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

MRI versus Dual-Energy CT in Local-Regional Staging of Gastric Cancer

Background Preoperative local-regional tumor staging of gastric cancer (GC) is critical for appropriate treatment planning. The comparative accuracy of multiparametric MRI (mpMRI) versus dual-energy CT (DECT) for staging of GC is not known. Purpose To compare the diagnostic accuracy of personalized mpMRI with that of DECT for local-regional T and N staging in patients with GC receiving curative surgical intervention. Materials and Methods Patients with GC who underwent gastric mpMRI and DECT before gastrectomy with lymphadenectomy were eligible for this single-center prospective noninferiority study between November 2021 and September 2022. mpMRI comprised T2-weighted imaging, multiorientational zoomed diffusion-weighted imaging, and extradimensional volumetric interpolated breath-hold examination dynamic contrast-enhanced imaging. Dual-phase DECT images were reconstructed at 40 keV and standard 120 kVp-like images. Using gastrectomy specimens as the reference standard, the diagnostic accuracy of mpMRI and DECT for T and N staging was compared by six radiologists in a pairwise blinded manner. Interreader agreement was assessed using the weighted κ and Kendall W statistics. The McNemar test was used for head-to-head accuracy comparisons between DECT and mpMRI. Results This study included 202 participants (mean age, 62 years ± 11 [SD]; 145 male). The interreader agreement of the six readers for T and N staging of GC was excellent for both mpMRI (κ = 0.89 and 0.85, respectively) and DECT (κ = 0.86 and 0.84, respectively). Regardless of reader experience, higher accuracy was achieved with mpMRI than with DECT for both T (61%-77% vs 50%-64%; all P < .05) and N (54%-68% vs 51%-58%; P = .497-.005) staging, specifically T1 (83% vs 65%) and T4a (78% vs 68%) tumors and N1 (41% vs 24%) and N3 (64% vs 45%) nodules (all P < .05). Conclusion Personalized mpMRI was superior in T staging and noninferior or superior in N staging compared with DECT for patients with GC. Clinical trial registration no. NCT05508126 © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Méndez and Martín-Garre in this issue.

Artificial intelligence-based tools with automated segmentation and measurement on CT images to assist accurate and fast diagnosis in acute pancreatitis

OBJECTIVES

To develop an artificial intelligence (AI) tool with automated pancreas segmentation and measurement of pancreatic morphological information on CT images to assist improved and faster diagnosis in acute pancreatitis.

METHODS

This study retrospectively contained 1124 patients suspected for AP and received non-contrast and enhanced abdominal CT examination between September 2013 and September 2022. Patients were divided into training (N = 688), validation (N = 145), testing dataset [N = 291; N = 104 for normal pancreas, N = 98 for AP, N = 89 for AP complicated with PDAC (AP&PDAC)]. A model based on convolutional neural network (MSAnet) was developed. The pancreas segmentation and measurement were performed via eight open-source models and MSAnet based tools, and the efficacy was evaluated using dice similarity coefficient (DSC) and intersection over union (IoU). The DSC and IoU for patients with different ages were also compared. The outline of tumour and oedema in the AP and were segmented by clustering. The diagnostic efficacy for radiologists with or without the assistance of MSAnet tool in AP and AP&PDAC was evaluated using receiver operation curve and confusion matrix.

RESULTS

Among all models, MSAnet based tool showed best performance on the training and validation dataset, and had high efficacy on testing dataset. The performance was age-affected. With assistance of the AI tool, the diagnosis time was significantly shortened by 26.8% and 32.7% for junior and senior radiologists, respectively. The area under curve (AUC) in diagnosis of AP was improved from 0.91 to 0.96 for junior radiologist and 0.98 to 0.99 for senior radiologist. In AP&PDAC diagnosis, AUC was increased from 0.85 to 0.92 for junior and 0.97 to 0.99 for senior.

CONCLUSION

MSAnet based tools showed good pancreas segmentation and measurement performance, which help radiologists improve diagnosis efficacy and workflow in both AP and AP with PDAC conditions.

ADVANCES IN KNOWLEDGE

This study developed an AI tool with automated pancreas segmentation and measurement and provided evidence for AI tool assistance in improving the workflow and accuracy of AP diagnosis.

Diagnostic accuracy of radiomics-based machine learning for neoadjuvant chemotherapy response and survival prediction in gastric cancer patients: A systematic review and meta-analysis

BACKGROUND

In recent years, researchers have explored the use of radiomics to predict neoadjuvant chemotherapy outcomes in gastric cancer (GC). Yet, a lingering debate persists regarding the accuracy of these predictions. Against this backdrop, this study was conducted to examine the accuracy of radiomics in predicting the response to neoadjuvant chemotherapy in GC patients.

METHODS

An exhaustive search of relevant studies was conducted in PubMed, Cochrane, Embase, and Web of Science databases up to February 21, 2023. The radiomics quality scoring (RQS) tool was employed to assess study quality. Tumor response to neoadjuvant chemotherapy and survival outcomes were examined as outcome measures.

RESULTS

Fourteen studies involving 3,373 GC patients who had received neoadjuvant chemotherapy were incorporated in our meta-analysis. The mean RQS score across all studies was 36.3%, ranging between 0 and 63.9%. On the validation cohort, when the modeling variables were restricted to radiomic features alone, the predictive performance was characterized by a c-index of 0.750 (95% CI: 0.710-0.790), with a sensitivity of 0.67 (95% CI: 0.58-0.75) and a specificity of 0.77 (95% CI: 0.69-0.84) for the prediction of neoadjuvant chemotherapy response. When clinical data was integrated with radiomic features as modeling variables, the predictive performance improved, yielding a c-index of 0.814 (95% CI: 0.780-0.847), a sensitivity of 0.78 [95% CI: 0.70-0.84], and a specificity of 0.73 [95% CI: 0.67-0.79].

CONCLUSIONS

Radiomics holds promise to noninvasively predict neoadjuvant chemotherapy response and survival outcomes among patients with locally advanced GC. Additionally, we underscore the need for future multicenter studies and the development of imaging-sourced tools for risk stratification in this patient population.

Predicting pathologic response to neoadjuvant chemotherapy in locally advanced gastric cancer: The establishment of a spectral CT-based nomogram from prospective datasets

BACKGROUND

To establish a spectral CT-based nomogram for predicting early neoadjuvant chemotherapy (NAC) response for locally advanced gastric cancer (LAGC).

METHODS

This study prospectively recruited 222 cases (177 male and 45 female patients, 9.59 ± 9.54 years) receiving NAC and radical gastrectomy. Triple enhanced spectral CT scans were performed before NAC initiation. According to post-operative tumor regression grade (TRG), patients were classified into responders (TRG = 0 + 1) or non-responders (TRG = 2 + 3), and split into a primary (156) and validation (66) dataset at 7:3 ratio chronologically. We compared clinicopathological data, follow-up information, iodine concentration (IC), normalized ICs (nICs) in arterial/venous/delayed phases (AP/VP/DP) between responders and non-responders. Independent risk factors of response were screened by multivariable logistic regression and adopted for model construction. Model was visualized by nomograms and its capability was determined through receiver operating characteristic (ROC) curves. Log-rank survival analysis was conducted to explore associations between TRG, nomogram and patients' survival.

RESULTS

This work identified Borrmann classification, ICDP, and nICDP were independent risk factors of response outcomes. A spectral CT-based nomogram was built accordingly and achieved an area under the curve (AUC) of 0.797 (0.692-0.879) and 0.741(0.661-0.811) for the primary and validation dataset, respectively, higher than AUC of individual parameters alone. The nomogram was related to disease-free survival in the validation dataset (Hazard ratio (HR): 5.19 [1.18-12.93], P = 0.02).

CONCLUSIONS

The spectral CT-based nomogram provides an efficient tool for predicting the pathologic response outcomes of GC after NAC and disease-free survival risk stratification.

Evaluation of optimal monoenergetic images acquired by dual-energy CT in the diagnosis of T staging of thoracic esophageal cancer

OBJECTIVES

The purpose of our study was to objectively and subjectively assess optimal monoenergetic image (MEI (+)) characteristics from dual-energy CT (DECT) and the diagnostic performance for the T staging in patients with thoracic esophageal cancer (EC).

METHODS

In this retrospective study, patients with histopathologically confirmed EC who underwent DECT from September 2019 to December 2020 were enrolled. One standard polyenergetic image (PEI) and five MEI (+) were reconstructed. Two readers independently assessed the lesion conspicuity subjectively and calculated the contrast-to-noise ratio (CNR) and the signal-to-noise ratio (SNR) of EC. Two readers independently assessed the T stage on the optimal MEI (+) and PEI subjectively. Multiple quantitative parameters were measured to assess the diagnostic performance to identify T1-2 from T3-4 in EC patients.

RESULTS

The study included 68 patients. Subjectively, primary tumor delineation received the highest ratings in MEI (+) _40 keV of the venous phase. Objectively, MEI (+) images showed significantly higher SNR compared with PEI (p < 0.05), peaking at MEI (+) _40 keV in the venous phase. CNR of tumor (MEI (+) _{40 keV -80 keV}) was all significantly higher than PEI in arterial and venous phases (p < 0.05), peaking at MEI (+) _40 keV in venous phases. The agreement between MEI (+) _40 keV and pathologic T categories was 81.63% (40/49). Rho values in venous phases had excellent diagnostic efficiency for identifying T1-2 from T3-4 (AUC = 0.84).

CONCLUSIONS

MEI (+) reconstructions at low keV in the venous phase improved the assessment of lesion conspicuity and also have great potential for preoperative assessment of T staging in patients with EC.

Dual energy CT applications in oncological imaging

Cancer is the second leading cause of death in the United States, killing more than 600.000 people each year.¹ Despite several screening programs available, cancer diagnosis is often made incidentally during imaging studies performed for other reasons. Once the diagnosis is made, treatment assessment and surveillance of these patients heavily rely on radiological tools. Computed tomography (CT) in particular is one of the most commonly ordered modalities due to wide availability even in the most remote locations, and fast results. However, conventional CT often cannot definitively characterize a neoplastic lesion unless it was tailored toward answering a specific question. Furthermore, characterizing small lesions can be difficult with CT. An innovative technique called dual-energy CT (DECT) offers solutions to some of the challenges of conventional CT in oncological imaging.

Noise-optimized virtual monoenergetic imaging technology of the third-generation dual-source computed tomography and its clinical applications

The third-generation dual-source computed tomography (DSCT) is among the most advanced imaging methods. It employs noise-optimized virtual monoenergetic imaging (VMI+) technology. It uses the frequency-split method to extract high-contrast image information from low-energy images and low-noise information from images reconstructed at an optimal energy level, combining them to obtain the final image with improved quality. This review is the first to summarize the results of clinical studies that primarily and recently evaluated the VMI+ technique based on tumor, blood vessel, and other lesion classification. We aim to assist radiologists in quickly selecting the appropriate energy level when performing image reconstruction for superior image quality in clinical work and providing several ideas for future scientific research of the VMI+ technique. Presently, VMI+ reconstruction is mostly used for images of various tumors or blood vessels, including coronary plaques, coronary stents, deep vein thromboses, pulmonary embolisms (PEs), active arterial hemorrhages, and endoleaks after endovascular aneurysm repair. In addition, VMI+ has been used for imaging children's heads, liver lesions, pancreatic lacerations, and reducing metal artifacts. Regarding the reconstruction at the optimal energy level, the VMI+ technique yielded a higher image quality than the pre-optimized virtual monoenergetic imaging (VMI) technique and single-energy CT. Moreover, either low concentrations of contrast medium or low iodine injection rates can be applied before VMI+ reconstruction at a low-energy level to reduce contrast agent-related kidney injury risk. After reconstructing an image at the optimal energy level, both the image's window width and level can also be adjusted to improve the image effect's reach and diagnosis suitability. To improve image quality and lesion-imaging clarity and reduce the use of contrast agents, VMI+ reconstruction technology has been applied clinically, in which the selection of energy level is the key to the whole reconstruction process. Our review summarizes these optimal levels for radiologists' reference and suggests new ideas for the direction of future VMI+ research.

Evaluation of dual-energy CT derived radiomics signatures in predicting outcomes in patients with advanced gastric cancer after neoadjuvant chemotherapy

BACKGROUND

To investigate the prognostic value of dual-energy CT (DECT) based radiomics to predict disease-free survival (DFS) and overall survival (OS) for patients with advanced gastric cancer (AGC) after neoadjuvant chemotherapy (NAC).

METHODS

From January 2014 to December 2018, a total of 156 AGC patients were enrolled and randomly allocated into a training cohort and a testing cohort at a ratio of 2:1. Volume of interest of primary tumor was delineated on eight image series. Four feature sets derived from pre-NAC and delta radiomics were generated for each survival arm. Random survival forest was used for generating the optimal radiomics signature (RS). Statistical metrics for model evaluation included Harrell's concordance index (C-index) and the average cumulative/dynamic AUC throughout follow-up. A clinical model and a combined Rad-clinical model were built for comparison.

RESULTS

The pre-IU (derived from iodine uptake images before NAC) RS performed best for DFS and OS in the testing cohort (C-indices, 0.784 and 0.698; the average cumulative/dynamic AUCs, 0.80 and 0.77). When compared with the clinical model, the radiomics model had significantly higher C-index to predict DFS in the testing cohort (0.784 vs. 0.635, p < 0.001), but no statistical difference was found for OS (0.698 vs. 0.680, p = 0.473). The combined Rad-clinical models showed improved performance in the testing cohort, with C-indices of 0.810 and 0.710 for DFS and OS, respectively.

CONCLUSION

DECT-derived radiomics serves as a promising non-invasive biomarker to predict survival for AGC patients after NAC, providing an opportunity for transforming proper treatment.

Dual-Energy Computed Tomography-Based Radiomics to Predict Peritoneal Metastasis in Gastric Cancer

Objective

To develop and validate a dual-energy computed tomography (DECT) derived radiomics model to predict peritoneal metastasis (PM) in patients with gastric cancer (GC).

Methods

This retrospective study recruited 239 GC (non-PM = 174, PM = 65) patients with histopathological confirmation for peritoneal status from January 2015 to December 2019. All patients were randomly divided into a training cohort (n = 160) and a testing cohort (n = 79). Standardized iodine-uptake (IU) images and 120-kV-equivalent mixed images (simulating conventional CT images) from portal-venous and delayed phases were used for analysis. Two regions of interest (ROIs) including the peritoneal area and the primary tumor were independently delineated. Subsequently, 1691 and 1226 radiomics features were extracted from the peritoneal area and the primary tumor from IU and mixed images on each phase. Boruta and Spearman correlation analysis were used for feature selection. Three radiomics models were established, including the R_IU model for IU images, the R_MIX model for mixed images and the combined radiomics model (the R_comb model). Random forest was used to tune the optimal radiomics model. The performance of the clinical model and human experts to assess PM was also recorded.

Results

Fourteen and three radiomics features with low redundancy and high importance were extracted from the IU and mixed images, respectively. The R_IU model showed significantly better performance to predict PM than the R_MIX model in the training cohort (AUC, 0.981 vs. 0.917, p = 0.034). No improvement was observed in the R_comb model (AUC = 0.967). The R_IU model was the optimal radiomics model which showed no overfitting in the testing cohort (AUC = 0.967, p = 0.528). The R_IU model demonstrated significantly higher predictive value on peritoneal status than the clinical model and human experts in the testing cohort (AUC, 0.785, p = 0.005; AUC, 0.732, p <0.001, respectively).

Conclusion

DECT derived radiomics could serve as a non-invasive and easy-to-use biomarker to preoperatively predict PM for GC, providing opportunity for those patients to tailor appropriate treatment.

Improved Visualization of Gastric Cancer and Increased Diagnostic Performance in Lesion Depiction and Depth Identification Using Monoenergetic Reconstructions from a Novel Dual-Layer Spectral Detector CT

RATIONALE AND OBJECTIVES

To determine the optimal keV for the visualization of gastric cancer and to investigate its value in depicting lesions and in identifying depth invasion using virtual monoenergetic images (VMIs) on a novel dual-layer spectral detector CT.

MATERIALS AND METHODS

Eighty-two gastric cancer patients were retrospectively enrolled, and 41 patients who did not undergo surgery were evaluated for image quality in VMIs at different keVs (40 keV-70 keV with 10 keV increments) and in conventional 120 kVp polyenergetic images (PEIs) reconstructed from the portal venous phase. Objective image quality was assessed by the contrast-to-noise ratio of the gastric cancer, while subjective performance was compared using a 5-point Likert scale. Another 41 patients who underwent surgery were examined to compare the diagnostic performance of the VMIs taken at the optimal keV and that of the 120 kVp-PEIs.

RESULTS

The contrast-to-noise ratio of gastric cancer at 40 keV (10.4 ± 4.6) was the highest among all the VMIs and was significantly superior to that of the 120 kVp-PEIs (3.5 ± 1.5, p < 0.001). Gastric-specific image quality was rated highest for the 40 keV-VMIs (4.92 ± 0.26), which was significantly superior to that of the 120 kVp-PEIs (4.15 ± 0.82, p < 0.001). In the diagnostic group, there were 13 pT1, 10 pT2, 9 pT3, and 9 pT4 gastric cancer patients. Compared with the 120 kVp-PEIs, the VMIs at 40 keV tended to have a higher detection rate of gastric cancer (82.9% vs. 92.7%, respectively, p = 0.125) and a significantly improved diagnostic accuracy in the T stage (from 41.5% to 78.11%, respectively) (p < 0.001), particularly in pT1 patients, whose diagnostic accuracy was improved by 53.8% (7.7% vs. 61.5%, respectively, p = 0.016).

CONCLUSION

VMIs at 40 keV performed the best, both objectively and subjectively, for gastric cancer, leading to improved lesion depiction and higher T stage accuracy.

Multifunctional gap-enhanced Raman tags for preoperative and intraoperative cancer imaging

Multi-modality imaging agents are desirable for tumor diagnosis because they can provide more alternative and reliable information for accurate detection and therapy of diseases than single imaging technique. However, most reported conventional imaging agents have not been found to successfully overcome the disadvantages of traditional diagnoses such as sensitivity, spatial resolution, short half-decay time and complexity. Therefore, exploring a multifunctional nanocomposite with the combination of their individual modality characteristics has great impact on preoperative imaging and intraoperative diagnosis of cancer. In our study, mesoporous silica gadolinium-loaded gap-enhanced Raman tags (Gd-GERTs) specifically for preoperative and intraoperative imaging are designed and their imaging capability and biosafety are examined. They exhibit strong attenuation property for computed X-ray tomography (CT) imaging, high T₁ relaxivity for magnetic resonance (MR) imaging capability and surface-enhanced Raman spectroscopy (SERS) signal with good dispersity and stability, which presents CT/MR/SERS multi-mode imaging performance of the tumor of mice within a given time. Furthermore, in vivo biodistribution and long-term toxicity studies reveal that the Gd-GERTs have good biocompatibility and bio-safety. Therefore, Gd-GERTs are of great potential as a multifunctional nanoplatform for accurate preoperative CT/MRI diagnosis and intraoperative Raman imaging-guide resection of cancers. STATEMENT OF SIGNIFICANCE: Recent advances in molecular imaging technology have provided a myriad of opportunities to prepare various nanomaterials for accurate diagnosis and response evaluation of cancer via different imaging modalities. However, single bioimaging modality is still challenging to overcome the issues such as sensitivity, spatial resolution, imaging speed and complexity for clinicians. In this work, we designed a kind of unique multifunctional nanoprobes with computed X-ray tomography/magnetic resonance/surface-enhanced Raman spectroscopy (CT/MR/SERS) triple-modal imaging capabilities. Multifunctional nanotags offer the capabilities of preoperative noninvasive CT/MR imaging for identification of tumors as well as intraoperative real-time SERS imaging for guidance of complete resection of tumors. These multifunctional nanoprobes show critical clinical significance on the improvement of tumor diagnosis and therapy.

Dual-energy computed tomography for evaluation of breast cancer: value of virtual monoenergetic images reconstructed with a noise-reduced monoenergetic reconstruction algorithm

PURPOSE

To evaluate the image quality and lesion visibility of virtual monoenergetic images (VMIs) reconstructed using a new monoenergetic reconstruction algorithm (nMERA) for evaluation of breast cancer.

MATERIALS AND METHODS

Forty-two patients with 46 breast cancers who underwent 4-phasic breast contrast-enhanced computed tomography (CT) using dual-energy CT (DECT) were enrolled. We selected the peak enhancement phase of the lesion in each patient. The selected phase images were generated by 120-kVp-equivalent linear blended (M120) and monoenergetic reconstructions from 40 to 80 keV using the standard reconstruction algorithm (sMERA: 40, 50, 60, 70, 80) and nMERA (40 +, 50 +, 60 +, 70 +, 80 +). The contrast-to-noise ratio (CNR) was calculated and objectively analyzed. Two independent readers subjectively scored tumor visibility and image quality each on a 5-point scale.

RESULTS

The CNR at 40 + and tumor visibility scores at 40 + and 50 + were significantly higher than those on M120. The CNR at 50 + was not significantly different from that on M120. However, the overall image quality score at 40 + was significantly lower than that at 50 + and on M120 (40 + vs M120, P < 0.0001 and 40 + vs 50 +, P = 0.0001).

CONCLUSIONS

VMI reconstructed with nMERA at 50 keV is preferable for evaluation of patients with breast cancer.