Deep learning in magnetic resonance enterography for Crohn's disease assessment: a systematic review.

Crohn's disease (CD) poses significant morbidity, underscoring the need for effective, non-invasive inflammatory assessment using magnetic resonance enterography (MRE). This literature review evaluates recent publications on the role of deep learning in improving MRE for CD assessment. We searched MEDLINE/PUBMED for studies that reported the use of deep learning algorithms for assessment of CD activity. The study was conducted according to the PRISMA guidelines. The risk of bias was evaluated using the QUADAS-2 tool. Five eligible studies, encompassing 468 subjects, were identified. Our study suggests that diverse deep learning applications, including image quality enhancement, bowel segmentation for disease burden quantification, and 3D reconstruction for surgical planning are useful and promising for CD assessment. However, most of the studies are preliminary, retrospective studies, and have a high risk of bias in at least one category. Future research is needed to assess how deep learning can impact CD patient diagnostics, particularly when considering the increasing integration of such models into hospital systems.

Jejunal Inflammation in Crohn's Disease: Comparison between Diffusion Weighted Magnetic Resonance Imaging and Video Capsule Endoscopy.

BACKGROUND: Jejunal disease is associated with worse prognosis in Crohn's disease. The added value of diffusion weighted imaging for evaluating jejunal inflammation related to Crohn's Disease is scarce. OBJECTIVES: To compare diffusion weighted imaging, video capsule endoscopy, and inflammatory biomarkers in the assessment of Crohn's disease involving the jejunum. METHODS: Crohn's disease patients in clinical remission were prospectively recruited and underwent magnetic resonance (MR)-enterography and video capsule endoscopy. C-reactive protein and fecal-calprotectin levels were obtained. MR-enterography images were evaluated for restricted diffusion, and apparent diffusion coefficient values were measured. The video capsule endoscopy-based Lewis score was calculated. Associations between diffusion weighted imaging, apparent diffusion coefficient, Lewis score, and inflammatory biomarkers were evaluated. RESULTS: The study included 51 patients, and 27/51 (52.9%) with video capsule endoscopies showed jejunal mucosal inflammation. Sensitivity and specificity of restricted diffusion for video capsule endoscopy mucosal inflammation were 59.3% and 37.5% for the first reader, and 66.7% and 37.5% for the second reader, respectively. Diffusion weighted imaging was not statistically associated with jejunal video capsule endoscopy inflammation (P = 0.813). CONCLUSIONS: Diffusion weighted imaging was not an effective test for evaluation of jejunal inflammation as seen by video capsule endoscopy in patients with quiescent Crohn's disease.

The Liver Tumor Segmentation Benchmark (LiTS).

In this work, we report the set-up and results of the Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI) 2017 and the International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2017 and 2018. The image dataset is diverse and contains primary and secondary tumors with varied sizes and appearances with various lesion-to-background levels (hyper-/hypo-dense), created in collaboration with seven hospitals and research institutions. Seventy-five submitted liver and liver tumor segmentation algorithms were trained on a set of 131 computed tomography (CT) volumes and were tested on 70 unseen test images acquired from different patients. We found that not a single algorithm performed best for both liver and liver tumors in the three events. The best liver segmentation algorithm achieved a Dice score of 0.963, whereas, for tumor segmentation, the best algorithms achieved Dices scores of 0.674 (ISBI 2017), 0.702 (MICCAI 2017), and 0.739 (MICCAI 2018). Retrospectively, we performed additional analysis on liver tumor detection and revealed that not all top-performing segmentation algorithms worked well for tumor detection. The best liver tumor detection method achieved a lesion-wise recall of 0.458 (ISBI 2017), 0.515 (MICCAI 2017), and 0.554 (MICCAI 2018), indicating the need for further research. LiTS remains an active benchmark and resource for research, e.g., contributing the liver-related segmentation tasks in http://medicaldecathlon.com/. In addition, both data and online evaluation are accessible via https://competitions.codalab.org/competitions/17094.

Diagnostic significance of mesenteric lymph node involvement in proximal small bowel Crohn's disease.

Background: The diagnosis of proximal small bowel involvement in Crohn's disease (CD) can be challenging at magnetic resonance enterography (MRE). The inflammatory process in CD can be associated with peri-intestinal inflammatory reactions, including the presence of inflamed mesenteric lymph nodes. Objectives: To evaluate the significance of inflamed mesenteric lymph nodes adjacent to the jejunum at MRE in CD and the association with proximal bowel disease as detected by video capsule endoscopy (VCE). Design: This retrospective study was performed in two tertiary medical centres, and included 64 patients with CD who underwent MRE as well as VCE within 1 year. Methods: Data were collected for examinations performed between August 2013 and February 2021. MRE images were independently reviewed by radiologists who were blinded to the clinical data. Association between the presence of mesenteric lymph nodes adjacent to jejunum at MRE and disease activity according to VCE Lewis scores of proximal small bowel was examined. Results: VCE detected proximal disease in 24/64 patients (37.5%). Presence of regional lymph nodes in the jejunal mesentery was significantly associated with jejunal disease as seen on VCE (p < 0.001). Of the 20 patients who had proximal mesenteric lymph nodes at MRE, 15 (75%) had jejunal disease at VCE (sensitivity, 62.5%; specificity, 87.5%; and negative and positive predictive values, 79.5 and 75%, respectively). The number of regional lymph nodes was positively correlated with jejunal disease (mean: 2.63 ± 2.90 versus 0.78 ± 2.60, p = 0.01). Other MRE features of lymph nodes were not significantly predictive of jejunal CD. Conclusion: In patients with CD, inflamed regional lymph nodes in the jejunal mesentery at MRE can be valuable to suggest proximal small bowel disease, even when bowel wall features at imaging do not suggest disease involvement. Plain language summary: The diagnosis of proximal small bowel involvement in Crohn's disease (CD) can be challenging at magnetic resonance enterography (MRE). We analysed MRE examinations in patients with CD for the presence of lymph nodes adjacent to the proximal small bowel. We included 64 patients with CD who had MRE examinations and video capsule endoscopy (VCE) examinations within 1 year. Of 64 patients, 24 had proximal small bowel disease according to VCE. We found that of 20 patients who had regional mesenteric lymph nodes in the jejunal mesentery at MRE, 15 had proximal bowel disease involvement. We also found that patients with jejunal disease had a larger number of regional lymph nodes compared to patients without jejunal disease. All but one patient had normal appearing bowel at MRE. But, using regional mesenteric lymphadenopathy at MRE as an indicator for disease, 15/24 (62.5%) patients with proximal small bowel disease were detected. We therefore conclude that regional mesenteric lymph nodes assessment at MRE can aid diagnose proximal bowel disease, even when the proximal bowel looks normal at imaging. Presence of proximal mesenteric lymph nodes at MRE in patients with CD possibly warrant further investigation of the proximal small bowel by endoscopic measures.

Magnetic resonance enterography in Crohn's disease patients: current state of the art and future perspectives.

INTRODUCTION: Crohn's disease (CD) is a major concern due to relatively high incidence and major complications like stricture or fistulas, often requiring surgical treatment. In recent years, magnetic resonance enterography (MRE) became a popular method of diagnosis and disease surveillance. The purpose of this review is to summarize and discuss the major and most recent advances in various aspects of MRE usage in diagnosong Crohn's disease, and to discuss advances in technique, disease activity monitoring and response to treatment. METHODS: A literature search was performed and relevant publications were included, with emphasis on articles from the past decade. AREAS COVERED: In this review we have presented articles with major advances in the field of MRE of CD patients such as proper sequence selection, recent advances in scoring of disease activity, differentiation between inflammation and fibrosis, response to treatment and technological advances such as the use of AI. EXPERT OPINION: The main goal in improving MRE performance will be sequence selection aimed at differenting between inflammation and stricture while shortening the study length adjusted to patient compliance, and developing a standardized scoring system for MRE reporting assisted by artificial intelligence.

Qualitative sonographic assessment of transmural ileal inflammation in Crohn's disease: a comparison with MRI activity score.

BACKGROUND: Intestinal ultrasound (IUS) is an accurate tool for monitoring Crohn's disease. To date, there is no clinically used validated quantitative ultrasonographic score for assessing disease activity. For magnetic resonance enterography (MRE), the magnetic resonance index of activity (MaRIA) is most used. The goal of this study was to devise a new quantitative IUS score for assessing Crohn's disease inflammation, by using a partial MaRIA score as a reference. METHODS: This was a retrospective cohort study. The study cohort included patients with Crohn's disease followed between January 2016 and December 2018. Inclusion criteria were age >18 and <3 months between MRE and IUS. Linear/logistic regression was performed for the correlation of ultrasonographic parameters with MaRIA score. Ultrasonograpic features included: bowel wall thickness, disrupted bowel wall stratification, mesenteric fat proliferation, presence of lymph nodes, hypervascularity present on color Doppler flow, and the presence of complications (strictures, inflammatory mass, and fistula). RESULTS: Forty-two patients were included. A stepwise multiple regression model was constructed to predict MaRIA score using ultrasound features. Two variables were found to be independently significant: terminal ileum (TI) thickness (r = 0.68, P = 0.001) and mesenteric fat proliferation (r = 0.45, P = 0.019). A model was constructed as follows: MaRIA = 7 + 2.5 * TI US thickness (mm) + 7 * US fat proliferation (0 = no, 1 = yes). This model has an R2 of 0.51 for explaining the variability in the results. CONCLUSIONS: IUS measurements are significantly correlated with MaRIA score in the terminal ileum and a simple computational model can be constructed.

Automated Detection of Crohn's Disease Intestinal Strictures on Capsule Endoscopy Images Using Deep Neural Networks.

BACKGROUND AND AIMS: Passable intestinal strictures are frequently detected on capsule endoscopy [CE]. Such strictures are a major component of inflammatory scores. Deep neural network technology for CE is emerging. However, the ability of deep neural networks to identify intestinal strictures on CE images of Crohn's disease [CD] patients has not yet been evaluated. METHODS: We tested a state-of-the-art deep learning network for detecting CE images of strictures. Images of normal mucosa, mucosal ulcers, and strictures of Crohn's disease patients were retrieved from our previously described CE image bank. Ulcers were classified as per degree of severity. We performed 10 cross-validation experiments. A clear patient-level separation was maintained between training and testing sets. RESULTS: Overall, the entire dataset included 27 892 CE images: 1942 stricture images, 14 266 normal mucosa images, and 11 684 ulcer images [mild: 7075, moderate: 2386, severe: 2223]. For classifying strictures versus non-strictures, the network exhibited an average accuracy of 93.5% [±6.7%]. The network achieved excellent differentiation between strictures and normal mucosa (area under the curve [AUC] 0.989), strictures and all ulcers [AUC 0.942], and between strictures and different grades of ulcers [for mild, moderate, and severe ulcers-AUCs 0.992, 0.975, and 0.889, respectively]. CONCLUSIONS: Deep neural networks are highly accurate in the detection of strictures on CE images in Crohn's disease. The network can accurately separate strictures from ulcers across the severity range. The current accuracy for the detection of ulcers and strictures by deep neural networks may allow for automated detection and grading of Crohn's disease-related findings on CE.

Serum MMP-9: a novel biomarker for prediction of clinical relapse in patients with quiescent Crohn's disease, a post hoc analysis.

BACKGROUND: Matrix metalloproteinase-9 (MMP-9) is a novel marker of intestinal inflammation. The aim of this study was to assess if serum MMP-9 levels predict clinical flare in patients with quiescent Crohn's disease (CD). METHODS: This study was a post hoc analysis of a prospective observational study in which quiescent CD patients were included and followed until clinical relapse or the end of a 2-year follow-up period. Serial C-reactive protein (CRP) and fecal calprotectin (FC) levels were measured, and the patients underwent repeated capsule endoscopies (CEs) every 6 months. Small bowel inflammation was quantified by Lewis score (LS) for CE. A baseline magnetic resonance enterography was also performed, and MaRIA score was calculated. Serum MMP-9 levels in baseline blood samples were quantified by ELISA. RESULTS: Out of 58 eligible enrolled patients, 16 had a flare. Higher levels of baseline MMP-9 were found in patients who developed subsequent symptomatic flare compared with patients who did not [median 661 ng/ml, 25-75 interquartile range (IQR; 478.2-1441.3) versus 525.5 ng/ ml (339-662.7), respectively, p = 0.01]. Patients with serum MMP-9 levels of 945 ng/ ml or higher were at increased risk for relapse within 24 months [area under the curve (AUC) of 0.72 [95% confidence interval (CI): 0.56-0.88]; hazard ratio 8.1 (95% CI 3.0-21.9, p < 0.001)]. Serum MMP-9 concentrations showed weak and moderate correlation to baseline LS and FC, respectively (r = 0.31, p = 0.02; r = 0.46, p < 0.001). No correlation was found between serum MMP-9 to CRP and MaRIA score. CONCLUSIONS: Serum MMP-9 may be a promising biomarker for prediction of clinical flare in CD patients with quiescent disease.

Simple Endoscopic Score of Crohn Disease and Magnetic Resonance Enterography in Children: Report From ImageKids Study.

OBJECTIVES: We aimed to explore the ability of magnetic resonance enterography (MRE) to impute the simple endoscopic score of Crohn disease (SES-CD) in children with CD, in whom failure of ileal intubation is common and may impair SES-CD calculation in clinical studies. METHODS: This is a substudy of the prospective ImageKids study in which children with CD underwent ileocolonoscopy (scored by SES-CD) and MRE (scored on a 100 mm visual analogue scale [VAS] and by MaRIA). Mucosal healing (MH) was defined as SES-CD <3, MRE-VAS <20 mm, and/or MaRIA <7. RESULTS: A total of 237 children (22 centers, age 11.5 ±â€Š3.3 years), were enrolled. Ileal intubation has failed in 40 of 237 (17%). The agreement between SES-CD and MRE was 75% (k = 0.508, P < 0.001) in the ileum, and 68% to 85% in the colonic segments (k = 0.21-0.50, P < 0.001). The sensitivity and specificity of ileal MRE-VAS for MH were 91.7% (95% confidence interval 0.84-0.96) and 53.1% (95% confidence interval 0.43-0.63), respectively. The ileal MaRIA score (calculated in 33/40) was higher in the children without ileal intubation than in the others (20.5 ±â€Š7.1 vs 15.1 ±â€Š10.8, respectively, P = 0.0018). In 7% (16/237) of children, isolated active ileal disease would have been missed when considering SES-CD only. A multivariable model predicted the ileal SES-CD subscore from the MaRIA: SES-CDileum = 1.145 + 0.169 × MaRIAileum rounded to the nearest whole number (R = 0.17). Applying this model to the children without ileal intubation revealed that 29 of 33 (88%) had ileal disease; 8 of 29 patients (28%) with normal colonic SES-CD had imputed ileal SES-CD ≥3. CONCLUSIONS: MRE is useful for imputing the ileal disease in pediatric clinical studies, overcoming the problem of ileal nonintubation.

Convolutional Neural Networks for Radiologic Images: A Radiologist's Guide.

Deep learning has rapidly advanced in various fields within the past few years and has recently gained particular attention in the radiology community. This article provides an introduction to deep learning technology and presents the stages that are entailed in the design process of deep learning radiology research. In addition, the article details the results of a survey of the application of deep learning-specifically, the application of convolutional neural networks-to radiologic imaging that was focused on the following five major system organs: chest, breast, brain, musculoskeletal system, and abdomen and pelvis. The survey of the studies is followed by a discussion about current challenges and future trends and their potential implications for radiology. This article may be used as a guide for radiologists planning research in the field of radiologic image analysis using convolutional neural networks.

CT Image-based Decision Support System for Categorization of Liver Metastases Into Primary Cancer Sites: Initial Results.

RATIONALE AND OBJECTIVES: This study aimed to provide decision support for the human expert, to categorize liver metastases into their primary cancer sites. Currently, once a liver metastasis is detected, the process of finding the primary site is challenging, time-consuming, and requires multiple examinations. The proposed system can support the human expert in localizing the search for the cancer source by prioritizing the examinations to probable cancer sites. MATERIALS AND METHODS: The suggested method is a learning-based approach, using computed tomography (CT) data as the input source. Each metastasis is circumscribed by a radiologist in portal phase and in non-contrast CT images. Visual features are computed from these images, combined into feature vectors, and classified using support vector machine classification. A variety of different features were explored and tested. A leave-one-out cross-validation technique was conducted for classification evaluation. The methods were developed on a set of 50 lesion cases taken from 29 patients. RESULTS: Experiments were conducted on a separate set of 142 lesion cases taken from 71 patients with four different primary sites. Multiclass categorization results (four classes) achieved low accuracy results. However, the proposed system was found to provide promising results of 83% and 99% for top-2 and top-3 classification tasks, respectively. Moreover, when compared to the experts' ability to distinguish the different metastases, the system shows improved results. CONCLUSIONS: Automated systems, such as the one proposed, show promising new results and demonstrate new capabilities that, in the future, will be able to provide decision and treatment support for radiologists and oncologists, toward more efficient detection and treatment of cancer.

Assessment of patency capsule retention using MR diffusion-weighted imaging.

OBJECTIVES: Evaluate the ability of MR diffusion-weighted imaging (DWI) to predict patency capsule retention in Crohn's disease (CD). METHODS: Clinical and imaging data were prospectively reviewed for 80 CD patients following patency capsule administration and MR-DWI under institutional review board (IRB) approval with informed consent. Two radiologists separately assessed the presence/absence of restricted diffusion in the distal ileum. Apparent diffusion coefficients (ADC) from three regions of interest on the ileal wall were averaged. The association between restricted diffusion and retention, and sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated. Ability of ADC to predict retention was assessed with receiver operating characteristic (ROC) curve analysis. RESULTS: Restricted diffusion in the distal ileum was associated with capsule retention (p = 0.001, p < 0.0001). Sensitivity, specificity, PPV and NPV of restricted diffusion for capsule retention were 100.0%, 46.2%, 30.0%, 100% and 100.0%, 56.9%, 34.9%, 100%, respectively, for two radiologists. Accuracy of ADC to predict retention was high (area under the curve = 0.851, p < 0.0001). An ADC of 1.47 mm2/s showed 90.0% sensitivity and 50.0% specificity for retention. CONCLUSIONS: Sensitivity and NPV of restricted diffusion for patency capsule retention were 100%, suggesting that DWI may predict gastrointestinal tract capability to pass video camera endoscopy. KEY POINTS: • Capsule endoscopy enables assessment of the gastrointestinal mucosa in Crohn's disease • Prior patency capsule administration is recommended to evaluate gastrointestinal tract patency • MR diffusion-weighted imaging may detect pathological constriction of the ileum • Restricted diffusion in the distal ileum was associated with capsule retention • MR-DWI may predict gastrointestinal tract capability to pass capsule endoscopy.

The Role of Liver Segment-to-Spleen Volume Ratio in the Staging of Hepatic Fibrosis in Patients with Hepatitis C Virus Infection.

BACKGROUND: Accurate assessment of liver fibrosis is crucial for the management of patients with hepatitis C virus (HCV) infection. OBJECTIVES: To evaluate the performance of liver segment-to-spleen volume ratio in predicting the severity of liver fibrosis. METHODS: Sixty-four consecutive HCV patients were enrolled in this retrospective study. All patients underwent contrast-enhanced computed tomography (CT) and were divided into three groups based on their hepatic fibrosis stage evaluated by shear-wave elastography (SWE): non-advanced (F0-F1, n=29), advanced (F2, n=19) and severe fibrosis (F3-F4, n=16). Using semi-automated liver segmentation software, we calculated the following liver segments and spleen volumes for each participant: total liver volume (TLV), caudate lobe (CV), left lateral segment (LLV), left medial segment (LMV), right lobe (RV) and spleen (SV), a well as their ratios: CV/SV, RV/SV, LLV/SV, LMV/SV and TLV/SV. RESULTS: RV/SV was found to discriminate between patients with non-advanced and advanced fibrosis (P = 0.001), whereas SV, CV, RV, TLV/SV, LMV/SV and RV/SV discriminated between patients with advanced and severe fibrosis (P < 0.05). RV/SV ≤ 3.6 and RV ≤ 2.9 were identified as the best cutoff values to differentiate non-advanced from advanced fibrosis and advanced from severe fibrosis with sensitivities of 72.2% and 92.7%, specificities of 72.7% and 77.8%, and with an area under the receiver operating characteristic (ROC) curve of 0.797 and 0.847, respectively (P ≤ 0.002). CONCLUSIONS: RV/SV may be used for the assessment and monitoring of liver fibrosis in HCV patients prior to the administration of antiviral therapy, considering SWE as the reference method.

Intraluminal Uterine Hypodensity in CT Scans of Postmenopausal Women: Recommendations for Interpretation.

OBJECTIVE: This study aims to investigate the association between intraluminal uterine hypodensity and uterine malignancy and establish thresholds that would minimize routine gynecological evaluation. METHODS: Two groups were recruited retrospectively: cancer group, which comprised 32 sequential endometrial cancer patients, and postmenopausal group, which comprised 63 women, with no known gynecologic malignancy.Two radiologists independently measured hypodensity, transversely in the axial plane and anterioposteriorly in the sagittal plane.The association between cancer and hypodensity was evaluated. Receiver operating characteristic curves were evaluated diameters predictive of cancer. RESULTS: Hypodensity was associated with cancer (cancer group, 93.8% vs. postmenopausal group, 38.1%; P < 0.0001). Hypodensity diameters correlated highly with prediction of cancer (transverse area under the curve, 0.899; anteroposterior area under the curve, 0.892). Diameters of 19.5 mm transverse and 6.0 mm anteroposterior yielded a sensitivity of 87% and 83% and specificity of 91% and 83%, respectively. CONCLUSIONS: Intrauterine hypodensity is a common finding in computed tomography scans of postmenopausal women. A transverse diameter of 19.5 mm and an anteroposterior diameter of 6.0 mm are suggested as thresholds for further gynecological sonographic evaluation.

Perianal Pediatric Crohn Disease Is Associated With a Distinct Phenotype and Greater Inflammatory Burden.

OBJECTIVES: Data on the outcomes of children with perianal Crohn disease (pCD) are limited, although its presence is often used for justifying early use of biologics. We aimed to assess whether pCD in children is associated with more severe outcomes as found in adults. METHODS: Data were extracted from the ImageKids database, a prospective, multicenter, longitudinal cohort study. The study enrolled 246 children at disease onset or thereafter. All patients underwent comprehensive clinical, endoscopic, and radiologic evaluation at enrollment; 98 children had repeat evaluation at 18 months. RESULTS: Of the 234 included patients (mean age 14.2 ±â€Š2.4 years; 131 [56%] boys), 57 (24%) had perianal findings, whereas only 21 (9%) had fistulizing perianal disease. Children with pCD had reduced weight and height z scores compared with non-pCD patients (-0.9 vs -0.35, P = 0.03 and -0.68 vs -0.23, respectively; P = 0.04), higher weighted pediatric CD activity index (32 [interquartile range 16-50] vs 20 [8-37]; P = 0.004), lower serum albumin (3.6 ±â€Š0.7 vs 4.5 ±â€Š0.8, P = 0.016), and higher magnetic resonance enterography global inflammatory score (P = 0.04). Children with pCD had more rectal (57% vs 38%, P = 0.04), and jejunal involvement (31% vs 11% P = 0.003) and a higher prevalence of granulomas (64% vs 23%, P = 0.0001). Magnetic resonance enterography-based damage scores did not differ between groups. Patients with skin tags/fissures only, had similar clinical, endoscopic, and radiologic characteristics as patients with no perianal findings. CONCLUSIONS: Pediatric patients with pCD with fistulizing disease have distinct phenotypic features and a predisposition to a greater inflammatory burden.

Task-Driven Dictionary Learning Based on Mutual Information for Medical Image Classification.

OBJECTIVE: We present a novel variant of the bag-of-visual-words (BoVW) method for automated medical image classification. METHODS: Our approach improves the BoVW model by learning a task-driven dictionary of the most relevant visual words per task using a mutual information-based criterion. Additionally, we generate relevance maps to visualize and localize the decision of the automatic classification algorithm. These maps demonstrate how the algorithm works and show the spatial layout of the most relevant words. RESULTS: We applied our algorithm to three different tasks: chest x-ray pathology identification (of four pathologies: cardiomegaly, enlarged mediastinum, right consolidation, and left consolidation), liver lesion classification into four categories in computed tomography (CT) images and benign/malignant clusters of microcalcifications (MCs) classification in breast mammograms. Validation was conducted on three datasets: 443 chest x-rays, 118 portal phase CT images of liver lesions, and 260 mammography MCs. The proposed method improves the classical BoVW method for all tested applications. For chest x-ray, area under curve of 0.876 was obtained for enlarged mediastinum identification compared to 0.855 using classical BoVW (with p-value 0.01). For MC classification, a significant improvement of 4% was achieved using our new approach (with p-value = 0.03). For liver lesion classification, an improvement of 6% in sensitivity and 2% in specificity were obtained (with p-value 0.001). CONCLUSION: We demonstrated that classification based on informative selected set of words results in significant improvement. SIGNIFICANCE: Our new BoVW approach shows promising results in clinically important domains. Additionally, it can discover relevant parts of images for the task at hand without explicit annotations for training data. This can provide computer-aided support for medical experts in challenging image analysis tasks.

CT measurement of breast glandular tissue and its association with testicular cancer.

OBJECTIVES: To evaluate the associations between breast glandular tissues diameters as determined by CT and b-hCG levels, histological types, tumour spread and prognosis in patients with testicular germ cell tumour. METHODS: Ninety-four patients with pre-treatment CT scan and markers (b-hCG, AFP, LDH) were retrospectively collected. A radiologist measured diameters in all CT examinations and correlation between diameters and log (b-hCG) was assessed (Pearson's coefficient). The ability of measured diameters to predict lymphatic and distant haematogenous metastatic spread was evaluated (ROC curves). The associations between measured diameter cut-off values of 20 and 25 mm and International Germ Cell Cancer Collaborative Group (IGCCCG) classification, lymphatic and distant haematogenous metastatic spread and histological subtypes were evaluated (chi squared test). RESULTS: Breast glandular diameters correlated to log(b-hCG) (r = 0.579) and predicted distant haematogenous metastatic spread (AUC = 0.78). Worse prognosis (intermediate or poor IGCCCG) was shown for 20 mm (27.3 vs. 4.2 %, p = 0.005) and 25 mm (33.3 vs. 6.1 %, p = 0.014). A diameter of 25 mm was associated with non-seminoma (91.7 vs. 48.8 %, p = 0.005). CONCLUSION: Breast glandular tissue diameters correlated with log(b-hCG) and predicted distant haematogenous metastases. Twenty and 25 mm were associated with worse prognosis and 25 mm was able to distinguish between seminoma and non-seminoma. KEY POINTS: • CT breast glandular tissue diameter correlates with log(b-HCG) • Gynaecomastia in CT is associated with worse prognosis • Gynaecomastia in CT is associated with non-seminoma histological subtype.

Splenic Pregnancy: A New Minimally Invasive Approach to Treatment.

The spleen is a rare site of abdominal ectopic pregnancy. In a review of the literature, we found 16 published cases of primary splenic pregnancies. Of the cases identified, all received surgical intervention, with one case successfully treated with laparoscopic methotrexate injection, and the rest underwent splenectomy. We would like to present a case of primary splenic pregnancy in a 35-year-old woman successfully treated with percutaneous image-guided injection of methotrexate and KCl.