Inferring FDG-PET-positivity of lymph node metastases in proven lung cancer from contrast-enhanced CT using radiomics and machine learning.

BACKGROUND: We evaluated the role of radiomics applied to contrast-enhanced computed tomography (CT) in the detection of lymph node (LN) metastases in patients with known lung cancer compared to 18F-fluorodeoxyglucose positron emission tomography (PET)/CT as a reference. METHODS: This retrospective analysis included 381 patients with 1,799 lymph nodes (450 malignant, 1,349 negative). The data set was divided into a training and validation set. A radiomics analysis with 4 filters and 6 algorithms resulting in 24 different radiomics signatures and a bootstrap algorithm (Bagging) with 30 bootstrap iterations was performed. A decision curve analysis was applied to generate a net benefit to compare the radiomics signature to two expert radiologists as one-by-one and as a prescreening tool in combination with the respective radiologist and only the radiologists. RESULTS: All 24 modeling methods showed good and reliable discrimination for malignant/benign LNs (area under the curve 0.75-0.87). The decision curve analysis showed a net benefit for the least absolute shrinkage and selection operator (LASSO) classifier for the entire probability range and outperformed the expert radiologists except for the high probability range. Using the radiomics signature as a prescreening tool for the radiologists did not improve net benefit. CONCLUSIONS: Radiomics showed good discrimination power irrespective of the modeling technique in detecting LN metastases in patients with known lung cancer. The LASSO classifier was a suitable diagnostic tool and even outperformed the expert radiologists, except for high probabilities. Radiomics failed to improve clinical benefit as a prescreening tool.

Use of TDI during MRI/US fusion-guided biopsy for suspected prostate cancer.

BACKGROUND: Tissue Doppler imaging (TDI) uses the Doppler principle to quantify the movement of biological tissues. OBJECTIVE: To investigate the contribution of TDI parameters derived during magnetic resonance imaging and ultrasound (MRI/US) fusion-guided biopsy for prostate cancer (PCa) discrimination. METHODS: From March 2016 to Dec. 2018, 75 men with suspected PCa prospectively underwent fusion-guided prostate biopsy. TDI overlaid on predefined target lesion were compared to the confirmed contralateral tumor-free area of the prostate gland (using Image J). Diagnostic value of TDI parameters was assessed using histopathology as standard of reference. RESULTS: Thirty-seven patients were diagnosed with PCa (49.3%), among them 27 with clinically significant PCa (Gleason score >  3 + 3 = 6 (ISUP 1). The LES/REF ratio was lower in confirmed PCa patients compared to patients without PCa (0.42, IQR, 0.22-0.59 vs. 0.52, IQR, 0.40-0.72, p = 0.017). TDI parameters allowed differentiation of low-risk from high-to-intermediate-risk PCa (ISUP 2 versus ISUP 3) based on lower pixel counts within the target ROI (1340, IQR 596-2430 vs. 2687, IQR 2453-3216, p = 0.004), lower pixel percentage (16.4 IQR 11.4-29.5 vs. 27.3, IQR 22.1-39.5; p = 0.005), and lower LES/REF ratios (0.29, IQR 0.19-0.51 vs. 0.52, IQR 0.47-0.74, p = 0.001). CONCLUSION: TDI of prostate lesions prelocated by MRI discriminates between cancerous and noncancerous lesions and further seems to enable characterization of PCa aggressiveness. This widely available US technique may improve confidence in target lesion localization for tissue sampling.

Diagnostic performance of PI-RADS version 2.1 compared to version 2.0 for detection of peripheral and transition zone prostate cancer.

The purpose of this study is to compare diagnostic performance of Prostate Imaging Reporting and Data System (PI-RADS) version (v) 2.1 and 2.0 for detection of Gleason Score (GS) ≥ 7 prostate cancer on MRI. Three experienced radiologists provided PI-RADS v2.0 scores and at least 12 months later v2.1 scores on lesions in 333 prostate MRI examinations acquired between 2012 and 2015. Diagnostic performance was assessed retrospectively by using MRI/transrectal ultrasound fusion biopsy and 10-core systematic biopsy as the reference. From a total of 359 lesions, GS ≥ 7 tumor was present in 135 lesions (37.60%). Area under the ROC curve (AUC) revealed slightly lower values for peripheral zone (PZ) and transition zone (TZ) scoring in v2.1, but these differences did not reach statistical significance. A significant number of score 2 lesions in the TZ were downgraded to score 1 in v2.1 showing 0% GS ≥ 7 tumor (0/11). The newly introduced diffusion-weighted imaging (DWI) upgrading rule in v2.1 was applied in 6 lesions from a total of 143 TZ lesions (4.2%). In summary, PI-RADS v2.1 showed no statistically significant differences in overall diagnostic performance of TZ and PZ scoring compared to v2.0. Downgraded BPH nodules showed favorable cancer frequencies. The new DWI upgrading rule for TZ lesions was applied in only few cases.

Contrast-Enhanced Ultrasound (CEUS) and Quantitative Perfusion Analysis in Patients with Suspicion for Prostate Cancer.

PURPOSE: The aim of this study was to investigate contrast-enhanced ultrasound (CEUS) parameters acquired by software during magnetic resonance imaging (MRI) US fusion-guided biopsy for prostate cancer (PCa) detection and discrimination. MATERIALS AND METHODS: From 2012 to 2015, 158 out of 165 men with suspicion for PCa and with at least 1 negative biopsy of the prostate were included and underwent a multi-parametric 3 Tesla MRI and an MRI/US fusion-guided biopsy, consecutively. CEUS was conducted during biopsy with intravenous bolus application of 2.4 mL of SonoVue® (Bracco, Milan, Italy). In the latter CEUS clips were investigated using quantitative perfusion analysis software (VueBox, Bracco). The area of strongest enhancement within the MRI pre-located region was investigated and all available parameters from the quantification tool box were collected and analyzed for PCa and its further differentiation was based on the histopathological results. RESULTS: The overall detection rate was 74 (47 %) PCa cases in 158 included patients. From these 74 PCa cases, 49 (66 %) were graded Gleason ≥ 3 + 4 = 7 (ISUP ≥ 2) PCa. The best results for cancer detection over all quantitative perfusion parameters were rise time (p = 0.026) and time to peak (p = 0.037). Within the subgroup analysis (> vs ≤ 3 + 4 = 7a (ISUP 2)), peak enhancement (p = 0.012), wash-in rate (p = 0.011), wash-out rate (p = 0.007) and wash-in perfusion index (p = 0.014) also showed statistical significance. CONCLUSION: The quantification of CEUS parameters was able to discriminate PCa aggressiveness during MRI/US fusion-guided prostate biopsy.

Maximizing Information From Routine Staging Computed Tomography in Functional Neuroendocrine Neoplasms: Are There Findings Indicating the Presence of Carcinoid Heart Disease?

PURPOSE: The aim of this study was to evaluate signs of right-sided heart dysfunction on staging computed tomography (CT) as indirect indicators of carcinoid heart disease. PATIENTS AND METHODS: Patients with functionally active neuroendocrine neoplasm and different grades of tricuspid valve regurgitation (TR) were identified. Two readers independently reviewed contrast-enhanced staging CT performed within 90 days before or after echocardiography. Logistic regression and receiver operating analyses were used to asses the predictive value of right ventricle-left ventricle ratio (RV-LV ratio), ventricular septal bowing, retrograde contrast filling of the hepatic veins during contrast injection, and time to aortal enhancement greater than 100 Hounsfield units during bolus tracking for TR. RESULTS: Forty-four examinations were evaluated (11 with TR = 0, 16 with TR = 1, 9 with TR = 2, and 8 with TR = 3). Right ventricle-LV ratio was found to predict TR less than or equal to 1 versus TR greater than 1 (P = 0.0188) and TR less than or equal to 1 versus TR equals 2 (P = 0.0082). A prolonged time to aortal enhancement greater than 100 Hounsfield units during bolus tracking predicted TR less than or equal to 1 versus TR greater than 1 (P = 0.0077). Area under the curve for RV-LV ratio was 0.86 when differentiating TR less than or equal to 1 versus TR equals 2. With a cutoff of 1.07, sensitivity was 0.89, and specificity was 0.72. CONCLUSIONS: In patients with functionally active neuroendocrine neoplasm, an RV-LV ratio of more than 1.07 predicted TR with a relatively high sensitivity and moderate specificity and thus could serve as an indicator of subclinical carcinoid heart disease on routine staging CT.