: The aim of this study was to evaluate test bolus scan technology on preoperative diagnostic performance, vascular enhancement, and artery visualization for gastric cancer. METHODS: The institutional review board approved this study. Fifty-four patients in protocol 1 were resigned to a fixed delay time scan method, and their images were obtained in the late arterial phase (AP) and portal venous phase (PP), with start delays of 40 and 70 seconds, respectively. Fifty-six patients in protocol 2 had undergone the test bolus method first and received the time to peak enhancement of the aorta. Their images were obtained in the AP and PP with start delays in the time to peak enhancement and 20 seconds after the AP, respectively. Two radiologists performed consensus interpretation of the preoperative TNM staging, vascular enhancement, tumor contrast-to-noise ratio (CNR) and artery visualization between the 2 protocols. RESULTS: There is no significant difference in the T, N, and M staging diagnostic accuracy between the protocols (P = 0.41, P > 0.99, and P = 0.34, respectively). For serosa-negative (T1, T2, and T3) tumors, the diagnostic accuracy obtained with protocol 2 was superior to that obtained with protocol 1 (P = 0.04). Protocol 2 was superior for perigastric vessel enhancement (left gastric artery, right gastroepiploic artery, and splenic artery; P < 0.001, P < 0.001, and P = 0.001, respectively). The stomach-to-tumor CNR during the PP of protocol 2 was significantly higher than that during either the AP or PP of protocol 1 (P = 0.004 and P = 0.001, respectively). The mean rankings of the artery visualization were significantly higher with protocol 2 than with protocol 1 (P < 0.001). CONCLUSIONS: The dual-phase scan with test bolus technology could improve the tumor CNR and had high staging accuracy for serosa-negative tumors as well as high perigastric artery enhancement, yielding satisfactory artery visualization for diagnosis.

Multiple instance learning for computer aided detection and diagnosis of gastric cancer with dual-energy CT imaging.

Multiple instance learning algorithms have been increasingly utilized in computer aided detection and diagnosis field. In this study, we propose a novel multiple instance learning method for the identification of tumor invasion depth of gastric cancer with dual-energy CT imaging. In the proposed scheme, two level features, bag-level features and instance-level features are extracted for subsequent processing and classification work. For instance-level features, there is some ambiguity in assigning labels to selected patches. An improved Citation-KNN method is presented to solve this problem. Compared with benchmarking state-of-the-art multiple instance learning algorithms using the same clinical dataset, the proposed algorithm can achieve improved results. The experimental evaluation is performed using leave-one-out cross validation with the total accuracy of 0.7692. The proposed multiple instance learning algorithm serves as an alternative method for computer aided diagnosis and identification of tumor invasion depth of gastric cancer with dual-energy CT imaging techniques.

Determination of 90Sr and 210Pb in food samples by liquid scintillation counting after sequential separation with an extraction chromatographic column.

90Sr and 210Pb are considered to be key radionuclides in internal exposure resulting from dietary intake, however, the established methods employed for their detection are time-comsuming. A method for the sequential separation of 90Sr and 210Pb using a Sr·spec resin by LSC measurement is developed, which is highly suitable for food safety monitoring as its minimal sample requirements. The sequential separation of Sr and Pb from the sample was using 0.05 mol/L HNO3 and 0.05 mol/L C6H5O7(NH4)3. The chemical recoveries of Sr and Pb measured using ICP-OES were 72-83% and 80-88%, respectively. The minimum detectable activities of 90Sr and 210Pb in the food sample were 36.2 mBq/kg and 28.6 mBq/kg, respectively, obtained from a 0.1 kg fresh sample and 300 min counting time. The method was validated using reference materials and compared with other methods. The feasibility of the developed method for other highly complex food matrices needs further investigation.

Effective mitigation of gadolinium deposition using the bidentate hydroxypyridinone ligand Me-3,2-HOPO.

With the extensive usage of gadolinium-based contrast agents (GBCAs) in magnetic resonance imaging (MRI), gadolinium deposition has been observed in the brain, kidneys, liver, etc., and this is also closely related to the development of nephrogenic systemic fibrosis (NSF) in patients with renal dysfunction. Chelation, thereby promoting the elimination of deposited Gd(III), seems to be promising for alleviating these problems. Despite many ligands suitable for chelation therapy having been studied, the decorporation of transition metals (e.g. iron, copper, lead, etc.) and actinides (e.g. uranium, plutonium, etc.) has long been a primary concern, whereas the study of Gd(III) has been extremely limited. Due to their excellent metal binding abilities in vivo and therapeutic effects toward neurodegenerative diseases, bidentate hydroxypyridinone ligands are expected to be able to remove Gd(III) from the brain, kidneys, bones, and liver. Herein, the Gd(III) decorporation efficacy of a bidentate hydroxypyridinone ligand (Me-3,2-HOPO) has been evaluated. The complexation behavior between Me-3,2-HOPO and Gd(III) in solution and solid states was characterized with the assistance of potentiometric titration and X-ray diffraction techniques, respectively. Solution-based thermodynamic studies illustrate that the dominant species of complex between Gd(III) and Me-3,2-HOPO (HL) is GdL2+ (log ß120 = 11.8 (3)) at pH 7.4. The structure of the Gd-Me-3,2-HOPO crystal obtained from a room temperature reaction reveals the formation of a Gd(III) dimer that is chelated by four ligands as a result of metal ion hydration and ligand complexation. Cellular Gd(III) removal assays illustrate that Me-3,2-HOPO could effectively reduce final amounts of gadolinium by 77.6% and 66.1% from rat renal proximal tubular epithelial (NRK-52E) cells and alpha mouse liver 12 (AML-12) cells, respectively. Our current results suggest the potential of bidentate HOPO ligands as an effective approach to treat patients suffering from Gd(III) toxicity.

Boosting Simultaneous Uranium Decorporation and Reactive Oxygen Species Scavenging Efficiency by Lacunary Polyoxometalates.

The chemical toxicity and the oxidative stress induced by the internal exposure of uranium is responsible for the long-term adverse effect of in vivo contamination of uranium. An agent with simultaneous removal capability of uranium and excess reactive oxygen species (ROS) is highly desired. Herein, the lacunary Keggin-type polyoxometalate (POM) is demonstrated to selectively bind with uranyl ions in the presence of excess essential divalent ions and exhibits a compelling ROS scavenging efficiency of 78.8%. In vivo uranium decorporation assays illustrate the uranium sequestration efficiencies of 74.0%, 49.4%, and 37.1% from kidneys by prophylactic, prompt, and delayed administration of lacunary POM solution, respectively. The superior ROS quenching and uranium removal performance in comparison with all reported bifunctional agents endow lacunary polyoxometalates as novel agents to effectively protect people from injuries caused by the internal exposure of actinides.

Resting-State Functional MRI Study: Connection Strength of Brain Networks in DR Patients.

PURPOSE: To explore the functional connection strength (FCS) changes of brain networks in diabetic retinopathy (DR) patients and uncover the underlying mechanism. METHODS AND MATERIALS: Twenty-one patients with DR and 21 age- and sex-matched healthy controls were enrolled from August 2012 to September 2014. Subjects were scanned using 3T MR with blood-oxygen-level dependent (BOLD) and 3-dimension fast spoiled gradient echo (3D-FSPGR) sequences. MR data was analyzed via preprocessing and functional network construction. After a group comparison, components of brain networks with significant group differences were extracted and the FCS of the brain network was evaluated. The brain areas were compared between patients and controls. P-values less than 0.05 were considered statistically significant. Connection strength was evaluated with alphasim, P<0.01. RESULTS: The component maps of altered brain networks with quantified FCS were obtained in DR patients, demonstrating more disconnections mainly in the bilateral Heschl's gyrus, left cuneus, left occipital lobe, bilateral amygdala, left parahippocampal, bilateral fusiform, and left superior parietal in the patients group compared to the healthy controls (P<0.01), while compensations may occur in the frontal-cingulum region, as well as among the right caudate, left thalamus, left inferior temporal lobe, and middle orbital frontal lobe. CONCLUSION: Brain network connections, decreased in the brain areas of which in charging with cognition and visual function, suggests that DR patients might have cognitive decline and visual function loss. However, there might be a frontal compensatory circle in patients with DR.

A 3,2-Hydroxypyridinone-based Decorporation Agent that Removes Uranium from Bones In Vivo.

Searching for actinide decorporation agents with advantages of high decorporation efficiency, minimal biological toxicity, and high oral efficiency is crucial for nuclear safety and the sustainable development of nuclear energy. Removing actinides deposited in bones after intake is one of the most significant challenges remaining in this field because of the instantaneous formation of highly stable actinide phosphate complexes upon contact with hydroxyapatite. Here we report a hydroxypyridinone-based ligand (5LIO-1-Cm-3,2-HOPO) exhibiting stronger affinity for U(VI) compared with the reported tetradentate hydroxypyridinone ligands. This is further revealed by the first principles calculation analysis on bonding between the ligand and uranium. Both in vitro uranium removal assay and in vivo decorporation experiments with mice show that 5LIO-1-Cm-3,2-HOPO can remove uranium from kidneys and bones with high efficiencies, while the decorporation efficiency is nearly independent of the treatment time. Moreover, this ligand shows a high oral decorporation efficiency, making it attractive for practical applications.

Structural and thermodynamic stability of uranyl-deferiprone complexes and the removal efficacy of U(vi) at the cellular level.

Deferiprone (3-hydroxy-1,2-dimethyl-4(1H)-pyridone, DFP), which is a drug clinically used for removing heavy metals in vivo, was explored for its removal efficiency towards uranium. The reaction of uranyl nitrate hexahydrate with DFP at room temperature yielded the compound [(UO2)(H2O)(C7NO2H8)2]·4H2O (1), which crystallizes from a mixed solution of methanol and water (pH = 7.0). X-ray diffraction shows that the stable complexation of uranyl occurs from the coordination of two bidentate DFP ligands perpendicular to the O[double bond, length as m-dash]U[double bond, length as m-dash]O unit with a fifth coordinating oxygen atom coming from one water molecule, resulting in a pentagonal bipyramidal geometry. The formation constants of uranyl and DFP complexes were measured and the species distribution diagram illustrates that UO2L2 (94.6%) is the dominant uranyl-DFP complex in 0.1 M KCl solution at physiological pH = 7.4. The results from both crystallographic and potentiometric studies imply that the metal : ligand ratio is 1 : 2. The effectiveness of using DFP to remove uranium was examined at the cellular level, and the results suggest that it can significantly reduce the cellular uptake and increase the cellular release of U(vi) in renal proximal tubular epithelial cells (NRK-52E).

Computer-aided diagnosis for preoperative invasion depth of gastric cancer with dual-energy spectral CT imaging.

RATIONALE AND OBJECTIVES: This study evaluates the accuracy of dual-energy spectral computed tomography (DEsCT) imaging with the aid of computer-aided diagnosis (CAD) system in assessing serosal invasion in patients with gastric cancer. MATERIALS AND METHODS: Thirty patients with gastric cancer were enrolled in this study. Two types of features (information) were collected with the use of DEsCT imaging: conventional features including patient's clinical information (eg, age, gender) and descriptive characteristics on the CT images (eg, location of the lesion, wall thickness at the gastric cardia) and additional spectral CT features extracted from monochromatic images (eg, 60 keV) and material-decomposition images (eg, iodine- and water-density images). The classification results of the CAD system were compared to pathologic findings. Important features can be found out using support vector machine classification method in combination with feature-selection technique thereby helping the radiologists diagnose better. RESULTS: Statistical analysis showed that for the collected cases, the feature "long axis" was significantly different between group A (serosa negative) and group B (serosa positive) (P < .05). By adding quantitative spectral features from several regions of interest (ROIs), the total classification accuracy was improved from 83.33% to 90.00%. Two feature ranking algorithms were used in the CAD scheme to derive the top-ranked features. The results demonstrated that low single-energy (approximately 60 keV) CT values, tumor size (long axis and short axis), iodine (water) density, and Effective-Z values of ROIs were important for classification. These findings concurred with the experience of the radiologist. CONCLUSIONS: The CAD system designed using machine-learning algorithms may be used to improve the identification accuracy in the assessment of serosal invasion in patients of gastric cancer with DEsCT imaging and provide some indicators which may be useful in predicting prognosis.

Evaluation of dual energy spectral CT in differentiating metastatic from non-metastatic lymph nodes in rectal cancer: Initial experience.

OBJECTIVES: To investigate the value of dual energy spectral CT (DEsCT) imaging in differentiating metastatic from non-metastatic lymph nodes in rectal cancer. METHODS: Fifty-five patients with rectal cancer underwent the arterial phase (AP) and portal venous phase (PP) contrast-enhanced DEsCT imaging. The virtual monochromatic images and iodine-based material decomposition images derived from DEsCT imaging were interpreted for lymph nodes (LNs) measurement. The short axis diameter and the normalized iodine concentration (nIC) of metastatic and non-metastatic LNs were measured. The two-sample t test was used to compare the short axis diameters and nIC values of metastatic and non-metastatic LNs. ROC analysis was performed to assess the diagnostic performance. RESULTS: One hundred and fifty two LNs including 92 non-metastatic LNs and 60 metastatic LNs were matched using the radiological-pathological correlation. The mean short axis diameter of metastatic LNs was significantly larger than that of the non-metastatic LNs (7.28±2.28mm vs. 4.90±1.64mm, P<0.001). The mean nIC value for metastatic LNs was significantly lower than that of non-metastatic LNs (0.24±0.08 vs. 0.34±0.21, P=0.001 in AP; 0.47±0.18 vs. 0.64±0.17, P<0.001 in PP). Combining nIC (PP) with the short axis diameter, the overall accuracy could be improved to 82.9%. CONCLUSIONS: With the combination of nIC value in PP and conventional size criterion, dual energy spectral imaging may be used to differentiate metastatic from non-metastatic lymph nodes in rectal cancer.