Search

Objectives: To investigate the differentiation of premalignant from benign colorectal polyps detected by CT colonography using deep learning., Methods: In this retrospective analysis of an average risk colorectal cancer screening sample, polyps of all size categories and morphologies were manually segmented on supine and prone CT colonography images and classified as premalignant (adenoma) or benign (hyperplastic polyp or regular mucosa) according to histopathology. Two deep learning models SEG and noSEG were trained on 3D CT colonography image subvolumes to predict polyp class, and model SEG was additionally trained with polyp segmentation masks. Diagnostic performance was validated in an independent external multicentre test sample. Predictions were analysed with the visualisation technique Grad-CAM++., Results: The training set consisted of 107 colorectal polyps in 63 patients (mean age: 63 ± 8 years, 40 men) comprising 169 polyp segmentations. The external test set included 77 polyps in 59 patients comprising 118 polyp segmentations. Model SEG achieved a ROC-AUC of 0.83 and 80% sensitivity at 69% specificity for differentiating premalignant from benign polyps. Model noSEG yielded a ROC-AUC of 0.75, 80% sensitivity at 44% specificity, and an average Grad-CAM++ heatmap score of ≥ 0.25 in 90% of polyp tissue., Conclusions: In this proof-of-concept study, deep learning enabled the differentiation of premalignant from benign colorectal polyps detected with CT colonography and the visualisation of image regions important for predictions. The approach did not require polyp segmentation and thus has the potential to facilitate the identification of high-risk polyps as an automated second reader., Key Points: • Non-invasive deep learning image analysis may differentiate premalignant from benign colorectal polyps found in CT colonography scans. • Deep learning autonomously learned to focus on polyp tissue for predictions without the need for prior polyp segmentation by experts. • Deep learning potentially improves the diagnostic accuracy of CT colonography in colorectal cancer screening by allowing for a more precise selection of patients who would benefit from endoscopic polypectomy, especially for patients with polyps of 6-9 mm size., (© 2022. The Author(s).)

Objectives: To investigate the prediction of 1-year survival (1-YS) in patients with metastatic colorectal cancer with use of a systematic comparative analysis of quantitative imaging biomarkers (QIBs) based on the geometric and radiomics analysis of whole liver tumor burden (WLTB) in comparison to predictions based on the tumor burden score (TBS), WLTB volume alone, and a clinical model., Methods: A total of 103 patients (mean age: 61.0 ± 11.2 years) with colorectal liver metastases were analyzed in this retrospective study. Automatic segmentations of WLTB from baseline contrast-enhanced CT images were used. Established biomarkers as well as a standard radiomics model building were used to derive 3 prognostic models. The benefits of a geometric metastatic spread (GMS) model, the Aerts radiomics prior model of the WLTB, and the performance of TBS and WLTB volume alone were assessed. All models were analyzed in both statistical and predictive machine learning settings in terms of AUC., Results: TBS showed the best discriminative performance in a statistical setting to discriminate 1-YS (AUC = 0.70, CI: [0.56, 0.90]). For the machine learning-based prediction for unseen patients, both a model of the GMS of WLTB (0.73, CI: [0.60, 0.84]) and the Aerts radiomics prior model (0.76, CI: [0.65, 0.86]) applied on the WLTB showed a numerically higher predictive performance than TBS (0.68, CI: [0.54, 0.79]), radiomics (0.65, CI: [0.55, 0.78]), WLTB volume alone (0.53, CI: [0.40. 0.66]), or the clinical model (0.56, CI: [0.43, 0.67])., Conclusions: The imaging-based GMS model may be a first step towards a more fine-grained machine learning extension of the TBS concept for risk stratification in mCRC patients without the vulnerability to technical variance of radiomics., Key Points: • CT-based geometric distribution and radiomics analysis of whole liver tumor burden in metastatic colorectal cancer patients yield prognostic information. • Differences in survival are possibly attributable to the spatial distribution of metastatic lesions and the geometric metastatic spread analysis of all liver metastases may serve as robust imaging biomarker invariant to technical variation. • Imaging-based prediction models outperform clinical models for 1-year survival prediction in metastatic colorectal cancer patients with liver metastases.

Objectives: To assess the value of a T1-3D black-blood turbo spin echo (TSE) sequence for the diagnosis of abdominal large vessel vasculitis (LVV)., Materials and Methods: The study included 20 patients with abdominal LVV and 17 controls, who underwent a 3T-MRI scan using a modified T1-3D volumetric isotropic TSE acquisition and a segmented T1-3D turbo field echo sequence (T1-mVISTA/T1-eTHRIVE). Two radiologists independently analyzed the aorta for concentric contrast enhancement, concentric wall thickening, image quality, and flow artifact intensity (CCE/CWT/IQ/FAI; 4-point scales). The mean aortic wall thickness (MAWT) in post-contrast T1-mVISTA was compared between patients and controls., Results: IQ of T1-mVISTA was rated good to excellent in 91.5% of 282 evaluated vessel segments with no or minor FAI present in 85.5%. The inter-observer reproducibility for the identification of CCE/CWT on T1-mVISTA was 0.92 and 0.93 (p < 0.001). The distribution of segmental inflammation in T1-mVISTA significantly correlated with T1-eTHRIVE (CCE, κ = 0.768; CWT, κ = 0.715; p < 0.001), resulting in a sensitivity, specificity, and positive predictive value of 100%, 81.3%, and 83.3%. The MAWT significantly differed between patients and controls (3.29 ± 0.81 vs. 2.24 ± 0.45 mm; p < 0.001)., Conclusions: T1-mVISTA enables the evaluation of the MAWT and allows the detection of abdominal LVV., Key Points: • 3D T1w-mVISTA accurately depicted the large abdominal vessels. • 3D T1w-mVISTA enables accurate measurements of the abdominal aortic wall thickness. • 3D T1w-mVISTA is useful for the detection of abdominal LVV.

Objectives: To evaluate the feasibility of T1w-3D black-blood turbo spin echo (TSE) sequence with variable flip angles for the diagnosis of thoracic large vessel vasculitis (LVV)., Methods: Thirty-five patients with LVV, diagnosed according to the current standard of reference, and 35 controls were imaged at 3.0T using 1.2 × 1.3 × 2.0 mm 3 fat-suppressed, T1w-3D, modified Volumetric Isotropic TSE Acquisition (mVISTA) pre- and post-contrast. Applying a navigator and peripheral pulse unit triggering (PPU), the total scan time was 10-12 min. Thoracic aorta and subclavian and pulmonary arteries were evaluated for image quality (IQ), flow artefact intensity, diagnostic confidence, concentric wall thickening and contrast enhancement (CWT, CCE) using a 4-point scale., Results: IQ was good in all examinations (3.25 ± 0.72) and good to excellent in 342 of 408 evaluated segments (83.8 %), while 84.1 % showed no or minor flow artefacts. The interobserver reproducibility for the identification of CCE and CWT was 0.969 and 0.971 (p < 0.001) with an average diagnostic confidence of 3.47 ± 0.64. CCE and CWT were strongly correlated (Cohen's k = 0.87; P < 0.001) and significantly more frequent in the LVV-group (52.8 % vs. 1.0 %; 59.8 % vs. 2.4 %; P < 0.001)., Conclusions: Navigated fat-suppressed T1w-3D black-blood MRI with PPU-triggering allows diagnosis of thoracic LVV., Key Points: • Cross-sectional imaging is frequently applied in the diagnosis of LVV. • Navigated, PPU-triggered, T1w-3D mVISTA pre- and post contrast takes 10-12 min. • In this prospective, single-centre study, T1w-3D mVISTA accurately depicted large thoracic vessels. • T1w-3D mVISTA visualized CWT/CCW as correlates of mural inflammation in LVV. • T1w-3D mVISTA might be an alternative diagnostic tool without ionizing radiation.