Search

Background: The aim of our current systematic dynamic phantom study was first, to optimize reconstruction parameters of coronary CTA (CCTA) acquired on photon counting CT (PCCT) for coronary artery calcium (CAC) scoring, and second, to assess the feasibility of calculating CAC scores from CCTA, in comparison to reference calcium scoring CT (CSCT) scans. Methods: In this phantom study, an artificial coronary artery was translated at velocities corresponding to 0, < 60, and 60–75 beats per minute (bpm) within an anthropomorphic phantom. The density of calcifications was 100 (very low), 200 (low), 400 (medium), and 800 (high) mgHA/cm3, respectively. CCTA was reconstructed with the following parameters: virtual non-iodine (VNI), with and without iterative reconstruction (QIR level 2, QIR off, respectively); kernels Qr36 and Qr44f; slice thickness/increment 3.0/1.5 mm and 0.4/0.2 mm. The agreement in risk group classification between CACCCTA and CACCSCT scoring was measured using Cohen weighted linear κ with 95% CI. Results: For CCTA reconstructed with 0.4 mm slice thickness, calcium detectability was perfect (100%). At < 60 bpm, CACCCTA of low, and medium density calcification was underestimated by 53%, and 15%, respectively. However, CACCCTA was not significantly different from CACCSCT of very low, and high-density calcifications. The best risk agreement was achieved when CCTA was reconstructed with QIR off, Qr44f, and 0.4 mm slice thickness (κ = 0.762, 95% CI 0.671–0.853). Conclusion: In this dynamic phantom study, the detection of calcifications with different densities was excellent with CCTA on PCCT using thin-slice VNI reconstruction. Agatston scores were underestimated compared to CSCT but agreement in risk classification was substantial. Clinical relevance statement: Photon counting CT may enable the implementation of coronary artery calcium scoring from coronary CTA in daily clinical practice. Key Points: Photon-counting CTA allows for excellent detectability of low-density calcifications at all heart rates. Coronary artery calcium scoring from coronary CTA acquired on photon counting CT is feasible, although improvement is needed. Adoption of the standard acquisition and reconstruction protocol for calcium scoring is needed for improved quantification of coronary artery calcium to fully employ the potential of photon counting CT. [ABSTRACT FROM AUTHOR]

Purpose: To determine the optimal window setting for virtual monoenergetic images (VMI) reconstructed from dual-layer spectral coronary computed tomography angiography (DE-CCTA) datasets. Material and methods: 50 patients (30 males; mean age 61.1 ± 12.4 years who underwent DE-CCTA from May 2021 to June 2022 for suspected coronary artery disease, were retrospectively included. Image quality assessment was performed on conventional images and VMI reconstructions at 70 and 40 keV. Objective image quality was assessed using contrast-to-noise ratio (CNR). Two independent observers manually identified the best window settings (B-W/L) for VMI 70 and VMI 40 visualization. B-W/L were then normalized with aortic attenuation using linear regression analysis to obtain the optimized W/L (O-W/L) settings. Additionally, subjective image quality was evaluated using a 5-point Likert scale, and vessel diameters were measured to examine any potential impact of different W/L settings. Results: VMI 40 demonstrated higher CNR values compared to conventional and VMI 70. B-W/L settings identified were 1180/280 HU for VMI 70 and 3290/900 HU for VMI 40. Subsequent linear regression analysis yielded O-W/L settings of 1155/270 HU for VMI 70 and 3230/880 HU for VMI 40. VMI 40 O-W/L received the highest scores for each parameter compared to conventional (all p < 0.0027). Using O-W/L settings for VMI 70 and VMI 40 did not result in significant differences in vessel measurements compared to conventional images. Conclusion: Optimization of VMI requires adjustments in W/L settings. Our results recommend W/L settings of 1155/270 HU for VMI 70 and 3230/880 HU for VMI 40. [ABSTRACT FROM AUTHOR]

Treatment response assessment of rectal cancer patients is a critical component of personalized cancer care and it allows to identify suitable candidates for organ-preserving strategies. This pilot study employed a novel multi-omics approach combining MRI-based radiomic features and untargeted metabolomics to infer treatment response at staging. The metabolic signature highlighted how tumor cell viability is predictively down-regulated, while the response to oxidative stress was up-regulated in responder patients, showing significantly reduced oxoproline values at baseline compared to non-responder patients (p-value < 10–4). Tumors with a high degree of texture homogeneity, as assessed by radiomics, were more likely to achieve a major pathological response (p-value < 10–3). A machine learning classifier was implemented to summarize the multi-omics information and discriminate responders and non-responders. Combining all available radiomic and metabolomic features, the classifier delivered an AUC of 0.864 (± 0.083, p-value < 10–3) with a best-point sensitivity of 90.9% and a specificity of 81.8%. Our results suggest that a multi-omics approach, integrating radiomics and metabolomic data, can enhance the predictive value of standard MRI and could help to avoid unnecessary surgical treatments and their associated long-term complications. [ABSTRACT FROM AUTHOR]

Purpose: To develop and assess the utility of synthetic dual-energy CT (sDECT) images generated from single-energy CT (SECT) using two state-of-the-art generative adversarial network (GAN) architectures for artificial intelligence-based image translation. Methods: In this retrospective study, 734 patients (389F; 62.8 years ± 14.9) who underwent enhanced DECT of the chest, abdomen, and pelvis between January 2018 and June 2019 were included. Using 70-keV as the input images (n = 141,009) and 50-keV, iodine, and virtual unenhanced (VUE) images as outputs, separate models were trained using Pix2PixHD and CycleGAN. Model performance on the test set (n = 17,839) was evaluated using mean squared error, structural similarity index, and peak signal-to-noise ratio. To objectively test the utility of these models, synthetic iodine material density and 50-keV images were generated from SECT images of 16 patients with gastrointestinal bleeding performed at another institution. The conspicuity of gastrointestinal bleeding using sDECT was compared to portal venous phase SECT. Synthetic VUE images were generated from 37 patients who underwent a CT urogram at another institution and model performance was compared to true unenhanced images. Results: sDECT from both Pix2PixHD and CycleGAN were qualitatively indistinguishable from true DECT by a board-certified radiologist (avg accuracy 64.5%). Pix2PixHD had better quantitative performance compared to CycleGAN (e.g., structural similarity index for iodine: 87% vs. 46%, p-value < 0.001). sDECT using Pix2PixHD showed increased bleeding conspicuity for gastrointestinal bleeding and better removal of iodine on synthetic VUE compared to CycleGAN. Conclusions: sDECT from SECT using Pix2PixHD may afford some of the advantages of DECT. [ABSTRACT FROM AUTHOR]

Oncotype Dx Recurrence Score (RS) has been validated in patients with ER + /HER2 − invasive breast carcinoma to estimate patient risk of recurrence and guide the use of adjuvant chemotherapy. We investigated the role of MRI-based radiomics features extracted from the tumor and the peritumoral tissues to predict the risk of tumor recurrence. A total of 62 patients with biopsy-proved ER + /HER2 − breast cancer who underwent pre-treatment MRI and Oncotype Dx were included. An RS > 25 was considered discriminant between low-intermediate and high risk of tumor recurrence. Two readers segmented each tumor. Radiomics features were extracted from the tumor and the peritumoral tissues. Partial least square (PLS) regression was used as the multivariate machine learning algorithm. PLS β-weights of radiomics features included the 5% features with the largest β-weights in magnitude (top 5%). Leave-one-out nested cross-validation (nCV) was used to achieve hyperparameter optimization and evaluate the generalizable performance of the procedure. The diagnostic performance of the radiomics model was assessed through receiver operating characteristic (ROC) analysis. A null hypothesis probability threshold of 5% was chosen (p < 0.05). The exploratory analysis for the complete dataset revealed an average absolute correlation among features of 0.51. The nCV framework delivered an AUC of 0.76 (p = 1.1∙10−3). When combining "early" and "peak" DCE images of only T or TST, a tendency toward statistical significance was obtained for TST with an AUC of 0.61 (p = 0.05). The 47 features included in the top 5% were balanced between T and TST (23 and 24, respectively). Moreover, 33/47 (70%) were texture-related, and 25/47 (53%) were derived from high-resolution images (1 mm). A radiomics-based machine learning approach shows the potential to accurately predict the recurrence risk in early ER + /HER2 − breast cancer patients. [ABSTRACT FROM AUTHOR]

Objectives: Establishing the reproducibility of expert-derived measurements on CTA exams of aortic dissection is clinically important and paramount for ground-truth determination for machine learning. Methods: Four independent observers retrospectively evaluated CTA exams of 72 patients with uncomplicated Stanford type B aortic dissection and assessed the reproducibility of a recently proposed combination of four morphologic risk predictors (maximum aortic diameter, false lumen circumferential angle, false lumen outflow, and intercostal arteries). For the first inter-observer variability assessment, 47 CTA scans from one aortic center were evaluated by expert-observer 1 in an unconstrained clinical assessment without a standardized workflow and compared to a composite of three expert-observers (observers 2–4) using a standardized workflow. A second inter-observer variability assessment on 30 out of the 47 CTA scans compared observers 3 and 4 with a constrained, standardized workflow. A third inter-observer variability assessment was done after specialized training and tested between observers 3 and 4 in an external population of 25 CTA scans. Inter-observer agreement was assessed with intraclass correlation coefficients (ICCs) and Bland-Altman plots. Results: Pre-training ICCs of the four morphologic features ranged from 0.04 (−0.05 to 0.13) to 0.68 (0.49–0.81) between observer 1 and observers 2–4 and from 0.50 (0.32–0.69) to 0.89 (0.78–0.95) between observers 3 and 4. ICCs improved after training ranging from 0.69 (0.52–0.87) to 0.97 (0.94–0.99), and Bland-Altman analysis showed decreased bias and limits of agreement. Conclusions: Manual morphologic feature measurements on CTA images can be optimized resulting in improved inter-observer reliability. This is essential for robust ground-truth determination for machine learning models. Key Points: • Clinical fashion manual measurements of aortic CTA imaging features showed poor inter-observer reproducibility. • A standardized workflow with standardized training resulted in substantial improvements with excellent inter-observer reproducibility. • Robust ground truth labels obtained manually with excellent inter-observer reproducibility are key to develop reliable machine learning models. [ABSTRACT FROM AUTHOR]

In this prospective, pilot study, we tested a kV-independent coronary artery calcium scoring CT protocol, using a novel reconstruction kernel (Sa36f). From December 2018 to November 2019, we performed an additional research scan in 61 patients undergoing clinical calcium scanning. For the standard protocol (120 kVp), images were reconstructed with a standard, medium-sharp kernel (Qr36d). For the research protocol (automated kVp selection), images were reconstructed with a novel kernel (Sa36f). Research scans were sequentially performed using a higher (cohort A, n = 31) and a lower (cohort B, n = 30) dose optimizer setting within the automatic system with customizable kV selection. Agatston scores, coronary calcium volumes, and radiation exposure of the standard and research protocol were compared. A phantom study was conducted to determine inter-scan variability. There was excellent correlation for the Agatston score between the two protocols (r = 0.99); however, the standard protocol resulted in slightly higher Agatston scores (29.4 [0–139.0] vs 17.4 [0–158.2], p = 0.028). The median calcium volumes were similar (11.5 [0–109.2] vs 11.2 [0–118.0] mm3; p = 0.176), and the number of calcified lesions was not significantly different (p = 0.092). One patient was reclassified to another risk category. The research protocol could be performed at a lower kV and resulted in a substantially lower radiation exposure, with a median volumetric CT dose index of 4.1 vs 5.2 mGy, respectively (p < 0.001). Our results showed that a consistent coronary calcium scoring can be achieved using a kV-independent protocol that lowers radiation doses compared to the standard protocol. Key Points: • The Sa36f kernel enables kV-independent Agatston scoring without changing the original Agatston weighting threshold. • Agatston scores and calcium volumes of the standard and research protocols showed an excellent correlation. • The research protocol resulted in a significant reduction in radiation exposure with a mean reduction of 22% in DLP and 25% in CTDIvol. [ABSTRACT FROM AUTHOR]

Objective: To (1) compare bladder cancer (BC) muscle invasiveness among three b-values using a contrast-free approach based on Vesical Imaging-Reporting and Data System (VI-RADS), to (2) determine if muscle-invasiveness assessment is affected by the reader experience, and to (3) compare BC conspicuity among three b-values, qualitatively and quantitatively. Methods: Thirty-eight patients who underwent a bladder MRI on a 3.0-T scanner were enrolled. The gold standard was histopathology report following transurethral resection of BC. Three sets of images, including T2w and different b-values for DWI, set 1 (b = 1000 s/mm2), set 2 (b = 1500 s/mm2), and set 3 (b = 2000 s/mm2), were reviewed by three differently experienced readers. Descriptive statistics and Intraclass Correlation Coefficient (ICC) were calculated. Comparisons among readers and DWI sets were performed with the Wilcoxon test. Receiver operating characteristic (ROC) analysis was performed. Areas under the curves (AUCs) and pairwise comparison were calculated. Results: AUCs of muscle-invasiveness assessment ranged from 0.896 to 0.984 (reader 1), 0.952–0.968 (reader 2), and 0.952–0.984 (reader 3) without significant differences among different sets and readers (p > 0.05). The mean conspicuity qualitative scores were higher in Set 1 (2.21–2.33), followed by Set 2 (2–2.16) and Set 3 (1.82–2.14). The quantitative conspicuity assessment showed that mean normalized intensity of tumor was significantly higher in Set 2 (4.217–4.737) than in Set 1 (3.923–4.492) and Set 3 (3.833–3.992) (p < 0.05). Conclusion: Muscle invasiveness can be assessed with high accuracy using a contrast-free protocol with T2W and DWI, regardless of reader's experience. b = 1500 s/mm2 showed the best tumor delineation, while b = 1000 s/mm2 allowed for better tumor–wall interface assessment. [ABSTRACT FROM AUTHOR]

Objectives: To determine whether single-phase dual-energy CT (DECT) differentiates vascular and nonvascular renal lesions in the portal venous phase (PVP). Optimal iodine threshold was determined and compared to Hounsfield unit (HU) measurements. Methods: We retrospectively included 250 patients (266 renal lesions) who underwent a clinically indicated PVP abdominopelvic CT on a rapid-kilovoltage-switching single-source DECT (rsDECT) or a dual-source DECT (dsDECT) scanner. Iodine concentration and HU measurements were calculated by four experienced readers. Diagnostic accuracy was determined using biopsy results and follow-up imaging as reference standard. Area under the curve (AUC) was calculated for each DECT scanner to differentiate vascular from nonvascular lesions and vascular lesions from hemorrhagic/proteinaceous cysts. Univariable and multivariable logistic regression analyses evaluated the association between variables and the presence of vascular lesions. Results: A normalized iodine concentration threshold of 0.25 mg/mL yielded high accuracy in differentiating vascular and nonvascular lesions (AUC 0.93, p < 0.001), with comparable performance to HU measurements (AUC 0.93). Both iodine concentration and HU measurements were independently associated with vascular lesions when adjusted for age, gender, body mass index, and lesion size (AUC 0.95 and 0.95, respectively). When combined, diagnostic performance was higher (AUC 0.96). Both absolute and normalized iodine concentrations performed better than HU measurements (AUC 0.92 vs. AUC 0.87) in differentiating vascular lesions from hemorrhagic/proteinaceous cysts. Conclusion: A single-phase (PVP) DECT scan yields high accuracy to differentiate vascular from nonvascular renal lesions. Iodine concentration showed a slightly higher performance than HU measurements in differentiating vascular lesions from hemorrhagic/proteinaceous cysts. Key Points: • A single-phase dual-energy CT scan in the portal venous phase differentiates vascular from nonvascular renal lesions with high accuracy (AUC 0.93). • When combined, iodine concentration and HU measurements showed the highest diagnostic performance (AUC 0.96) to differentiate vascular from nonvascular renal lesions. • Compared to HU measurements, iodine concentration showed a slightly higher performance in differentiating vascular lesions from hemorrhagic/proteinaceous cysts. [ABSTRACT FROM AUTHOR]

After an ischemic event, disruptive changes in the healthy myocardium may gradually develop and may ultimately turn into fibrotic scar. While these structural changes have been described by conventional imaging modalities mostly on a macroscopic scale—i.e., late gadolinium enhancement at magnetic resonance imaging (MRI)—in recent years, novel imaging methods have shown the potential to unveil an even more detailed picture of the postischemic myocardial phenomena. These new methods may bring advances in the understanding of ischemic heart disease with potential major changes in the current clinical practice. In this review article, we provide an overview of the emerging methods for the non-invasive characterization of ischemic heart disease, including coronary ultrafast Doppler angiography, photon-counting computed tomography (CT), micro-CT (for preclinical studies), low-field and ultrahigh-field MRI, and 11C-methionine positron emission tomography. In addition, we discuss new opportunities brought by artificial intelligence, while addressing promising future scenarios and the challenges for the application of artificial intelligence in the field of cardiac imaging. [ABSTRACT FROM AUTHOR]

Objectives: Currently, hurdles to implementation of artificial intelligence (AI) in radiology are a much-debated topic but have not been investigated in the community at large. Also, controversy exists if and to what extent AI should be incorporated into radiology residency programs. Methods: Between April and July 2019, an international survey took place on AI regarding its impact on the profession and training. The survey was accessible for radiologists and residents and distributed through several radiological societies. Relationships of independent variables with opinions, hurdles, and education were assessed using multivariable logistic regression. Results: The survey was completed by 1041 respondents from 54 countries. A majority (n = 855, 82%) expects that AI will cause a change to the radiology field within 10 years. Most frequently, expected roles of AI in clinical practice were second reader (n = 829, 78%) and work-flow optimization (n = 802, 77%). Ethical and legal issues (n = 630, 62%) and lack of knowledge (n = 584, 57%) were mentioned most often as hurdles to implementation. Expert respondents added lack of labelled images and generalizability issues. A majority (n = 819, 79%) indicated that AI should be incorporated in residency programs, while less support for imaging informatics and AI as a subspecialty was found (n = 241, 23%). Conclusions: Broad community demand exists for incorporation of AI into residency programs. Based on the results of the current study, integration of AI education seems advisable for radiology residents, including issues related to data management, ethics, and legislation. Key Points: • There is broad demand from the radiological community to incorporate AI into residency programs, but there is less support to recognize imaging informatics as a radiological subspecialty. • Ethical and legal issues and lack of knowledge are recognized as major bottlenecks for AI implementation by the radiological community, while the shortage in labeled data and IT-infrastructure issues are less often recognized as hurdles. • Integrating AI education in radiology curricula including technical aspects of data management, risk of bias, and ethical and legal issues may aid successful integration of AI into diagnostic radiology. [ABSTRACT FROM AUTHOR]

Objectives: Radiologists' perception is likely to influence the adoption of artificial intelligence (AI) into clinical practice. We investigated knowledge and attitude towards AI by radiologists and residents in Europe and beyond. Methods: Between April and July 2019, a survey on fear of replacement, knowledge, and attitude towards AI was accessible to radiologists and residents. The survey was distributed through several radiological societies, author networks, and social media. Independent predictors of fear of replacement and a positive attitude towards AI were assessed using multivariable logistic regression. Results: The survey was completed by 1,041 respondents from 54 mostly European countries. Most respondents were male (n = 670, 65%), median age was 38 (24–74) years, n = 142 (35%) residents, and n = 471 (45%) worked in an academic center. Basic AI-specific knowledge was associated with fear (adjusted OR 1.56, 95% CI 1.10–2.21, p = 0.01), while intermediate AI-specific knowledge (adjusted OR 0.40, 95% CI 0.20–0.80, p = 0.01) or advanced AI-specific knowledge (adjusted OR 0.43, 95% CI 0.21–0.90, p = 0.03) was inversely associated with fear. A positive attitude towards AI was observed in 48% (n = 501) and was associated with only having heard of AI, intermediate (adjusted OR 11.65, 95% CI 4.25–31.92, p < 0.001), or advanced AI-specific knowledge (adjusted OR 17.65, 95% CI 6.16–50.54, p < 0.001). Conclusions: Limited AI-specific knowledge levels among radiology residents and radiologists are associated with fear, while intermediate to advanced AI-specific knowledge levels are associated with a positive attitude towards AI. Additional training may therefore improve clinical adoption. Key Points: • Forty-eight percent of radiologists and residents have an open and proactive attitude towards artificial intelligence (AI), while 38% fear of replacement by AI. • Intermediate and advanced AI-specific knowledge levels may enhance adoption of AI in clinical practice, while rudimentary knowledge levels appear to be inhibitive. • AI should be incorporated in radiology training curricula to help facilitate its clinical adoption. [ABSTRACT FROM AUTHOR]

Ground-glass opacities (GGOs) are a non-specific high-resolution computed tomography (HRCT) finding tipically observed in early Coronavirus disesase 19 (COVID-19) pneumonia. However, GGOs are also seen in other acute lung diseases, thus making challenging the differential diagnosis. To this aim, we investigated the performance of a radiomics-based machine learning method to discriminate GGOs due to COVID-19 from those due to other acute lung diseases. Two sets of patients were included: a first set of 28 patients (COVID) diagnosed with COVID-19 infection confirmed by real-time polymerase chain reaction (RT-PCR) between March and April 2020 having (a) baseline HRCT at hospital admission and (b) predominant GGOs pattern on HRCT; a second set of 30 patients (nCOVID) showing (a) predominant GGOs pattern on HRCT performed between August 2019 and April 2020 and (b) availability of final diagnosis. Two readers independently segmented GGOs on HRCTs using a semi-automated approach, and radiomics features were extracted using a standard open source software (PyRadiomics). Partial least square (PLS) regression was used as the multivariate machine-learning algorithm. A leave-one-out nested cross-validation was implemented. PLS β-weights of radiomics features, including the 5% features with the largest β-weights in magnitude (top 5%), were obtained. The diagnostic performance of the radiomics model was assessed through receiver operating characteristic (ROC) analysis. The Youden's test assessed sensitivity and specificity of the classification. A null hypothesis probability threshold of 5% was chosen (p < 0.05). The predictive model delivered an AUC of 0.868 (Youden's index = 0.68, sensitivity = 93%, specificity 75%, p = 4.2 × 10–7). Of the seven features included in the top 5% features, five were texture-related. A radiomics-based machine learning signature showed the potential to accurately differentiate GGOs due to COVID-19 pneumonia from those due to other acute lung diseases. Most of the discriminant radiomics features were texture-related. This approach may assist clinician to adopt the appropriate management early, while improving the triage of patients. [ABSTRACT FROM AUTHOR]

Objectives: (1) To investigate whether a contrast-free biparametric MRI (bp-MRI) including T2-weighted images (T2W) and diffusion-weighted images (DWI) can be considered an accurate alternative to the standard multiparametric MRI (mp-MRI), consisting of T2, DWI, and dynamic contrast-enhanced (DCE) imaging for the muscle-invasiveness assessment of bladder cancer (BC), and (2) to evaluate how the diagnostic performance of differently experienced readers is affected according to the type of MRI protocol.Methods: Thirty-eight patients who underwent a clinically indicated bladder mp-MRI on a 3-T scanner were prospectively enrolled. Trans-urethral resection of bladder was the gold standard. Two sets of images, set 1 (bp-MRI) and set 2 (mp-MRI), were independently reviewed by four readers. Descriptive statistics, including sensitivity and specificity, were calculated for each reader. Receiver operating characteristic (ROC) analysis was performed, and the areas under the curve (AUCs) were calculated for the bp-MRI and the standard mp-MRI. Pairwise comparison of the ROC curves was performed.Results: The AUCs for bp- and mp-MRI were respectively 0.91-0.92 (reader 1), 0.90 (reader 2), 0.95-0.90 (reader 3), and 0.90-0.87 (reader 4). Sensitivity was 100% for both protocols and specificity ranged between 79.31 and 89.66% and between 79.31 and 83.33% for bp-MRI and mp-MRI, respectively. No significant differences were shown between the two MRI protocols (p > 0.05). No significant differences were shown accordingly to the reader's experience (p > 0.05).Conclusions: A bp-MRI protocol consisting of T2W and DWI has comparable diagnostic accuracy to the standard mp-MRI protocol for the detection of muscle-invasive bladder cancer. The experience of the reader does not significantly affect the diagnostic performance using VI-RADS.Key Points: • The contrast-free MRI protocol shows a comparable accuracy to the standard multiparametric MRI protocol in the bladder cancer muscle-invasiveness assessment. • VI-RADS classification helps non-expert radiologists to assess the muscle-invasiveness of bladder cancer. • DCE should be carefully interpreted by less experienced readers due to inflammatory changes representing a potential pitfall. [ABSTRACT FROM AUTHOR]

Purpose: To determine whether multiphasic dual-energy (DE) CT iodine quantitation correlates with the severity of chronic liver disease. Methods: We retrospectively included 40 cirrhotic and 28 non-cirrhotic patients who underwent a multiphasic liver protocol DECT. All three phases (arterial, portal venous (PVP), and equilibrium) were performed in DE mode. Iodine (I) values (mg I/ml) were obtained by placing regions of interest in the liver, aorta, common hepatic artery, and portal vein (PV). Iodine slopes (λ) were calculated as follows: (Iequilibrium–Iarterial)/time and (Iequilibrium–IPVP)/time. Spearman correlations between λ and MELD scores were evaluated, and the area under the curve of the receiver operating characteristic (AUROC) was calculated to distinguish cirrhotic and non-cirrhotic patients. Results: Cirrhotic and non-cirrhotic patients had significantly different λequilibrium-arterial [IQR] for the caudate (λ = 2.08 [1.39–2.98] vs 1.46 [0.76–1.93], P = 0.007), left (λ = 2.05 [1.50–2.76] vs 1.51 [0.59–1.90], P = 0.002) and right lobes (λ = 1.72 [1.12–2.50] vs 1.13 [0.41–0.43], P = 0.003) and for the PV (λ = 3.15 [2.20–5.00] vs 2.29 [0.85–2.71], P = 0.001). λequilibrium-PVP were significantly different for the right (λ = 0.11 [− 0.45–1.03] vs − 0.44 [− 0.83–0.12], P = 0.045) and left lobe (λ = 0.30 [− 0.25–0.98] vs − 0.10 [− 0.35–0.24], P = 0.001). Significant positive correlations were found between MELD scores and λequilibrium-arterial for the caudate lobe (ρ = 0.34, P = 0.004) and λequilibrium-PVP for the caudate (ρ = 0.26, P = 0.028) and right lobe (ρ = 0.33, P = 0.007). AUROC in distinguishing cirrhotic and non-cirrhotic patients were 0.72 (P = 0.002), 0.71 (P = 0.003), and 0.75 (P = 0.001) using λequilibrium-arterial for the left lobe, right lobe, and PV, respectively. The λequilibrium-PVP AUROC of the right lobe was 0.73 (P = 0.001). Conclusion: Multiphasic DECT iodine quantitation over time is significantly different between cirrhotic and non-cirrhotic patients, correlates with the MELD score, and it could potentially serve as a non-invasive measure of cirrhosis and disease severity with acceptable diagnostic accuracy. [ABSTRACT FROM AUTHOR]

Neoadjuvant chemo-radiotherapy (CRT) followed by total mesorectal excision (TME) represents the standard treatment for patients with locally advanced (≥ T3 or N+) rectal cancer (LARC). Approximately 15% of patients with LARC shows a complete response after CRT. The use of pre-treatment MRI as predictive biomarker could help to increase the chance of organ preservation by tailoring the neoadjuvant treatment. We present a novel machine learning model combining pre-treatment MRI-based clinical and radiomic features for the early prediction of treatment response in LARC patients. MRI scans (3.0 T, T2-weighted) of 72 patients with LARC were included. Two readers independently segmented each tumor. Radiomic features were extracted from both the "tumor core" (TC) and the "tumor border" (TB). Partial least square (PLS) regression was used as the multivariate, machine learning, algorithm of choice and leave-one-out nested cross-validation was used to optimize hyperparameters of the PLS. The MRI-Based "clinical-radiomic" machine learning model properly predicted the treatment response (AUC = 0.793, p = 5.6 × 10–5). Importantly, the prediction improved when combining MRI-based clinical features and radiomic features, the latter extracted from both TC and TB. Prospective validation studies in randomized clinical trials are warranted to better define the role of radiomics in the development of rectal cancer precision medicine. [ABSTRACT FROM AUTHOR]

Magnetic resonance imaging (MRI) plays a leading role in the non-invasive evaluation of bone marrow (BM). Normal BM pattern depends on the ratio and distribution of yellow and red marrow, which are subject to changes with age, pathologies, and treatments. Neonates show almost entirely red marrow. Over time, yellow marrow conversion takes place with a characteristic sequence leading to a red marrow persistence in proximal metaphyses of long bones. In adults, normal BM is composed of both red (40% water, 40% fat) and yellow marrow (15% water, 80% fat). Due to the higher content of fat, yellow marrow normally appears hyperintense on T1-weighted (T1w) fast spin echo (FSE) sequences and hypo-/iso-intense in short tau inversion recovery (STIR) T2-weighted (T2w); red marrow appears slightly hyperintense in T1w FSE and hyper-/iso-intense in STIR T2w. Pathologic BM has reduced fat and increased water percentages, resulting hypointense in T1w FSE and hyperintense in STIR T2w. In oncologic patients, BM MRI signal largely depends on the treatment (irradiation and/or chemotherapy) and its timing. BM fat and water amount and location in normal red/yellow and pathologic marrow are responsible for different signals in MRI sequences whose knowledge by radiologists may help to differentiate between normal and pathologic findings. Our aim was to discuss and illustrate the MRI of BM physiologic conversion and pathologic reconversion occurring in malignancies and after treatments in cancer patients. [ABSTRACT FROM AUTHOR]

Purpose: To evaluate the cost-effectiveness of DECT versus multiphasic CT and MRI for characterizing small incidentally detected indeterminate renal lesions using a Markov Monte Carlo decision-analytic model. Background: Incidental renal lesions are commonly encountered due to the increasing utilization of medical imaging and the increasing prevalence of renal lesions with age. Currently recommended imaging modalities to further characterize incidental indeterminate renal lesions have some inherent drawbacks. Single-phase DECT may overcome these limitations, but its cost-effectiveness remains uncertain. Materials and methods: A decision-analytic (Markov) model was constructed to estimate life expectancy and lifetime costs for otherwise healthy 64-year-old patients with small (≤ 4 cm) incidentally detected, indeterminate renal lesions on routine imaging (e.g., ultrasound or single-phase CT). Three strategies for evaluating renal lesions for enhancement were compared: multiphase SECT (e.g., true unenhanced and nephrographic phase), multiphasic MRI, and single-phase DECT (nephrographic phase in dual-energy mode). The model incorporated modality-specific diagnostic test performance, incidence, and prevalence of incidental renal cell carcinomas (RCCs), effectiveness, costs, and health outcomes. An incremental cost-effectiveness analysis was performed to identify strategy preference at willingness-to-pay (WTP) thresholds of $50,000 and $100,000 per quality-adjusted life-year (QALY) gained. Deterministic and probabilistic sensitivity analysis were performed. Results: In the base case analysis, expected mean costs per patient undergoing characterization of incidental renal lesions were $2567 for single-phase DECT, $3290 for multiphasic CT, and $3751 for multiphasic MRI. Associated quality-adjusted life-years were the highest for single-phase DECT at 0.962, for multiphasic MRI it was 0.940, and was the lowest for multiphasic CT at 0.925. Because of lower associated costs and higher effectiveness, the single-phase DECT strategy dominated the other two strategies. Conclusions: Single-phase DECT is potentially more cost-effective than multiphasic SECT and MRI for evaluating small incidentally detected indeterminate renal lesions. [ABSTRACT FROM AUTHOR]

The development of deep learning algorithms for complex tasks in digital medicine has relied on the availability of large labeled training datasets, usually containing hundreds of thousands of examples. The purpose of this study was to develop a 3D deep learning model, AppendiXNet, to detect appendicitis, one of the most common life-threatening abdominal emergencies, using a small training dataset of less than 500 training CT exams. We explored whether pretraining the model on a large collection of natural videos would improve the performance of the model over training the model from scratch. AppendiXNet was pretrained on a large collection of YouTube videos called Kinetics, consisting of approximately 500,000 video clips and annotated for one of 600 human action classes, and then fine-tuned on a small dataset of 438 CT scans annotated for appendicitis. We found that pretraining the 3D model on natural videos significantly improved the performance of the model from an AUC of 0.724 (95% CI 0.625, 0.823) to 0.810 (95% CI 0.725, 0.895). The application of deep learning to detect abnormalities on CT examinations using video pretraining could generalize effectively to other challenging cross-sectional medical imaging tasks when training data is limited. [ABSTRACT FROM AUTHOR]

Purpose: To compare tumor detectability and conspicuity of standard b = 1000 s/mm2 (b1000) versus ultrahigh b = 2000 s/mm2 (b2000) diffusion-weighted imaging (DWI) in rectal cancer. Methods: Fifty-five patients for a total of 81 3T DWI-MR scans were retrospectively evaluated by two differently experienced readers. A comparison between b1000 and b2000 for tumor detectability and conspicuity was performed. The conspicuity was qualitatively and quantitatively assessed by using three-point scale and whole tumor volume manual delineation, respectively. Receiver-operating characteristic curve (ROC) with area under the curve (AUC) analysis provided diagnostic accuracy in tumor detectability of restaging MR scans. Qualitative scores and quantitative features including mean signal intensity, variance, 10th percentile and 90th percentile, were compared using the Wilcoxon test. Interobserver agreement (IOA) for qualitative and quantitative data was calculated using Cohen's Kappa and intraclass correlation coefficient (ICC) respectively. Results: Diagnostic accuracy was comparable between b1000 and b2000 for both readers (p > 0.05). Overall quality scores were significantly better for b2000 than b1000 (2.29 vs 1.65 Reader 1, p = 0.01; 2.18 vs 1.69 Reader 2, p = 0.04). IOA was equally good for both b values (k = 0.86 b1000, k = 0.86 b2000). Quantitative analysis revealed more uniform signal (measured in variance) of b2000 in both healthy surrounding tissue (p < 0.05) and tumor (p < 0.05), with less outliers (measured using 10th and 90th percentile). Additionally, b2000 offered lower mean signal intensity in tissue sorrounding the tumor (p < 0.05). Finally, ICC improved from 0.92 (b1000) to 0.97 (b2000). Conclusion: Ultrahigh b value (b2000) may improve rectal cancer conspicuity and introbserver agreement maintaining comparable diagnostic accuracy to standard b1000. [ABSTRACT FROM AUTHOR]

Purpose: To investigate the diagnostic accuracy of MRI for placenta accreta spectrum (PAS) and clinical outcome prediction in women with placenta previa, using a novel MRI-based predictive model. Methods: Thirty-eight placental MRI exams performed on a 1.5T scanner were retrospectively reviewed by two radiologists in consensus. The presence of T2 dark bands, myometrial thinning, abnormal vascularity, uterine bulging, placental heterogeneity, placental protrusion sign, placental recess, and percretism signs was scored using a 5-point scale. Pathology and clinical intrapartum findings were the standard of reference for PAS, while intrapartum/peripartum bleeding and emergency hysterectomy defined the clinical outcome. Receiver-operating characteristic (ROC) analysis and discriminant function analysis were performed to test the predictive power of MRI findings for both PAS and clinical outcome prediction. Results: Abnormal vascularity and percretism signs were the two most predictive MRI features of PAS. The area under the curve (AUC) of the predictive function was 0.833 (cutoff 0.39, 67% sensitivity, 100% specificity, p = 0.001). Percretism signs and myometrial thinning were the two most predictive MRI features of poor outcome. AUC of the predictive function was 0.971 (cutoff − 0.55, 100% sensitivity, 77% specificity, p < 0.001). Conclusion: The diagnostic accuracy of MRI, especially considering the combination of the most predictive MRI findings, is higher when the target of the prediction is the clinical outcome rather than the PAS. [ABSTRACT FROM AUTHOR]

To develop a quantitative T1-mapping-based synthetic inversion recovery (IRsynth) approach to calculate the optimal inversion time (TI0) for late gadolinium enhancement (LGE) imaging. Prospectively enrolled patients (n = 130, 58 ± 16 years) underwent cardiac MRI on a 1.5T system including Look-Locker TI-scout (LL), modified LL IR (MOLLI)-based T1-mapping, and LGE acquisitions. Patients were randomized into two groups: LL group (TI-scout followed T1-mapping) or MOLLI group (T1-mapping followed TI-scout). In both groups, the second acquisition was used to determine the TI0 for LGE. IRsynth images were generated from T1-maps between TI = 200-400 ms in 5 ms increments. Image quality was rated on a 3-point scale and the remote/background signal intensity ratio (SIR) was calculated. In the LL group (n = 53), the TI-scout-based TI0 was significantly shorter compared to IRsynth [230 ms (219-242) vs. 280 ms (263-297), P < 0.0001]. The TI0 used for LGE was set 30-40 ms longer [261 ms (247-276), P < 0.0001] than the TI-scout-based TI0, resulting in a TI0 ~ 20 ms shorter than what was obtained by IRsynth (P = 0.0156). In the MOLLI group (n = 63), IRsynth-based TI0 was significantly longer than the TI-scout-based TI0 [298 ms (262-334) vs. 242 ms (217-267), P = 0.0313]. The quality of myocardial nulling was rated higher [2.4 (2.2-2.5) vs. 2.0 (1.8-2.1), P = 0.0042] and the remote/background SIR was found to be more optimal (1.6 [1.1-2.1] vs. 2.6 [1.8-3.3], P = 0.0256) in the MOLLI group. T1-based IRsynth selects TI0 for LGE more accurately than conventional TI-scout imaging. IRsynth improves TI0 selection by providing excellent visualization of the representative image contrast for LGE images, reducing operator dependence in LGE acquisition. [ABSTRACT FROM AUTHOR]

Diagnosis of endoleak following endovascular aortic repair (EVAR) relies on manual review of multi-slice CT angiography (CTA) by physicians which is a tedious and time-consuming process that is susceptible to error. We evaluate the use of a deep neural network for the detection of endoleak on CTA for post-EVAR patients using a novel data efficient training approach. 50 CTAs and 20 CTAs with and without endoleak respectively were identified based on gold standard interpretation by a cardiovascular subspecialty radiologist. The Endoleak Augmentor, a custom designed augmentation method, provided robust training for the machine learning (ML) model. Predicted segmentation maps underwent post-processing to determine the presence of endoleak. The model was tested against 3 blinded general radiologists and 1 blinded subspecialist using a held-out subset (10 positive endoleak CTAs, 10 control CTAs). Model accuracy, precision and recall for endoleak diagnosis were 95%, 90% and 100% relative to reference subspecialist interpretation (AUC = 0.99). Accuracy, precision and recall was 70/70/70% for generalist1, 50/50/90% for generalist2, and 90/83/100% for generalist3. The blinded subspecialist had concordant interpretations for all test cases compared with the reference. In conclusion, our ML-based approach has similar performance for endoleak diagnosis relative to subspecialists and superior performance compared with generalists. [ABSTRACT FROM AUTHOR]