Comparable Diagnostic Performance Research Articles

Clinical application of single-shot echo-planar diffusion-weighted imaging with compressed SENSE in prostate MRI at 3T: preliminary experience.

Image quality (IQ) of diffusion-weighted imaging (DWI) with single-shot echo-planar imaging (ssEPI) suffers from low signal-to-noise ratio (SNR) in high b-value acquisitions. Compressed SENSE (C-SENSE), which combines SENSE with compressed sensing, enables SNR to be improved by reducing noise. The aim of this study was to compare IQ and prostate cancer (PC) detectability between DWI with ssEPI using SENSE (EPIS) and using C-SENSE (EPICS). Twenty-five patients with pathologically proven PC underwent multi-parametric magnetic resonance imaging at 3T. DW images acquired with EPIS and EPICS were assessed for the following: lesion conspicuity (LC), SNR, contrast-to-noise ratio (CNR), mean and standard deviation (SD) of apparent diffusion coefficient (ADC) of lesion (lADCm and lADCsd), coefficient of variation of lesion ADC (lADCcv), and mean ADC of benign prostate (bADCm). LC were comparable between EPIS and EPICS (p > 0.050), and SNR and CNR were significantly higher in EPICS than EPIS (p = 0.001 and p < 0.001). In both EPIS and EPICS, lADCm was significantly lower than bADCm (p < 0.001). In addition, lADCcv was significantly lower in EPICS than in EPIS (p < 0.001). Compared with EPIS, EPICS has improved IQ and comparable diagnostic performance in PC.

Dose Reduction and Diagnostic Performance of Tin Filter-Based Spectral Shaping CT in Patients with Colorectal Cancer.

Routine CT examinations are crucial in colorectal cancer patients (CCPs); however, the high frequency of radiation exposure is a significant concern. This study investigated the radiation dose, image quality, and diagnostic performance of tin filter-based spectral shaping chest–abdominal–pelvic (CAP) CT for CCPs. We reviewed 44 CCPs who underwent single-phase enhanced tin-filtered 100 kV (TF100kV) and standard 120 kV (ST120kV) CAP CT on separate days. Radiation metrics including the volume CT dose index (CTDIvol), dose-length product (DLP), and effective dose (ED) were calculated for both protocols. Two radiologists assessed the presence of the following lesions: lung metastasis, liver metastasis, lymph node metastasis, peritoneal dissemination, and bone metastasis. The area under the receiver operating characteristic curve (AUC) was calculated for the diagnostic performance of each protocol. Radiation metrics of the TF100kV protocol were significantly lower than those of the ST120kV protocol (CDTIvol 1.60 ± 0.31 mGy vs. 14.4 ± 2.50, p < 0.0001; DLP 107.1 (95.9–125.5) mGy·cm vs. 996.7 (886.2–1144.3), p < 0.0001; ED 1.93 (1.73–2.26) mSv vs. 17.9 (16.0–20.6), p < 0.0001, respectively). TF100kV protocol achieved comparable diagnostic performance to that of the ST120kV protocol (AUC for lung metastasis: 1.00 vs. 0.94; liver metastasis: 0.88 vs. 0.83, respectively). TF100kV protocol could substantially reduce the radiation dose by 89% compared to that with the ST120kV protocol while maintaining good diagnostic performance in CCPs.

Deep learning-based image reconstruction of 40-keV virtual monoenergetic images of dual-energy CT for the assessment of hypoenhancing hepatic metastasis.

To evaluate the diagnostic value of deep learning model (DLM) reconstructed dual-energy CT (DECT) low-keV virtual monoenergetic imaging (VMI) for assessing hypoenhancing hepatic metastases. This retrospective study included 131 patients who underwent contrast-enhanced DECT (80-kVp and 150-kVp with a tin filter) in the portal venous phase for hepatic metastasis surveillance. Linearly blended images simulating 100-kVp images (100-kVp), standard 40-keV VMI images (40-keV VMI), and post-processed 40-keV VMI using a vendor-agnostic DLM (i.e., DLM 40-keV VMI) were reconstructed. Lesion conspicuity and diagnostic acceptability were assessed by three independent reviewers and compared using the Wilcoxon signed-rank test. The contrast-to-noise ratios (CNRs) were also measured placing ROIs in metastatic lesions and liver parenchyma. The detection performance of hepatic metastases was assessed by using a jackknife alternative free-response ROC method. The consensus by two independent radiologists was used as the reference standard. DLM 40-keV VMI, compared to 40-keV VMI and 100-kVp, showed a higher lesion-to-liver CNR (8.25 ± 3.23 vs. 6.05 ± 2.38 vs. 5.99 ± 2.00), better lesion conspicuity (4.3 (4.0-4.7) vs. 3.7 (3.7-4.0) vs. 3.7 (3.3-4.0)), and better diagnostic acceptability (4.3 (4.0-4.3) vs. 3.0 (2.7-3.3) vs. 4.0 (4.0-4.3)) (p < 0.001 for all). For lesion detection (246 hepatic metastases in 68 patients), the figure of merit was significantly higher with DLM 40-keV VMI than with 40-keV VMI (0.852 vs. 0.822, p = 0.012), whereas no significant difference existed between DLM 40-keV VMI and 100-kVp (0.852 vs. 0.842, p = 0.31). DLM 40-keV VMI provided better image quality and comparable diagnostic performance for detecting hypoenhancing hepatic metastases compared to linearly blended images. • DLM 40-keV VMI provides a superior image quality compared with 40-keV or 100-kVp for assessing hypoenhancing hepatic metastasis. • DLM 40-keV VMI has the highest CNR and lesion conspicuity score for hypoenhancing hepatic metastasis due to noise reduction and structural preservation. • DLM 40-keV VMI provides higher lesion detectability than standard 40-keV VMI (p = 0.012).

Accurate preoperative staging with [68Ga]Ga-FAPI PET/CT for patients with oral squamous cell carcinoma: a comparison to 2-[18F]FDG PET/CT.

To evaluate the potential value of [68Ga]Ga-labelled fibroblast activation protein inhibitor ([68Ga]Ga-FAPI) positron emission tomography/computed tomography (PET/CT) in preoperative staging for patients with oral squamous cell carcinoma (OSCC) as compared to 2-[18F]fluoro-2-deoxy-D-glucose (2-[18F]FDG) PET/CT. Thirty-six treatment-naïve patients with OSCC who underwent 2-[18F]FDG and [68Ga]Ga-FAPI PET/CT for preoperative staging were enrolled. The maximum standardised uptake value (SUVmax) of the primary tumour and suspected cervical metastatic lymph nodes, and the tumour-to-background ratio (TBR) of the primary tumour, were measured. The accuracy of two imaging modalities for preoperative diagnosis of metastatic lymph nodes was analysed. Histopathology served as the standard of reference. Thirty-seven primary lesions of 36 patients were accurately detected by both [68Ga]Ga-FAPI and 2-[18F]FDG PET/CT. Regarding primary tumours, the SUVmax and TBR of the two imaging modalities in stage T3-T4 were significantly higher than those of stage T1-T2 (all p < 0.05). On the patient analysis, the accuracy for the evaluation of N1-N3 neck status was 52.6% (10/19) for [68Ga]Ga-FAPI PET/CT and 57.9% (11/19) for 2-[18F]FDG PET/CT. Notably, the accuracy for the evaluation of the N0 neck status between [68Ga]Ga-FAPI and 2-[18F]FDG PET/CT was 100% (17/17) and 29% (5/17), respectively. Based on the patient, neck side and neck level, [68Ga]Ga-FAPI PET/CT resulted in higher specificity and accuracy in diagnosing metastatic neck lymph nodes than 2-[18F]FDG PET/CT (all p < 0.05). [68Ga]Ga-FAPI PET/CT is a promising tool for preoperative staging of OSCC, and appears to reduce the false positivity seen with 2-[18F]FDG PET/CT for the detection of neck lymph node metastases. • [68Ga]Ga-FAPI PET/CT is a promising tool targeting cancer-associated fibroblasts with comparable diagnostic performance to 2-[18F]FDG PET/CT for identifying the primary lesions of OSCC. • [68Ga]Ga-FAPI PET/CT showed higher specificity and accuracy for the evaluation of neck lymph node metastases of OSCC than 2-[18F]FDG PET/CT, especially for N0 neck status.

Optimized cutoffs of gamma-glutamyl transpeptidase-to-platelet ratio, aspartate aminotransferase-to-platelet ratio index, and fibrosis-4 scoring systems for exclusion of cirrhosis in patients with chronic hepatitis B.

Accurate prediction of the extent of fibrosis is of great clinical importance in patients infected with chronic hepatitis B (CHB). This study aimed to compare the performance of gamma‐glutamyl transpeptidase‐to‐platelet ratio (GPR), aspartate aminotransferase‐to‐platelet ratio index (APRI), and fibrosis‐4 (FIB‐4) in evaluating liver fibrosis stages and to identify optimized cutoffs to exclude cirrhosis. Consecutive patients with CHB with liver biopsies were enrolled and randomly divided into derivation and validation cohorts. Areas under the receiver operating characteristic curve were used to evaluate the diagnostic performance of APRI, FIB‐4, and GPR to distinguish fibrosis stages. New cutoffs with a sensitivity of at least 90% and a negative predictive value (NPV) of more than 95% were identified. A total of 880 individuals were enrolled in this study. The derivation data set consisted of 617 patients, with 82 patients with cirrhosis. In the validation cohort (n = 263), 29 patients had cirrhosis. APRI, FIB‐4, and GPR had comparable diagnostic performance for diagnosing significant fibrosis. GPR outperformed APRI (p < 0.05) in the prediction of cirrhosis. A newly identified GPR score of 0.35 had a sensitivity and NPV of 93.9% and 98.0%, respectively, and misclassified 5 of 82 (6.1%) patients with cirrhosis in the derivation group. All new cutoffs identified in this study also reached our goal in the validation cohort. The new GPR score could rule out a larger proportion of individuals without cirrhosis, and the subgroup analysis showed more stable performance. However, the lower cutoff dose increases the need for further testing compared to the conventional cutoff. Conclusion: A newly identified cutoff for GPR (<0.35) could rule out more patients without cirrhosis compared to APRI and FIB‐4 and have low misclassification rates.

An artificial intelligence system using maximum intensity projection MR images facilitates classification of non-mass enhancement breast lesions.

To build an artificial intelligence (AI) system to classify benign and malignant non-mass enhancement (NME) lesions using maximum intensity projection (MIP) of early post-contrast subtracted breast MR images. This retrospective study collected 965 pure NME lesions (539 benign and 426 malignant) confirmed by histopathology or follow-up in 903 women. The 754 NME lesions acquired by one MR scanner were randomly split into the training set, validation set, and test set A (482/121/151 lesions). The 211 NME lesions acquired by another MR scanner were used as test set B. The AI system was developed using ResNet-50 with the axial and sagittal MIP images. One senior and one junior radiologist reviewed the MIP images of each case independently and rated its Breast Imaging Reporting and Data System category. The performance of the AI system and the radiologists was evaluated using the area under the receiver operating characteristic curve (AUC). The AI system yielded AUCs of 0.859 and 0.816 in the test sets A and B, respectively. The AI system achieved comparable performance as the senior radiologist (p = 0.558, p = 0.041) and outperformed the junior radiologist (p < 0.001, p = 0.009) in both test sets A and B. After AI assistance, the AUC of the junior radiologist increased from 0.740 to 0.862 in test set A (p < 0.001) and from 0.732 to 0.843 in test set B (p < 0.001). Our MIP-based AI system yielded good applicability in classifying NME lesions in breast MRI and can assist the junior radiologist achieve better performance. • Our MIP-based AI system yielded good applicability in the dataset both from the same and a different MR scanner in predicting malignant NME lesions. • The AI system achieved comparable diagnostic performance with the senior radiologist and outperformed the junior radiologist. • This AI system can assist the junior radiologist achieve better performance in the classification of NME lesions in MRI.

Accelerating susceptibility-weighted imaging with deep learning by complex-valued convolutional neural network (ComplexNet): validation in clinical brain imaging.

Susceptibility-weighted imaging (SWI) is crucial for the characterization of intracranial hemorrhage and mineralization, but has the drawback of long acquisition times. We aimed to propose a deep learning model to accelerate SWI, and evaluate the clinical feasibility of this approach. A complex-valued convolutional neural network (ComplexNet) was developed to reconstruct high-quality SWI from highly accelerated k-space data. ComplexNet can leverage the inherently complex-valued nature of SWI data and learn richer representations by using complex-valued network. SWI data were acquired from 117 participants who underwent clinical brain MRI examination between 2019 and 2021, including patients with tumor, stroke, hemorrhage, traumatic brain injury, etc. Reconstruction quality was evaluated using quantitative image metrics and image quality scores, including overall image quality, signal-to-noise ratio, sharpness, and artifacts. The average reconstruction time of ComplexNet was 19 ms per section (1.33 s per participant). ComplexNet achieved significantly improved quantitative image metrics compared to a conventional compressed sensing method and a real-valued network with acceleration rates of 5 and 8 (p < 0.001). Meanwhile, there was no significant difference between fully sampled and ComplexNet approaches in terms of overall image quality and artifacts (p > 0.05) at both acceleration rates. Furthermore, ComplexNet showed comparable diagnostic performance to the fully sampled SWI for visualizing a wide range of pathology, including hemorrhage, cerebral microbleeds, and brain tumor. ComplexNet can effectively accelerate SWI while providing superior performance in terms of overall image quality and visualization of pathology for routine clinical brain imaging. • The complex-valued convolutional neural network (ComplexNet) allowed fast and high-quality reconstruction of highly accelerated SWI data, with an average reconstruction time of 19 ms per section. • ComplexNet achieved significantly improved quantitative image metrics compared to a conventional compressed sensing method and a real-valued network with acceleration rates of 5 and 8 (p < 0.001). • ComplexNet showed comparable diagnostic performance to the fully sampled SWI for visualizing a wide range of pathology, including hemorrhage, cerebral microbleeds, and brain tumor.

Application Values of 2D and 3D Radiomics Models Based on CT Plain Scan in Differentiating Benign from Malignant Ovarian Tumors.

Background Accurate identification of ovarian tumors as benign or malignant is highly crucial. Radiomics is a new branch of imaging that has emerged in recent years to replace the traditional naked eye qualitative diagnosis. Objective This study is aimed at exploring the difference in the application potential of two- (2D) and three-dimensional (3D) radiomics models based on CT plain scan in differentiating benign from malignant ovarian tumors. Method A retrospective analysis was performed on 140 patients with ovarian tumors confirmed by surgery and pathology in our hospital from July 2017 to August 2020. These 140 patients were divided into benign group and malignant group according to the pathological results. The ITK-SNAP software was used to outline the regions-of-interest (ROI) of 2D or 3D tumors on the CT plain scan image of each patient; the texture features were extracted through analysis kit (AK), and the cases were randomly divided into training groups (n = 99) and validation group (n = 41) in a ratio of 7 : 3. The least absolute shrinkage and selection operator (LASSO) algorithm was used to perform dimensionality reduction, followed by the construction of the radiomics nomogram model using the logistic regression method. The receiver operating characteristic (ROC) curve was drawn, and the calibration curve and decision curve analysis (DCA) were used to evaluate and verify the results of the radiomics nomogram and compare the differences between 2D and 3D diagnostic performance. Results There were 396 quantitative radiomics feature parameters extracted from 2D group and the 3D group, respectively. The area under the curve (AUC) of the radiomics nomogram of the 2D training group and the validation group were 0.96 and 0.97, respectively. The accuracy, specificity, and sensitivity of the training set were 92.9%, 88.9%, and 96.3%, respectively, and those of the validation set were 90.2%, 82.6%, and 100.0%, respectively. The AUCs of the radiomics nomogram of the 3D training group and validation group were 0.96% and 0.99%, respectively. The accuracy, sensitivity, and specificity of the training set were 92.9%, 96.3%, and 88.9%, respectively, and those of the validation set were 97.6%, 95.7%, and 100.0%, respectively. DeLong's test indicated that there was no statistical significance between the two sets (P > 0.05). Conclusions For the differential diagnosis of benign and malignant ovarian tumors, the 2D and 3D radiomics nomogram models exhibited comparable diagnostic performance. Considering that the 2D model was cost-effective and time-efficient, it was more recommended to use 2D features in future research.

Clinical Feasibility of Ultrafast Contrast-Enhanced T1-Weighted 3D-EPI for Evaluating Intracranial Enhancing Lesions in Oncology Patients: Comparison with Standard 3D MPRAGE Sequence.

Contrast-enhanced 3D T1WI is a preferred sequence for brain tumor imaging despite the long scan time. This study investigated the clinical feasibility of ultrafast contrast-enhanced T1WI by 3D echo-planar imaging compared with a standard contrast-enhanced 3D MPRAGE sequence for evaluating intracranial enhancing lesions in oncology patients. Sixty-one patients in oncology underwent brain MR imaging including both contrast-enhanced T1WI, 3D-EPI and 3D MPRAGE, in a single examination session for evaluating intracranial tumors. Two neuroradiologists evaluated image quality, lesion conspicuity, diagnostic confidence, number and size of the lesions, and contrast-to-noise ratio measurements from the 2 different sequences. Ultrafast 3D-EPI T1WI did not reveal significant differences in diagnostic confidence, contrast-to-noise ratiolesion/parenchyma, and the number of enhancing lesions compared with MPRAGE (P > .05). However, ultrafast 3D-EPI T1WI revealed inferior image quality, inferior anatomic delineation and greater susceptibility artifacts with fewer motion artifacts than images obtained with MPRAGE. The mean contrast-to-noise ratioWM/GM and visual conspicuity of the lesion on ultrafast 3D-EPI T1WI were lower than those of MPRAGE (P < .001). Ultrafast 3D-EPI T1WI showed comparable diagnostic performance with sufficient image quality and a 7-fold reduction in scan time for evaluating intracranial enhancing lesions compared with standard MPRAGE, even though it was limited by an inferior image quality and frequent susceptibility artifacts. Therefore, we believe that ultrafast 3D-EPI T1WI may be a viable option in oncology patients prone to movement during imaging studies.

Black-blood Venous Imaging (BBVI): A Contrast-Free and High-Resolution Magnetic Resonance Approach for Diagnosing IVCS – a Proof of Concept Study

BackgroundIliac vein compression syndrome (IVCS) diagnosis heavily relies on an imaging test. However, non-invasive and contrast-free imaging test for the diagnosis of IVCS remains a big challenge. To address this issue, this prospective study aimed to assess the image quality and diagnostic performance of a magnetic resonance imaging technique, black-blood venous imaging (BBVI), in detecting IVCS by comparing it with contrast-enhanced computed tomography venography (CTV) and using invasive digital subtraction angiography (DSA) as the reference.MethodsWe enrolled 105 patients, and all patients underwent BBVI, CTV, and DSA examinations. We compared the consistency of CTV and BBVI image quality and their consistency in diagnosing the rate of iliac vein stenosis in IVCS patients. Using the consensus DSA as a reference, the sensitivity, specificity, positive and negative predictive values, and accuracy of BBVI and CTV and their diagnostic agreement with DSA were calculated.ResultsBBVI demonstrated high sensitivity, specificity, and accuracy for the diagnosis of IVCS, without contrast agents. BBVI and CTV are quite in diagnosis IVCS. Quite SE (67.8% vs 68.3%), SP (94.8% vs 94.8%), PPV (98.0% vs 98.0%), NPV (46.2% vs 46.9%) and ACC (75.3% vs 75.7%) were obtained by BBVI in comparison with CTV.ConclusionBBVI has comparable diagnostic performance with CTV. It may be a viable alternative to CTV techniques in screening the IVCS without contrast agents and free of ionizing radiation.

Diffusion-weighted magnetic resonance imaging at 1.5 T for peripheral zone prostate cancer: the influence of the b-value combination on the diagnostic performance of apparent diffusion coefficient.

PurposeDiffusion-weighted imaging as a noninvasive functional modality plays a valuable role in the evaluation of prostate cancer. However, there is still no agreement on the number and range of b-values to be used. Therefore, the purpose of this study is to investigate the influence of b-value choice on the diagnostic performance of apparent diffusion coefficient (ADC) values for prostate cancer detection.Material and methodsFifty-nine consecutive patients with abnormal digital rectal examination findings and raised serum prostate-specific antigen were chosen for magnetic resonance imaging of the prostate before systematic 12-core trans-rectal ultrasound-guided prostate biopsies. ADC values for each ROI were calculated from different b-value combinations (0-1600 s/mm2) by a monoexponential model. Mann-Whitney and the paired-sample t-test were used to compare the mean ADC values for malignant lesions and noncancerous tissues. ROC curve analysis was used to evaluate the diagnostic performance of ADC values in distinguishing prostate cancer from normal-tissue ROIs.ResultsThe differences between mean ADC values of malignant lesions and contralateral healthy tissues were significant for all the pairs of b-value combinations. The pair of b-values 50 and 1200 provided the highest AUC (0.94), with a sensitivity of 90.2%, a specificity of 92.6%, and an accuracy of 91.2% at an ADC cut-off of 1.23 × 10-3 mm2/s.ConclusionsOur study showed that using a 1.5-Tesla MRI scanner the diagnostic performance of ADC values estimated from the b-value pair 50 and 1200 s/mm2 was highest. However, some other b-value pairs provided statically comparable diagnostic performance.

Prospective evaluation of thoracic diseases using a compact flat-panel detector spiral computed tomographic scanner

ObjectiveTo prospectively evaluate the image quality and diagnostic performance of a compact flat-panel detector (FD) scanner for thoracic diseases compared to a clinical CT scanner. Materials and methodsThe institutional review board approved this single-center prospective study, and all participants provided informed consent. From December 2020 to May 2021, 30 patients (mean age, 67.1 ± 8.3 years) underwent two same-day low-dose chest CT scans using clinical state-of-art and compact FDCT scanners. Image quality was assessed visually and quantitatively. Two readers evaluated the diagnostic performance for nodules, parenchymal opacifications, bronchiectasis, linear opacities, and pleural abnormalities in 40 paired CT scans. The other 20 paired CT scans were used to examine the agreement of semi-quantitative CT scoring regarding bronchiectasis, bronchiolitis, nodules, airspace consolidations, and cavities. ResultsFDCT images had significantly lower visual image quality than clinical CT images (all p < 0.001). The two CT image sets showed no significant differences in signal-to-noise and contrast-to-noise ratios (56.8 ± 12.5 vs. 57.3 ± 15.2; p = 0.985 and 62.9 ± 11.7 vs. 60.7 ± 16.9; p = 0.615). The pooled sensitivity was comparable for nodules, parenchymal opacifications, linear opacities, and pleural abnormalities (p = 0.065–0.625), whereas the sensitivity was significantly lower in FDCT images than in clinical CT images for micronodules (p = 0.007) and bronchiectasis (p = 0.004). The specificity was mostly 1.0. Semi-quantitative CT scores were similar between the CT image sets (p > 0.05), and intraclass correlation coefficients were around 0.950 or higher, except for bronchiectasis (0.869). ConclusionCompact FDCT images provided lower image quality but comparable diagnostic performance to clinical CT images for nodules, parenchymal opacifications, linear opacities, and pleural abnormalities.

Performance of an Artificial Intelligence-based Application for the Detection of Plaque-based Stenosis on Monoenergetic Coronary CT Angiography: Validation by Invasive Coronary Angiography

To explore the value of an artificial intelligence (AI)-based application for identifying plaque-specific stenosis and obstructive coronary artery disease from monoenergetic spectral reconstructions on coronary computed tomography angiography (CTA). This retrospective study enrolled 71 consecutive patients (52 men, 19 women; 63.3 ± 10.7 years) who underwent coronary CTA and invasive coronary angiography for diagnosing coronary artery disease. The conventional 120 kVp images and eight different virtual monoenergetic images (VMIs) (from 40 keV to 140 keV at increment of 10 keV) were reconstructed. An AI system automatically detected plaques from the conventional 120 kVp images and VMIs and calculated the degree of stenosis, which was further compared to invasive coronary angiography. The assessment was performed at a segment, vessel, and patient level. Vessel and segment-based analyses showed comparable diagnostic performance between conventional CTA images and VMIs from 50 keV to 90 keV. For vessel-based analysis, the sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of conventional CTA were 74.3% (95% CI: 64.9%-82.0%), 85.6% (95% CI: 77.0%-91.4%), 84.3% (95% CI: 75.2%-90.7%), 76.1% (95% CI: 67.1%-83.3%) and 79.8% (95% CI: 73.7%-84.9%), respectively; the average sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy values of the VMIs ranging from 50 keV to 90 keV were 71.6%, 90.7%, 87.5%, 64.1% and 81.6%, respectively. For plaque-based assessment, diagnostic performance of the average VMIs ranging from 50 keV to 100 keV showed no significant statistical difference in diagnostic accuracy compared to those of conventional CTA images in detecting calcified (91.4% vs. 93.8%, p > 0.05), noncalcified (92.6% vs. 85.2%, p > 0.05) or mixed (80.2% vs. 81.2%, p > 0.05) stenosis, although the specificity was slightly higher (53.4% vs. 40.0%, p > 0.05) in detecting stenosis caused by mixed plaques. For VMIs above 100 keV, the diagnostic accuracy dropped significantly. Our study showed that the performance of an AI-based application employed to detect significant coronary stenosis in virtual monoenergetic reconstructions ranging from 50 keV to 90 keV was comparable to conventional 120 kVp reconstructions.

Head-to-Head Comparison of 68Ga-FAPI-46 and 18F-FDG PET/CT for Evaluation of Head and Neck Squamous Cell Carcinoma: A Single-Center Exploratory Study.

68Ga-conjugated fibroblast activation protein inhibitor (68Ga-FAPI) has become an attractive agent for PET. This study aimed to compare 68Ga-FAPI-46 PET/CT with 18F-FDG PET/CT for detecting primary cancer and metastatic lesions in patients with head and neck squamous cell carcinoma (HNSCC). Methods: Twelve patients and 28 patients with HNSCC underwent 68Ga-FAPI-46 and 18F-FDG PET/CT for initial staging and recurrence detection, respectively. The concordance and diagnostic accuracy of both tracers were analyzed. Semiquantitative parameters, including SUVmax, SUVmean, and tumor-to-background ratio, were compared. Fibroblast activation protein (FAP) expression tumor volume and total lesion FAP expression of 68Ga-FAPI-46 were compared with metabolic tumor volume and total lesion glycolysis of 18F-FDG, respectively. Differences between semiquantitative parameters were analyzed using paired t testing. Results:68Ga-FAPI-46 PET/CT was 83.3% and 96.4% concordant with 18F-FDG PET/CT for initial staging and recurrence detection, respectively. Eighteen lesions had histopathologic validation, and both tracers displayed 100% sensitivity, 50% specificity, and 94.4% accuracy for lesion-based analysis. FAP expression tumor volume was greater than metabolic tumor volume (P < 0.05), but no significant differences were observed for the other parameters. Conclusion:68Ga-FAPI-46 PET/CT showed good concordance with, and comparable diagnostic performance to, 18F-FDG PET/CT for initial staging and recurrence detection in HNSCC patients.

Role of Advanced MRI Techniques in the Quantitative Assessment of Liver Fat—A Multimodality-Based Comparative Study of Diagnostic Performance in a Tertiary Care Institute

Abstract Objective To evaluate the accuracy of noninvasive imaging methods including gray-scale ultrasound, ultrasound shear wave elastography, unenhanced computed tomography (CT), and proton density fat fraction (PDFF) on magnetic resonance imaging (MRI) using three-dimensional (3D) multiecho multipoint chemical shift–encoded spoiled gradient echo (q-DIXON) sequence in the quantification of hepatic steatosis, with proton MR spectroscopy (H1-MRS) as the reference standard in Indian population. Methods Our study included 100 consecutive adult patients referred to the department of radiology in our hospital for imaging of liver. Fat content of liver was recorded using MRI (H1-MRS and q-DIXON), unenhanced CT (average liver attenuation [ALI] and liver attenuation index [LAI]) and ultrasonography (USG) (gray-scale grading and shear wave elastography [SWE]). Data were analyzed by linear regression and Bland–Altman analysis for each technique compared with H1-MRS. The diagnostic performances of all the methods were compared using DeLong test, for detection of mild and moderate-to-severe hepatic steatosis, separately. Results MRI q-DIXON PDFF showed excellent correlation (r = 0.917, r2 = 0.840) and strong agreement (1.48 ±3.01) with H1-MRS-derived PDFF measurements. Unenhanced CT-based methods showed moderate correlation with modest agreement (r = −0.826, r2 = 0.681, −40.18 ± 16.05 for ALI and r = −0.858, r2 = 0.735, 13.4 ± 15.3 for LAI) whereas USG gray-scale assessment showed low correlation (weighted Kappa value 0.366) with H1-MRS PDFF. No correlation was found between USG-SWE results and PDFF measured with H1-MRS. Comparison of areas under curve (AUCs) using DeLong test revealed that MRI q-DIXON method performed the best for diagnosis of hepatic steatosis compared with rest. For moderate to severe steatosis, MRI q-DIXON and unenhanced CT-based methods had comparable diagnostic performance with AUCs not showing statistically significant differences. Conclusion MRI q-DIXON shows strongest correlation with MRS and should be preferred for estimation of hepatic fat, especially when MRS is not available. Unenhanced CT shows limited diagnostic performance in detecting mild steatosis; however, it certainly has a role in diagnosing moderate-to-severe hepatic steatosis, such as evaluating donor candidates for living donor liver transplantation. USG, using both the traditional four-grade visual assessment and elastography in the present form, appears to have limited role in liver fat quantification.

Turbo Gradient and Spin Echo PROPELLER-Diffusion Weighted Imaging for Orbital Tumors: A Comparative Study With Readout-Segmented Echo-Planar Imaging.

Purpose: To qualitatively and quantitatively compare the image quality and diagnostic performance of turbo gradient and spin echo PROPELLER diffusion-weighted imaging (TGSE-PROPELLER-DWI) vs. readout-segmented echo-planar imaging (rs-EPI) in the evaluation of orbital tumors.Materials and Methods: A total of 43 patients with suspected orbital tumors were enrolled to perform the two DWIs with comparable spatial resolution on 3T. The overall image qualities, geometric distortions, susceptibility artifacts, and lesion conspicuities were scored by using a four-point scale (1, poor; 4, excellent). Quantitative measurements, including contrast-to-noise ratios (CNRs), apparent diffusion coefficients (ADCs), geometric distortion rates (GDRs), and lesion sizes, were calculated and compared. The two ADCs for differentiating malignant from benign orbital tumors were evaluated. Wilcoxon signed-rank test, Kappa statistic, and receiver operating characteristics (ROC) curves were used.Results: TGSE-PROPELLER-DWI performed superior in all subjective scores and quantitative GDR evaluation than rs-EPI (p < 0.001), and excellent interobserver agreement was obtained for Kappa value ranging from 0.876 to 1.000. ADClesion of TGSE-PROPELLER-DWI was significantly higher than those of rs-EPI (p < 0.001). Mean ADC of malignant tumors was significantly lower than that of benign tumors both in two DWIs. However, the AUC for differentiating malignant and benign tumors showed no significant difference in the two DWIs (0.860 vs. 0.854, p = 0.7448). Sensitivity and specificity could achieve 92.86% and 72.73% for TGSE-PROPELLER-DWI with a cutoff value of 1.23 × 10–3 mm2/s, and 85.71% and 81.82% for rs-EPI with a cutoff value of 0.99 × 10–3 mm2/s.Conclusion: Compared with rs-EPI, TGSE-PROPELLER-DWI showed minimized geometric distortion and susceptibility artifacts significantly improved the image quality for orbital tumors and achieved comparable diagnostic performance in differentiating malignant and benign orbital tumors.

Machine learning of treadmill exercise test to improve selection for testing for coronary artery disease

Background and aimsThe high false-positive rate of the treadmill exercise test (TET) may lead to unnecessary invasive coronary angiography. We aimed to develop a machine learning-based algorithm to improve the diagnostic performance of TET. MethodsStudy included 2325 patients who underwent TET and subsequent coronary angiography within one-year interval. The mean age was 58.7 (48.1–69.3) years, 1731 (74.5%) were male, 1858 (79.9%) had positive TET result, and 812 (34.9%) had obstructive coronary artery disease (≥70% stenosis in at least one vessel). The study population were randomly divided into training (70%) and testing (30%) groups for algorithm development. A total of 93 features, including exercise performance, hemodynamics and ST-segment changes were extracted from the TET results. Clinical features included comorbidity, smoking, height, weight, and Framingham risk score. Support vector machine, logistic regression, random forest, k-nearest neighbor and extreme gradient boosting machine learning algorithms were used to build the predictive models. The performance of each model was compared with that of conventional TET. ResultsFour of the five models exhibited comparable diagnostic performance and were better than conventional TET. The random forest algorithm had an area under the curve (AUC) of 0.73. When used with clinical features, the AUC improved to 0.74. The major advantage of the algorithm is the reduction of the false-positive rate compared with conventional TET (55% vs. 76.3%, respectively), while maintaining comparable sensitivity (85%). ConclusionsUsing the information obtained from conventional TET, a more accurate diagnosis can be made by incorporating an artificial intelligence-based model.

Deep learning models for screening of high myopia using optical coherence tomography

This study aimed to validate and evaluate deep learning (DL) models for screening of high myopia using spectral-domain optical coherence tomography (OCT). This retrospective cross-sectional study included 690 eyes in 492 patients with OCT images and axial length measurement. Eyes were divided into three groups based on axial length: a “normal group,” a “high myopia group,” and an “other retinal disease” group. The researchers trained and validated three DL models to classify the three groups based on horizontal and vertical OCT images of the 600 eyes. For evaluation, OCT images of 90 eyes were used. Diagnostic agreements of human doctors and DL models were analyzed. The area under the receiver operating characteristic curve of the three DL models was evaluated. Absolute agreement of retina specialists was 99.11% (range: 97.78–100%). Absolute agreement of the DL models with multiple-column model was 100.0% (ResNet 50), 90.0% (Inception V3), and 72.22% (VGG 16). Areas under the receiver operating characteristic curves of the DL models with multiple-column model were 0.99 (ResNet 50), 0.97 (Inception V3), and 0.86 (VGG 16). The DL model based on ResNet 50 showed comparable diagnostic performance with retinal specialists. The DL model using OCT images demonstrated reliable diagnostic performance to identify high myopia.

Diagnostic Performance of Fractional Flow Reserve From CT Coronary Angiography With Analytical Method.

The aim of this study was to evaluate a new analytical method for calculating non-invasive fractional flow reserve (FFRAM) to diagnose ischemic coronary lesions. Patients with suspected or known coronary artery disease (CAD) who underwent computed tomography coronary angiography (CTCA) and invasive coronary angiography (ICA) with FFR measurements from two sites were prospectively recruited. Obstructive CAD was defined as diameter stenosis (DS) ≥50% on CTCA or ICA. FFRAM was derived from CTCA images and anatomical features using analytical method and was compared with computational fluid dynamics (CFD)-based FFR (FFRB) and invasive ICA-based FFR. FFRAM, FFRB, and invasive FFR ≤ 0.80 defined ischemia. A total of 108 participants (mean age 60, range: 30–83 years, 75% men) with 169 stenosed coronary arteries were analyzed. The per-vessel accuracy, sensitivity, specificity, and positive predictive and negative predictive values were, respectively, 81, 75, 86, 81, and 82% for FFRAM and 87, 88, 86, 83, and 90% for FFRB. The area under the receiver operating characteristics curve for FFRAM (0.89 and 0.87) and FFRB (0.90 and 0.86) were higher than both CTCA- and ICA-derived DS (all p < 0.0001) on per-vessel and per-patient bases for discriminating ischemic lesions. The computational time for FFRAM was much shorter than FFRB (2.2 ± 0.9 min vs. 48 ± 36 min, excluding image acquisition and segmentation). FFRAM calculated from a novel and expeditious non-CFD approach possesses a comparable diagnostic performance to CFD-derived FFRB, with a significantly shorter computational time.

A plasma and serum 5-HIAA assay with comparable diagnostic performance in patients with neuroendocrine tumours

A plasma and serum 5-HIAA assay with comparable diagnostic performance in patients with neuroendocrine tumours