Computerized assessment of background parenchymal enhancement on breast dynamic contrast-enhanced-MRI including electronic lesion removal

Purpose

Current clinical assessment qualitatively describes background parenchymal enhancement (BPE) as minimal, mild, moderate, or marked based on the visually perceived volume and intensity of enhancement in normal fibroglandular breast tissue in dynamic contrast-enhanced (DCE)-MRI. Tumor enhancement may be included within the visual assessment of BPE, thus inflating BPE estimation due to angiogenesis within the tumor. Using a dataset of 426 MRIs, we developed an automated method to segment breasts, electronically remove lesions, and calculate scores to estimate BPE levels.

Approach

A U-Net was trained for breast segmentation from DCE-MRI maximum intensity projection (MIP) images. Fuzzy c -means clustering was used to segment lesions; the lesion volume was removed prior to creating projections. U-Net outputs were applied to create projection images of both, affected, and unaffected breasts before and after lesion removal. BPE scores were calculated from various projection images, including MIPs or average intensity projections of first- or second postcontrast subtraction MRIs, to evaluate the effect of varying image parameters on automatic BPE assessment. Receiver operating characteristic analysis was performed to determine the predictive value of computed scores in BPE level classification tasks relative to radiologist ratings.

Results

Statistically significant trends were found between radiologist BPE ratings and calculated BPE scores for all breast regions (Kendall correlation, p < 0.001 ). Scores from all breast regions performed significantly better than guessing (p < 0.025 from the z -test). Results failed to show a statistically significant difference in performance with and without lesion removal. BPE scores of the affected breast in the second postcontrast subtraction MIP after lesion removal performed statistically greater than random guessing across various viewing projections and DCE time points.

Conclusions

Results demonstrate the potential for automatic BPE scoring to serve as a quantitative value for objective BPE level classification from breast DCE-MR without the influence of lesion enhancement.

Passing the JMI Torch: Thank You to the Medical Imaging Community for our Amazing Journey

JMI Editor-in-Chief Maryellen L. Giger reflects with gratitude on the past decade of JMI and passes the leadership "torch" to Bennett A. Landman.

U-Net breast lesion segmentations for breast dynamic contrast-enhanced magnetic resonance imaging

Purpose

Given the dependence of radiomic-based computer-aided diagnosis artificial intelligence on accurate lesion segmentation, we assessed the performances of 2D and 3D U-Nets in breast lesion segmentation on dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) relative to fuzzy c-means (FCM) and radiologist segmentations.

Approach

Using 994 unique breast lesions imaged with DCE-MRI, three segmentation algorithms (FCM clustering, 2D and 3D U-Net convolutional neural networks) were investigated. Center slice segmentations produced by FCM, 2D U-Net, and 3D U-Net were evaluated using radiologist segmentations as truth, and volumetric segmentations produced by 2D U-Net slices and 3D U-Net were compared using FCM as a surrogate reference standard. Fivefold cross-validation by lesion was conducted on the U-Nets; Dice similarity coefficient (DSC) and Hausdorff distance (HD) served as performance metrics. Segmentation performances were compared across different input image and lesion types.

Results

2D U-Net outperformed 3D U-Net for center slice (DSC, HD p < 0.001 ) and volume segmentations (DSC, HD p < 0.001 ). 2D U-Net outperformed FCM in center slice segmentation (DSC p < 0.001 ). The use of second postcontrast subtraction images showed greater performance than first postcontrast subtraction images using the 2D and 3D U-Net (DSC p < 0.05 ). Additionally, mass segmentation outperformed nonmass segmentation from first and second postcontrast subtraction images using 2D and 3D U-Nets (DSC, HD p < 0.001 ).

Conclusions

Results suggest that 2D U-Net is promising in segmenting mass and nonmass enhancing breast lesions from first and second postcontrast subtraction MRIs and thus could be an effective alternative to FCM or 3D U-Net.

Toward fairness in artificial intelligence for medical image analysis: identification and mitigation of potential biases in the roadmap from data collection to model deployment

Purpose

To recognize and address various sources of bias essential for algorithmic fairness and trustworthiness and to contribute to a just and equitable deployment of AI in medical imaging, there is an increasing interest in developing medical imaging-based machine learning methods, also known as medical imaging artificial intelligence (AI), for the detection, diagnosis, prognosis, and risk assessment of disease with the goal of clinical implementation. These tools are intended to help improve traditional human decision-making in medical imaging. However, biases introduced in the steps toward clinical deployment may impede their intended function, potentially exacerbating inequities. Specifically, medical imaging AI can propagate or amplify biases introduced in the many steps from model inception to deployment, resulting in a systematic difference in the treatment of different groups.

Approach

Our multi-institutional team included medical physicists, medical imaging artificial intelligence/machine learning (AI/ML) researchers, experts in AI/ML bias, statisticians, physicians, and scientists from regulatory bodies. We identified sources of bias in AI/ML, mitigation strategies for these biases, and developed recommendations for best practices in medical imaging AI/ML development.

Results

Five main steps along the roadmap of medical imaging AI/ML were identified: (1) data collection, (2) data preparation and annotation, (3) model development, (4) model evaluation, and (5) model deployment. Within these steps, or bias categories, we identified 29 sources of potential bias, many of which can impact multiple steps, as well as mitigation strategies.

Conclusions

Our findings provide a valuable resource to researchers, clinicians, and the public at large.

MIDRC CRP10 AI interface-an integrated tool for exploring, testing and visualization of AI models

Objective. Developing Machine Learning models (N Gorre et al 2023) for clinical applications from scratch can be a cumbersome task requiring varying levels of expertise. Seasoned developers and researchers may also often face incompatible frameworks and data preparation issues. This is further complicated in the context of diagnostic radiology and oncology applications, given the heterogenous nature of the input data and the specialized task requirements. Our goal is to provide clinicians, researchers, and early AI developers with a modular, flexible, and user-friendly software tool that can effectively meet their needs to explore, train, and test AI algorithms by allowing users to interpret their model results. This latter step involves the incorporation of interpretability and explainability methods that would allow visualizing performance as well as interpreting predictions across the different neural network layers of a deep learning algorithm.Approach. To demonstrate our proposed tool, we have developed the CRP10 AI Application Interface (CRP10AII) as part of the MIDRC consortium. CRP10AII is based on the web service Django framework in Python. CRP10AII/Django/Python in combination with another data manager tool/platform, data commons such as Gen3 can provide a comprehensive while easy to use machine/deep learning analytics tool. The tool allows to test, visualize, interpret how and why the deep learning model is performing. The major highlight of CRP10AII is its capability of visualization and interpretability of otherwise Blackbox AI algorithms.Results. CRP10AII provides many convenient features for model building and evaluation, including: (1) query and acquire data according to the specific application (e.g. classification, segmentation) from the data common platform (Gen3 here); (2) train the AI models from scratch or use pre-trained models (e.g. VGGNet, AlexNet, BERT) for transfer learning and test the model predictions, performance assessment, receiver operating characteristics curve evaluation; (3) interpret the AI model predictions using methods like SHAPLEY, LIME values; and (4) visualize the model learning through heatmaps and activation maps of individual layers of the neural network.Significance. Unexperienced users may have more time to swiftly pre-process, build/train their AI models on their own use-cases, and further visualize and explore these AI models as part of this pipeline, all in an end-to-end manner. CRP10AII will be provided as an open-source tool, and we expect to continue developing it based on users' feedback.

A Framework for Evaluating the Technical Performance of Multiparameter Quantitative Imaging Biomarkers (mp-QIBs)

Multiparameter quantitative imaging incorporates anatomical, functional, and/or behavioral biomarkers to characterize tissue, detect disease, identify phenotypes, define longitudinal change, or predict outcome. Multiple imaging parameters are sometimes considered separately but ideally are evaluated collectively. Often, they are transformed as Likert interpretations, ignoring the correlations of quantitative properties that may result in better reproducibility or outcome prediction. In this paper we present three use cases of multiparameter quantitative imaging: i) multidimensional descriptor, ii) phenotype classification, and iii) risk prediction. A fourth application based on data-driven markers from radiomics is also presented. We describe the technical performance characteristics and their metrics common to all use cases, and provide a structure for the development, estimation, and testing of multiparameter quantitative imaging. This paper serves as an overview for a series of individual articles on the four applications, providing the statistical framework for multiparameter imaging applications in medicine.

Advancing Research on Medical Image Perception by Strengthening Multidisciplinary Collaboration

Medical image interpretation is central to detecting, diagnosing, and staging cancer and many other disorders. At a time when medical imaging is being transformed by digital technologies and artificial intelligence, understanding the basic perceptual and cognitive processes underlying medical image interpretation is vital for increasing diagnosticians’ accuracy and performance, improving patient outcomes, and reducing diagnostician burn-out. Medical image perception remains substantially understudied. In September of 2019, the National Cancer Institute convened a multidisciplinary panel of radiologists and pathologists together with researchers working in medical image perception and adjacent fields of cognition and perception for the “Cognition and Medical Image Perception Think Tank.” The Think Tank’s key objectives were: to identify critical unsolved problems related to visual perception in pathology and radiology from the perspective of diagnosticians; to discuss how these clinically relevant questions could be addressed through cognitive and perception research; to identify barriers and solutions for transdisciplinary collaborations; to define ways to elevate the profile of cognition and perception research within the medical image community; to determine the greatest needs to advance medical image perception; and to outline future goals and strategies to evaluate progress. The Think Tank emphasized diagnosticians’ perspectives as the crucial starting point for medical image perception research, with diagnosticians describing their interpretation process and identifying perceptual and cognitive problems that arise. This paper reports the deliberations of the Think Tank participants to address these objectives and highlight opportunities to expand research on medical image perception.

Report from the RSNA COVID-19 Task Force: COVID-19 Impact on Academic Radiology Research-A Survey of Vice Chairs of Research

Objective

Survey vice chairs of research from academic radiology departments on the impact of coronavirus disease 2019 (COVID-19) on research activities.

Methods

The survey asked respondents to quantify changes in research performed during the shutdown and ramp-up, relative to pre–COVID-19 levels. Respondents estimated research activity changes by overall research type (wet, instrumentation, or core facilities: prospective non–COVID-19 clinical research and computational laboratories) and then by the research activity type (data analysis, grant or manuscript writing, clinician involvement, summer student participation, and international research fellow appointments).The χ² test was used for comparison between shutdown and ramp-up, with Yates correction when necessary.

Results

Of 105 vice chairs contacted, 46 (43.8%) responded. For 95.5%, wet, instrumentation, or core facilities research decreased to ≤50% during shutdown and for 83.3% during ramp-up (P < .0001). In addition, 89.2% and 46.5% indicated reduction to ≤25% of non–COVID-19 clinical research during shutdown and ramp-up, respectively (P < .0001). Only computational research increased to 120% during shutdown (39.5%) or ramp-up (50%) (P = .8984). For data analysis from closed laboratories, 75% and 86% showed decreased activity during shutdown and ramp-up, respectively (P = .28). Increased grant writing during shutdown and ramp-up was reported by 45.5% and 23.3% (P = .093). For 52.3% and 23.3%, manuscript writing and submission increased during shutdown and ramp-up, respectively (P < .02). Clinician research involvement trended toward relative decreases during shutdown (84.1% versus 60.5%, P = .05). There was similar drop in summer student participation (shutdown: 86.4%, ramp-up: 83.7%, P = .95) and international researcher appointment (shutdown: 85.7%, ramp-up: 86.1%; P = .96).

Conclusion

Many radiology research activities diminished during the COVID-19 shutdown and to a lesser extent during the ramp-up. Activities that could be done remotely, such as computational analysis and grant and manuscript writing and submission, increased.

Medical imaging of COVID-19

Editor-in-Chief Maryellen Giger introduces the JMI Special Issue on COVID-19 Medical Imaging Research.

Abstract PS3-01: Quantitative dynamic contrast-enhanced (DCE) MRI radiomic phenotypes for prediction of nodal and distal metastasis in breast cancer patients

Background and Purpose:Image-based tumor phenotypes by using computer extraction techniques have been studied for evaluation of breast cancer invasiveness, stage, lymph node involvement, molecular subtypes and genomics. In this project we aimed to investigate ability of computer-extracted breast MR imaging radiomic phenotypes to predict nodal and distant metastasis in breast cancer patients.

MATERIALS AND METHODS:This retrospective IRB approved study included 416 biopsy proven breast cancer patients who had pretreatment DCE MRI in a single institution between 2014 and 2018. Patient’s demographic, clinical data, pathology at diagnosis and surgery, nodal and distant metastasis (M1) at follow up were documented. Using QuantX imaging software, the tumor volume of interest was automatically-segmented using the multiple dynamic phases of DCE MRI. A total of 33 radiomic features describing tumor phenotype were extracted from each tumor site. A linear discriminant analysis (LDA) as a classifier with nested feature selection 10-fold cross validation was used to build the radiomic signature for prediction of nodal and distant metastasis occurrence. Receiver operating characteristic (ROC) and precision-recall analyses were used to evaluate performance, with 95% confidence intervals from 1000 bootstraps, and Kaplan-Meier was used to calculate the progression-free survival estimates and associated hazard ratio at the median cutpoint of the probability of metastasis calculated by the LDA in the 10-fold cross-validation.

RESULTS:The quantitative DCE MRI radiomic model was able to differentiate between breast cancer patients with and without distant metastatic disease at follow up with area under the ROC of 0.75 (95% CI 0.65; 0.82) and precision-recall curves 0.46 (0.33;0.69), hazard ratio at median cut point is 3.76 (2.27; 6.24), p&lt;0.001. Volume, surface area, sphericity, margin, maximum uptake, and washout rate variation features played the most important role in differentiating between breast cancer patients with and without distant metastasis.

The DCE radiomic model was able predict presence of ipsilateral nodal disease (≥1 positive lymph nodes) at surgery with AUC 0.66 (95% CI: 0.60; 0.71), ≥4 positive lymph nodes at surgery with AUC 0.67 (95% CI: 0.60; 0.74), and N2/N3 disease with AUC 0.64 (95% CI: 0.56; 0.72). Effective radius was most important feature for nodal disease prediction.

CONCLUSIONS:Our results show that DCE MRI based radiomic phenotypes were able to predict nodal involvement and distant metastasis in breast cancer patients. Quantitative breast DCE MRI radiomics shows promise for noninvasive image based phenotyping for prediction of nodal and distant metastatic disease in breast cancer patients.

Citation Format: Gaiane Margishvili Rauch, Karen Drukker, Nabil Elshafeey, Rania M.m. Mohamed, Medina Boge, Beatriz E. Adrada, Rosalind P Candelaria, Mo Salama, Irene Shkatova, Maryellen Giger, Wei T Yang. Quantitative dynamic contrast-enhanced (DCE) MRI radiomic phenotypes for prediction of nodal and distal metastasis in breast cancer patients [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS3-01.

Deep convolutional neural networks in the classification of dual-energy thoracic radiographic views for efficient workflow: analysis on over 6500 clinical radiographs

DICOM header information is frequently used to classify medical image types; however, if a header is missing fields or contains incorrect data, the utility is limited. To expedite image classification, we trained convolutional neural networks (CNNs) in two classification tasks for thoracic radiographic views obtained from dual-energy studies: (a) distinguishing between frontal, lateral, soft tissue, and bone images and (b) distinguishing between posteroanterior (PA) or anteroposterior (AP) chest radiographs. CNNs with AlexNet architecture were trained from scratch. 1910 manually classified radiographs were used for training the network to accomplish task (a), then tested with an independent test set (3757 images). Frontal radiographs from the two datasets were combined to train a network to accomplish task (b); tested using an independent test set of 1000 radiographs. ROC analysis was performed for each trained CNN with area under the curve (AUC) as a performance metric. Classification between frontal images (AP/PA) and other image types yielded an AUC of 0.997 [95% confidence interval (CI): 0.996, 0.998]. Classification between PA and AP radiographs resulted in an AUC of 0.973 (95% CI: 0.961, 0.981). CNNs were able to rapidly classify thoracic radiographs with high accuracy, thus potentially contributing to effective and efficient workflow.

Integrating structured abstracts in the Journal of Medical Imaging

An editorial by Editor-in-Chief Maryellen Giger explains the journal's transition to structured abstracts.

Transfer Learning From Convolutional Neural Networks for Computer-Aided Diagnosis: A Comparison of Digital Breast Tomosynthesis and Full-Field Digital Mammography

RATIONALE AND OBJECTIVES

With the growing adoption of digital breast tomosynthesis (DBT) in breast cancer screening, we compare the performance of deep learning computer-aided diagnosis on DBT images to that of conventional full-field digital mammography (FFDM).

MATERIALS AND METHODS

In this study, we retrospectively collected FFDM and DBT images of 78 biopsy-proven lesions from 76 patients. A region of interest was selected for each lesion on FFDM, synthesized 2D, and DBT key slice images. Features were extracted from each lesion using a pretrained convolutional neural network (CNN) and served as input to a support vector machine classifier trained in the task of predicting likelihood of malignancy.

RESULTS

From receiver operating characteristic (ROC) analysis of all 78 lesions, the synthesized 2D image performed best in both the cradiocaudal view (area under the ROC curve [AUC] = 0.81, SE = 0.05) and mediolateral oblique view (AUC = 0.88, SE = 0.04) in the task of lesion characterization. When cradiocaudal and mediolateral oblique data of each lesion were merged through soft voting, DBT key slice image performed best (AUC = 0.89, SE = 0.04). When only masses and architectural distortions (ARDs) were considered, DBT performed significantly better than FFDM (p = 0.024).

CONCLUSION

DBT performed significantly better than FFDM in the merged view classification of mass and ARD lesions. The increased performance suggests that the information extracted by the CNN from DBT images may be more relevant to lesion malignancy status than the information extracted from FFDM images. Therefore, this study provides supporting evidence for the efficacy of computer-aided diagnosis on DBT in the evaluation of mass and ARD lesions.

Radiomics robustness assessment and classification evaluation: A two-stage method demonstrated on multivendor FFDM

PURPOSE

Radiomic texture analysis is typically performed on images acquired under specific, homogeneous imaging conditions. These controlled conditions may not be representative of the range of imaging conditions implemented clinically. We aim to develop a two-stage method of radiomic texture analysis that incorporates the reproducibility of individual texture features across imaging conditions to guide the development of texture signatures which are robust across mammography unit vendors.

METHODS

Full-field digital mammograms were retrospectively collected for women who underwent screening mammography on both a Hologic Lorad Selenia and GE Senographe 2000D system. Radiomic features were calculated on manually placed regions of interest in each image. In stage one (robustness assessment), we identified a set of nonredundant features that were reproducible across the two different vendors. This was achieved through hierarchical clustering and application of robustness metrics. In stage two (classification evaluation), we performed stepwise feature selection and leave-one-out quadratic discriminant analysis (QDA) to construct radiomic signatures. We refer to this two-state method as robustness assessment, classification evaluation (RACE). These radiomic signatures were used to classify the risk of breast cancer through receiver operator characteristic (ROC) analysis, using the area under the ROC curve as a figure of merit in the task of distinguishing between women with and without high-risk factors present. Generalizability was investigated by comparing the classification performance of a feature set on the images from which they were selected (intravendor) to the classification performance on images from the vendor on which it was not selected (intervendor). Intervendor and intravendor performances were also compared to the performance obtained by implementing ComBat, a feature-level harmonization method and to the performance by implementing ComBat followed by RACE.

RESULTS

Generalizability, defined as the difference between intervendor and intravendor classification performance, was shown to monotonically decrease as the number of clusters used in stage one increased (Mann-Kendall P < 0.001). Intravendor performance was not shown to be statistically different from ComBat harmonization while intervendor performance was significantly higher than ComBat. No significant difference was observed between either of the single methods and the use of ComBat followed by RACE.

CONCLUSIONS

A two-stage method for robust radiomic signature construction is proposed and demonstrated in the task of breast cancer risk assessment. The proposed method was used to assess generalizability of radiomic texture signatures at varying levels of feature robustness criteria. The results suggest that generalizability of feature sets monotonically decreases as reproducibility of features decreases. This trend suggests that considerations of feature robustness in feature selection methodology could improve classifier generalizability in multifarious full-field digital mammography datasets collected on various vendor units. Additionally, harmonization methods such as ComBat may hold utility in classification schemes and should continue to be investigated.

Opportunities and challenges to utilization of quantitative imaging: Report of the AAPM practical big data workshop

BACKGROUND

This article is a summary of the quantitative imaging subgroup of the 2017 AAPM Practical Big Data Workshop (PBDW-2017) on progress and challenges in big data applied to cancer treatment and research supplemented by a draft white paper following an American Association of Physicists in Medicine FOREM meeting on Imaging Genomics in 2014.

AIMS

The goal of PBDW-2017 was to close the gap between theoretical vision and practical experience with encountering and solving challenges in curating and analyzing data.

CONCLUSIONS

Recommendations based on the meetings are summarized.

Special Section Guest Editorial: Radiomics and Deep Learning

This guest editorial introduces and summarizes the JMI Special Section on Radiomics and Deep Learning.

Abstract 2633: Radiogenomics of breast cancer using DCE-MRI and gene expression profiling

The utilization of radiomics (high-throughput extraction and analysis of imaging features) to abstract underlying genomic features of an evolving malignancy is an emerging field that has the potential to detail biological information of a tumor in a non-invasive and more accessible manner. Furthermore, applying radiomics confers a comprehensive spatial view of the tumor, a potential advantage over the limitations of sampling a small region of tissue that may not accurately represent the underlying complexity of the entire tumor. Most studies to date that incorporate radiogenomics in the analysis of breast cancers have focused on few basic clinical data or individual genetic mutations such as BRCA or HER2 status.

Here, we use an automated quantitative radiomics analysis platform developed at the University of Chicago that enables computerized feature extraction of tumors to analyze magnetic resonant imaging scans of 50 breast cancer patients (mean age of diagnosis [range]: 54 [24-89]; receptor status: HER2+: 14, Triple negative: 7; stage [1 through 4]: 10%, 40%, 42%, 8%) who have had comprehensive gene expression profiling performed using Agilent Human Gene Expression arrays. Our imaging platform extracts 38 features across six major phenotypes (size, shape, morphology, enhancement texture, kinetic curve assessment, and enhancement variance kinetics) (see Table for listing of 24 selected features). Existing radiomic analysis derived from a TCGA/TCIA dataset suggests that there are many correlations between imaging phenotypes and various genetic pathways, such as VEGF signaling and volume of enhancing voxels, base excision repair and enhancement texture entropy, and TGF-beta signaling and enhancement texture variance. We confirm these relationships as well as establish novel associations using a robust imaging dataset. By associating specific radiomic features with gene expression profile of tumors, we have the opportunity to extract detailed biological information non-invasively through clinical imaging. Selected imaging phenotypes extracted from MRI scansPhenotype categoryImage phenotypeDescriptionSizeVolumeVolume of lesionSizeEffective diameterDiameter of a sphere with the same volume as the lesionSizeSurface areaLesion surface areaShapeSphericitySimilarity of the lesion shape to a sphereShapeIrregularityDeviation of the lesion surface from the surface of a sphereShapeSurface area / volumeRatio of surface area to volumeMorphologyMargin sharpnessMean of the image gradient at the lesion marginMorphologyVariance of margin sharpnessVariance of the image gradient at the lesion marginMorphologyVariance of radial gradient histogramDegree to which the enhancement structure extends in a radial pattern originating from the center of the lesionEnhancement TextureContrastLocal image variationsEnhancement TextureEntropyRandomness of the gray-levelsEnhancement TextureDifference varianceVariations of difference of gray-levels between voxel-pairsEnhancement TextureAngular second momentImage homogeneityEnhancement TextureMaximum correlation coefficientNonlinear gray-level dependenceEnhancement TextureSum averageOverall brightnessEnhancement TextureSum of squaresSpread in the gray-level distributionKinetic Curve AssessmentMaximum enhancementMaximum contrast enhancementKinetic Curve AssessmentTime to peakTime at which the maximum enhancement occursKinetic Curve AssessmentUptake rateUptake speed of the contrast enhancementKinetic Curve AssessmentCurve shape indexDifference between late and early enhancementKinetic Curve AssessmentTotal rate variationHow rapidly the contrast will enter and exit from the lesionEnhancement-Variance KineticsMaximum variance of enhancementMaximum spatial variance of contrast enhancement over timeEnhancement-Variance KineticsTime to peak maximum varianceTime at which the maximum variance occursEnhancement-Variance KineticsEnhancement variance increasing rateRate of increase of the enhancement-variance during uptake

Citation Format: Albert C. Yeh, Stephanie McGregor, Hui Li, Yuan Ji, Yitan Zhu, Tatyana Grushko, Alexandra Edwards, Fan Lui, Jing Zhang, Qiu Niu, Yonglan Zheng, Toshio Yoshimatsu, Galina Khramtsova, Karen Drukker, Gregory Karczmar, Hiroyuki Abe, Jeffrey Mueller, Maryellen Giger, Olufunmilayo Olopade. Radiogenomics of breast cancer using DCE-MRI and gene expression profiling. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2633.

Long-term results of microcoil embolization for colonic haemorrhage: how common is rebleeding?

OBJECTIVE

To determine the long-term results of patients undergoing transcatheter coil embolization for the treatment of acute colonic haemorrhage.

METHODS

Patients undergoing angiography for suspected colonic bleeding between January 2002 and December 2012 were reviewed (average age, 60 years; 38.4% male). Baseline, procedural and outcome parameters were recorded following the Society of Interventional Radiology guidelines. Primary outcome measures included early (<30 days) and delayed (>30 days) rebleeding events and adverse procedure-related complication. Average follow-up time was 996 days (median, 232 days; range, 30-3663 days).

RESULTS

One or multiple sites of bleeding were identified in 40 cases. Coil embolization was performed in 39 patients, 26 (66.7%, 26/39) of whom were treated successfully without technical/clinical failure (n = 12) or loss to follow-up (n = 1). Three patients (11.5%, 3/26) rebled in the early period within 30 days; one patient went on to hemicolectomy. Four patients (15.3%, 4/26) experienced delayed rebleeding after 30 days; two of whom also underwent hemicolectomy. No major complication occurred. One minor complication of short segment arterial dissection was seen in the clinical failure group. One case of asymptomatic ischaemia was identified on a patient undergoing pre-operative colonoscopy for elective bowel resection. No instances of ischaemic stricture were seen. All-cause mortality of successfully treated and all patients at 1 year was 31% (8/26) and 30% (12/40), respectively.

CONCLUSION

Transcatheter coil embolization is a durable treatment option with a technical and clinical success rate of 67% in the setting of acute colonic haemorrhage. A modest level of rebleeding was seen among successfully treated patients in both the early and delayed periods; in the majority of patients, embolization proved to be definitive therapy.

ADVANCES IN KNOWLEDGE

Transcatheter coil embolization is a durable and potentially definitive therapy in the management of acute colonic haemorrhage.

Radiologically Guided Placement of Mushroom-retained Gastrostomy Catheters: Long-term Outcomes of Use in 300 Patients at a Single Center

PURPOSE

To assess long-term outcomes including risk of complications and nutritional benefits of mushroom-retained (pull-type) gastrostomy catheters placed in patients by interventional radiologists.

MATERIALS AND METHODS

All patients who received pull-type gastrostomy tubes between 2010 and 2013 were retrospectively reviewed, including 142 men (average weight, 169.6 lb [76.32 kg]; mean age, 65.2 years; range, 22-92 years) and 158 women (average weight, 150.4 lb [67.68 kg]; mean age, 65.2 years; range, 18-98 years). Indications for placement were cerebrovascular accident (n = 80), failure to thrive (n = 71), other central nervous system disorder (n = 51), head and neck cancer (n = 47), and other malignancy (n = 51). Complications were recorded per Society of Interventional Radiology practice guidelines. Patient weight was documented at specific follow-up intervals. Statistical analysis was performed by using the Student t test and one-way analysis of variance for the effects of sex and indication for placement, respectively, on average weight change.

RESULTS

The technical success rate was 98.4% (300 of 305 patients). Major and minor complications occurred at a rate of 3.7% (n = 11) and 13% (n = 39), respectively. Follow-up weight during the early (≤45 days), intermediate (≤180 days), and long-term (>180 days) periods was available for 71% (n = 214), 36% (n = 108), and 15% (n = 44) of the 300 patients, respectively. Weight gain occurred in 77% (160 of 214), 60% (65 of 108), and 73% (32 of 44) of the patients, respectively. Patients who gained weight gained 6.7, 10.6, and 16.3 lb (3.02, 4.77, and 7.34 kg) during each follow-up period, respectively. Average weight gain at follow-up in all patients was 4.2, 0.6, and 5.4 lb (1.89, 0.27, and 2.43 kg), respectively. No significant differences in average weight change were seen among groups when they were classified according to sex or indication for placement.

CONCLUSION

Placement of mushroom-retained gastrostomy catheters is a viable long-term treatment option for enteral nutrition, with complication rates similar to those reported for other gastrostomy techniques. Improvement in nutrition status measured as weight gain was seen in most patients in both early and long-term periods.

Dual-lumen chest port infection rates in patients with head and neck cancer

PURPOSE

The aim of this study was to investigate dual-lumen chest port infection rates in patients with head and neck cancer (HNC) compared to those with other malignancies (non-HNC).

MATERIALS AND METHODS

An IRB-approved retrospective study was performed on 1,094 consecutive chest ports placed over a 2-year period. Patients with poor follow-up (n = 53), no oncologic history (n = 13), or single-lumen ports (n = 183) were excluded yielding a study population of 845 patients. The electronic medical records were queried for demographic information, data regarding ports and infections, and imaging review.

RESULTS

HNC patients experienced more infections (42 vs. 30), an increased infection rate per 1,000 catheter days (0.68 vs. 0.21), and more early infections within 30 days compared to non-HNC patients (10 vs. 6) (p < 0.001, p < 0.001, p = 0.02, respectively). An existing tracheostomy at the time of port placement was associated with infection in the HNC group (p = 0.02) but was not an independent risk factor for infection in the study population overall (p = 0.06). There was a significant difference in age, male gender, and right-sided ports between the HNC and non-HNC groups (p < 0.01, p < 0.001, and p = 0.01), although these were not found to be independent risk factors for infection (p = 0.32, p = 0.76, p = 0.16).

CONCLUSION

HNC patients are at increased risk for infection of dual-lumen chest ports placed via a jugular approach compared to patients with other malignancies. Tracheostomy is associated with infection in HNC patients but is not an independent risk factor for infection in the oncologic population as a whole.

Comparison of barbed versus conventional sutures for wound closure of radiologically implanted chest ports

PURPOSE

To retrospectively compare the incidences of complications with barbed suture versus conventional interrupted suture for incision closure in implantable chest ports.

MATERIALS AND METHODS

A total of 715 power-injectable dual-lumen chest ports placed between 2011 and 2013 were studied. Primary outcomes included wound dehiscence, local port infection, local infections treated by wound packing, early infections within 30 days, and total infections. A multivariate analysis of independent risk factors for port infection was also performed.

RESULTS

A total of 442 ports were closed with nonbarbed suture, versus 273 closed with barbed suture. Mean catheter-days in the traditional and barbed groups were 257.9 (range, 3-722) and 189.1 (range, 13-747), respectively (P < .01). The rate of dehiscence with traditional suture (1.6%; seven of 442) was significantly higher than that with barbed suture (zero of 273; P = .04). Percentage of total infections was also significantly higher with traditional suture (9.5% vs 5.1%; P = .03). No difference in rate of infection per 1,000 catheter-days was seen between traditional and barbed suture groups (0.0035 vs 0.0026; P = .17). The rate of local infection with traditional suture was significantly higher (2.7% vs 0.4%; P = .02). Additionally, multivariate analysis identified the use of traditional suture as the only independent risk factor for infection (39% vs 25%; P = .03).

CONCLUSIONS

Barbed suture for incision closure in implantable dual-lumen chest ports was associated with lower rates of dehiscence and potentially lower rates of local infectious complications compared with traditional nonbarbed suture.

Potential of computer-aided diagnosis of high spectral and spatial resolution (HiSS) MRI in the classification of breast lesions

PURPOSE

To compare the performance of computer-aided diagnosis (CADx) analysis of precontrast high spectral and spatial resolution (HiSS) MRI to that of clinical dynamic contrast-enhanced MRI (DCE-MRI) in the diagnostic classification of breast lesions.

MATERIALS AND METHODS

Thirty-four malignant and seven benign lesions were scanned using two-dimensional (2D) HiSS and clinical 4D DCE-MRI protocols. Lesions were automatically segmented. Morphological features were calculated for HiSS, whereas both morphological and kinetic features were calculated for DCE-MRI. After stepwise feature selection, Bayesian artificial neural networks merged selected features, and receiver operating characteristic (ROC) analysis evaluated the performance with leave-one-lesion-out validation.

RESULTS

AUC (area under the ROC curve) values of 0.92 ± 0.06 and 0.90 ± 0.05 were obtained using CADx on HiSS and DCE-MRI, respectively, in the task of classifying benign and malignant lesions. While we failed to show that the higher HiSS performance was significantly better than DCE-MRI, noninferiority testing confirmed that HiSS was not worse than DCE-MRI.

CONCLUSION

CADx of HiSS (without contrast) performed similarly to CADx on clinical DCE-MRI; thus, computerized analysis of HiSS may provide sufficient information for diagnostic classification. The results are clinically important for patients in whom contrast agent is contra-indicated. Even in the limited acquisition mode of 2D single slice HiSS, by using quantitative image analysis to extract characteristics from the HiSS images, similar performance levels were obtained as compared with those from current clinical 4D DCE-MRI. As HiSS acquisitions become possible in 3D, CADx methods can also be applied. Because HiSS and DCE-MRI are based on different contrast mechanisms, the use of the two protocols in combination may increase diagnostic accuracy.