CT and MR Protocol Standardization Across a Large Health System: Providing a Consistent Radiologist, Patient, and Referring Provider Experience

Lessons learned: strategies for implementing and the ongoing use of LI-RADS in your practice

The establishment of the Liver Imaging Reporting and Data System (LI-RADS) in 2011 provided a comprehensive approach to standardized imaging, interpretation, and reporting of liver observations in patients diagnosed with or at risk for hepatocellular carcinoma (HCC). Each set of algorithms provides criteria pertinent to the various components of HCC management including surveillance, diagnosis, staging, and treatment response supported by a detailed lexicon of terms applicable to a wide range of liver imaging scenarios. Before its widespread adoption, the variability in the terminology of diagnostic criteria and definitions of imaging features led to significant challenges in patient management and made it difficult to replicate findings or apply them consistently. The integration of LI-RADS into the clinical setting has enhanced the efficiency and clarity of communication between radiologists, referring providers, and patients by employing a uniform language that averts miscommunications. LI-RADS has been strengthened with its integration into the American Association for Study of Liver Diseases practice guidelines. We will provide the background on the initial development of LI-RADS and reasons for development to serve as a starting point for conveying the system's benefits and evolution over the years. We will also suggest strategies for the implementation and maintenance of a LI-RADS program will be discussed.

Solving the Pervasive Problem of Protocol Non-Compliance in MRI using an Open-Source tool mrQA

Pooling data across diverse sources acquired by multisite consortia requires compliance with a predefined reference protocol i.e., ensuring different sites and scanners for a given project have used identical or compatible MR physics parameter values. Traditionally, this has been an arduous and manual process due to difficulties in working with the complicated DICOM standard and lack of resources allocated towards protocol compliance. Moreover, issues of protocol compliance is often overlooked for lack of realization that parameter values are routinely improvised/modified locally at various sites. The inconsistencies in acquisition protocols can reduce SNR, statistical power, and in the worst case, may invalidate the results altogether. An open-source tool, mrQA was developed to automatically assess protocol compliance on standard dataset formats such as DICOM and BIDS, and to study the patterns of non-compliance in over 20 open neuroimaging datasets, including the large ABCD study. The results demonstrate that the lack of compliance is rather pervasive. The frequent sources of non-compliance include but are not limited to deviations in Repetition Time, Echo Time, Flip Angle, and Phase Encoding Direction. It was also observed that GE and Philips scanners exhibited higher rates of non-compliance relative to the Siemens scanners in the ABCD dataset. Continuous monitoring for protocol compliance is strongly recommended before any pre/post-processing, ideally right after the acquisition, to avoid the silent propagation of severe/subtle issues. Although, this study focuses on neuroimaging datasets, the proposed tool mrQA can work with any DICOM-based datasets.

Review of standard paediatric neuroradiology MRI protocols from 12 UK tertiary paediatric hospitals: is there much variation between centres?

AIM

To investigate how magnetic resonance imaging (MRI) examinations are protocolled in tertiary paediatric neuroradiology centres around the UK for some of the more common presentations encountered in paediatric neuroradiology, and to identify any variations of note.

MATERIALS AND METHODS

All 19 UK tertiary paediatric neuroradiology centres registered with the British Society of Neuroradiologists-Paediatric Group were contacted and asked if they could provide a copy of their standard MRI protocols. Twelve responded (63%) and 10 of the more common presentations were selected and the standard acquired sequences obtained at each participating centre were compared. Where available the collated protocols were also compared against current published guidance.

RESULTS

The basic sequences carried out by centres around the UK are similar; however, there are lots of variations overall. The only standardised protocol currently being implemented nationally in paediatric imaging is that for brain tumours. Otherwise, chosen protocols are generally dependent on the preferences and technical capabilities of individual centres. Suggested published protocols also exist for non-accidental injury (NAI), multiple sclerosis, epilepsy, and head and neck imaging.

CONCLUSIONS

The differences in MRI protocolling depend in part on technical capabilities and in part on the experience and preferences of the paediatric neuroradiologists at each centre. For most presentations, there is no consensus as to what constitutes the perfect protocol. The present results will be useful for specialist centres who may wish to review their current protocols, and for more generalist centres to use as a reference to guide their MRI protocolling.

Must-have Qualities of Clinical Research on Artificial Intelligence and Machine Learning

In the field of computer science, known as artificial intelligence, algorithms imitate reasoning tasks that are typically performed by humans. The techniques that allow machines to learn and get better at tasks such as recognition and prediction, which form the basis of clinical practice, are referred to as machine learning, which is a subfield of artificial intelligence. The number of artificial intelligence-and machine learnings-related publications in clinical journals has grown exponentially, driven by recent developments in computation and the accessibility of simple tools. However, clinicians are often not included in data science teams, which may limit the clinical relevance, explanability, workflow compatibility, and quality improvement of artificial intelligence solutions. Thus, this results in the language barrier between clinicians and artificial intelligence developers. Healthcare practitioners sometimes lack a basic understanding of artificial intelligence research because the approach is difficult for non-specialists to understand. Furthermore, many editors and reviewers of medical publications might not be familiar with the fundamental ideas behind these technologies, which may prevent journals from publishing high-quality artificial intelligence studies or, worse still, could allow for the publication of low-quality works. In this review, we aim to improve readers’ artificial intelligence literacy and critical thinking. As a result, we concentrated on what we consider the 10 most important qualities of artificial intelligence research: valid scientific purpose, high-quality data set, robust reference standard, robust input, no information leakage, optimal bias-variance tradeoff, proper model evaluation, proven clinical utility, transparent reporting, and open science. Before designing a study, one should have defined a sound scientific purpose. Then, it should be backed by a high-quality data set, robust input, and a solid reference standard. The artificial intelligence development pipeline should prevent information leakage. For the models, optimal bias-variance tradeoff should be achieved, and generalizability assessment must be adequately performed. The clinical value of the final models must also be established. After the study, thought should be given to transparency in publishing the process and results as well as open science for sharing data, code, and models. We hope this work may improve the artificial intelligence literacy and mindset of the readers.

Utilization of a cloud-based radiology analytics platform to monitor imaging volumes at a large tertiary center

Rationale and objective

In this study, we evaluate the ability of a novel cloud-based radiology analytics platform to continuously monitor imaging volumes at a large tertiary center following institutional protocol and policy changes.

Materials and methods

We evaluated response to environmental factors through the lens of the COVID-19 pandemic. Analysis involved 11 CT/18 MR imaging systems at a large tertiary center. A vendor neutral, cloud-based analytics tool (CBRAP) was used to retrospectively collect information via DICOM headers on imaging exams between Oct. 2019 to Aug. 2021. Exams were stratified by modality (CT or MRI) and organized by body region. Pre-pandemic scan volumes (Oct 2019-Feb. 2010) were compared with volumes during/after two waves of COVID-19 in Illinois (Mar. to May 2020 & Oct. to Dec. 2020) using a t-test or Mann-Whitney U test.

Results

The CBRAP was able to analyze 169,530 CT and 110,837 MR images, providing a detailed snapshot of baseline and post-pandemic CT and MR imaging across the radiology enterprise at our tertiary center. The CBRAP allowed for further subdivision in its reporting, showing monthly trends in average scan volumes specifically in the head, abdomen, spine, MSK, thorax, neck, GU system, or breast.

Conclusion

The CBRAP retrieved data for 300,000 + imaging exams across multiple modalities at a large tertiary center in a highly populated, urban environment. The ability to analyze large imaging volumes across multiple waves of COVID-19 and evaluate quality-improvement endeavors/imaging protocol changes displays the usefulness of the CBRAP as an advanced imaging analytics tool.

Automated Protocoling for MRI Exams-Challenges and Solutions

Automated protocoling for MRI examinations is an amendable target for workflow automation with artificial intelligence. However, there are still challenges to overcome for a successful and robust approach. These challenges are outlined and analyzed in this work. Through a literature review, we analyzed limitations of currently published approaches for automated protocoling. Then, we assessed these limitations quantitatively based on data from a private radiology practice. For this, we assessed the information content provided by the clinical indication by computing the overlap coefficients for the sets of ICD-10-coded admitting diagnoses of different MRI protocols. Additionally, we assessed the heterogeneity of protocol trees from three different MRI scanners based on the overlap coefficient, on MRI protocol and sequence level. Additionally, we applied sequence name standardization to demonstrate its effect on the heterogeneity assessment, i.e., the overlap coefficient, of different protocol trees. The overlap coefficient for the set of ICD-10-coded admitting diagnoses for different protocols ranges from 0.14 to 0.56 for brain/head MRI exams and 0.04 to 0.57 for spine exams. The overlap coefficient across the set of sequences used at two different scanners increases when applying sequence name standardization (from 0.81/0.86 to 0.93). Automated protocoling for MRI examinations has the potential to reduce the workload for radiologists. However, an automated protocoling approach cannot be solely based on admitting diagnosis as it does not provide sufficient information. Moreover, sequence name standardization increases the overlap coefficient across the set of sequences used at different scanners and therefore facilitates transfer learning.

Strategies for Dose Optimization: Views From Health Care Systems

BACKGROUND

Advances in CT have facilitated widespread use of medical imaging while increasing patient lifetime exposure to ionizing radiation.

PURPOSE

To describe dose optimization strategies used by health care organizations to optimize radiation dose and image quality.

MATERIALS AND METHODS

A qualitative study of semistructured interviews conducted with 26 leaders from 19 health care systems in the United States, Europe, and Japan. Interviews focused on strategies that were used to optimize radiation dose at the organizational level. A directed content analysis approach was used in data analysis.

RESULTS

Analysis identified seven organizational strategies used by these leaders for optimizing CT dose: (1) engaging radiologists and technologists, (2) establishing a CT dose committee, (3) managing organizational change, (4) providing leadership and support, (5) monitoring and benchmarking, (6) modifying CT protocols, and (7) changes in equipment and work rules.

CONCLUSIONS

Leaders in these health systems engaged in specific strategies to optimize CT dose within their organizations. The strategies address challenges health systems encounter in optimizing CT dose at the organizational level and offer an evolving framework for consideration in dose optimization efforts for enhancing safety and use of medical imaging.

An Image Quality-informed Framework for CT Characterization

Background Lack of standardization in CT protocol choice contributes to radiation dose variation. Purpose To create a framework to assess radiation doses within broad CT categories defined according to body region and clinical imaging indication and to cluster indications according to the dose required for sufficient image quality. Materials and Methods This was a retrospective study using Digital Imaging and Communications in Medicine metadata. CT examinations in adults from January 1, 2016 to December 31, 2019 from the University of California San Francisco International CT Dose Registry were grouped into 19 categories according to body region and required radiation dose levels. Five body regions had a single dose range (ie, extremities, neck, thoracolumbar spine, combined chest and abdomen, and combined thoracolumbar spine). Five additional regions were subdivided according to dose. Head, chest, cardiac, and abdomen each had low, routine, and high dose categories; combined head and neck had routine and high dose categories. For each category, the median and 75th percentile (ie, diagnostic reference level [DRL]) were determined for dose-length product, and the variation in dose within categories versus across categories was calculated and compared using an analysis of variance. Relative median and DRL (95% CI) doses comparing high dose versus low dose categories were calculated. Results Among 4.5 million examinations, the median and DRL doses varied approximately 10 times between categories compared with between indications within categories. For head, chest, abdomen, and cardiac (3 266 546 examinations [72%]), the relative median doses were higher in examinations assigned to the high dose categories than in examinations assigned to the low dose categories, suggesting the assignment of indications to the broad categories is valid (head, 3.4-fold higher [95% CI: 3.4, 3.5]; chest, 9.6 [95% CI: 9.3, 10.0]; abdomen, 2.4 [95% CI: 2.4, 2.5]; and cardiac, 18.1 [95% CI: 17.7, 18.6]). Results were similar for DRL doses (all P < .001). Conclusion Broad categories based on image quality requirements are a suitable framework for simplifying radiation dose assessment, according to expected variation between and within categories. © RSNA, 2021 See also the editorial by Mahesh in this issue.

Upstream Machine Learning in Radiology

Machine learning (ML) and Artificial intelligence (AI) has the potential to dramatically improve radiology practice at multiple stages of the imaging pipeline. Most of the attention has been garnered by applications focused on improving the end of the pipeline: image interpretation. However, this article reviews how AI/ML can be applied to improve upstream components of the imaging pipeline, including exam modality selection, hardware design, exam protocol selection, data acquisition, image reconstruction, and image processing. A breadth of applications and their potential for impact is shown across multiple imaging modalities, including ultrasound, computed tomography, and MRI.

Normalization of breast MRIs using cycle-consistent generative adversarial networks

OBJECTIVES

Dynamic Contrast Enhanced-Magnetic Resonance Imaging (DCE-MRI) is widely used to complement ultrasound examinations and x-ray mammography for early detection and diagnosis of breast cancer. However, images generated by various MRI scanners (e.g., GE Healthcare, and Siemens) differ both in intensity and noise distribution, preventing algorithms trained on MRIs from one scanner to generalize to data from other scanners. In this work, we propose a method to solve this problem by normalizing images between various scanners.

METHODS

MRI normalization is challenging because it requires normalizing intensity values and mapping noise distributions between scanners. We utilize a cycle-consistent generative adversarial network to learn a bidirectional mapping and perform normalization between MRIs produced by GE Healthcare and Siemens scanners in an unpaired setting. Initial experiments demonstrate that the traditional CycleGAN architecture struggles to preserve the anatomical structures of the breast during normalization. Thus, we propose two technical innovations in order to preserve both the shape of the breast as well as the tissue structures within the breast. First, we incorporate mutual information loss during training in order to ensure anatomical consistency. Second, we propose a modified discriminator architecture that utilizes a smaller field-of-view to ensure the preservation of finer details in the breast tissue.

RESULTS

Quantitative and qualitative evaluations show that the second innovation consistently preserves the breast shape and tissue structures while also performing the proper intensity normalization and noise distribution mapping.

CONCLUSION

Our results demonstrate that the proposed model can successfully learn a bidirectional mapping and perform normalization between MRIs produced by different vendors, potentially enabling improved diagnosis and detection of breast cancer. All the data used in this study are publicly available at https://wiki.cancerimagingarchive.net/pages/viewpage.action?pageId=70226903.

Making Radiomics More Reproducible across Scanner and Imaging Protocol Variations: A Review of Harmonization Methods

Radiomics converts medical images into mineable data via a high-throughput extraction of quantitative features used for clinical decision support. However, these radiomic features are susceptible to variation across scanners, acquisition protocols, and reconstruction settings. Various investigations have assessed the reproducibility and validation of radiomic features across these discrepancies. In this narrative review, we combine systematic keyword searches with prior domain knowledge to discuss various harmonization solutions to make the radiomic features more reproducible across various scanners and protocol settings. Different harmonization solutions are discussed and divided into two main categories: image domain and feature domain. The image domain category comprises methods such as the standardization of image acquisition, post-processing of raw sensor-level image data, data augmentation techniques, and style transfer. The feature domain category consists of methods such as the identification of reproducible features and normalization techniques such as statistical normalization, intensity harmonization, ComBat and its derivatives, and normalization using deep learning. We also reflect upon the importance of deep learning solutions for addressing variability across multi-centric radiomic studies especially using generative adversarial networks (GANs), neural style transfer (NST) techniques, or a combination of both. We cover a broader range of methods especially GANs and NST methods in more detail than previous reviews.

Enhanced Magnetic Resonance Image Synthesis with Contrast-Aware Generative Adversarial Networks

A magnetic resonance imaging (MRI) exam typically consists of the acquisition of multiple MR pulse sequences, which are required for a reliable diagnosis. With the rise of generative deep learning models, approaches for the synthesis of MR images are developed to either synthesize additional MR contrasts, generate synthetic data, or augment existing data for AI training. While current generative approaches allow only the synthesis of specific sets of MR contrasts, we developed a method to generate synthetic MR images with adjustable image contrast. Therefore, we trained a generative adversarial network (GAN) with a separate auxiliary classifier (AC) network to generate synthetic MR knee images conditioned on various acquisition parameters (repetition time, echo time, and image orientation). The AC determined the repetition time with a mean absolute error (MAE) of 239.6 ms, the echo time with an MAE of 1.6 ms, and the image orientation with an accuracy of 100%. Therefore, it can properly condition the generator network during training. Moreover, in a visual Turing test, two experts mislabeled 40.5% of real and synthetic MR images, demonstrating that the image quality of the generated synthetic and real MR images is comparable. This work can support radiologists and technologists during the parameterization of MR sequences by previewing the yielded MR contrast, can serve as a valuable tool for radiology training, and can be used for customized data generation to support AI training.

MR-contrast-aware image-to-image translations with generative adversarial networks

Purpose

A magnetic resonance imaging (MRI) exam typically consists of several sequences that yield different image contrasts. Each sequence is parameterized through multiple acquisition parameters that influence image contrast, signal-to-noise ratio, acquisition time, and/or resolution. Depending on the clinical indication, different contrasts are required by the radiologist to make a diagnosis. As MR sequence acquisition is time consuming and acquired images may be corrupted due to motion, a method to synthesize MR images with adjustable contrast properties is required.

Methods

Therefore, we trained an image-to-image generative adversarial network conditioned on the MR acquisition parameters repetition time and echo time. Our approach is motivated by style transfer networks, whereas the “style” for an image is explicitly given in our case, as it is determined by the MR acquisition parameters our network is conditioned on.

Results

This enables us to synthesize MR images with adjustable image contrast. We evaluated our approach on the fastMRI dataset, a large set of publicly available MR knee images, and show that our method outperforms a benchmark pix2pix approach in the translation of non-fat-saturated MR images to fat-saturated images. Our approach yields a peak signal-to-noise ratio and structural similarity of 24.48 and 0.66, surpassing the pix2pix benchmark model significantly.

Conclusion

Our model is the first that enables fine-tuned contrast synthesis, which can be used to synthesize missing MR-contrasts or as a data augmentation technique for AI training in MRI. It can also be used as basis for other image-to-image translation tasks within medical imaging, e.g., to enhance intermodality translation (MRI → CT) or 7 T image synthesis from 3 T MR images.

Implementing Shared, Standardized Imaging Protocols to Improve Cross-Enterprise Workflow and Quality

Value-based imaging requires appropriate utilization and the delivery of consistently high-quality imaging at an acceptable cost. Challenges include developing standardized imaging protocols, ensuring consistent application by technologists, and monitoring quality. These challenges increase as enterprises grow in geographical extent and complexity through mergers or partnerships. Our imaging enterprise includes a university hospital and clinic system, a large county hospital and healthcare system, and a pediatric hospital and health system. Studies across the three systems are interpreted by one large academic radiology group with expertise in various subspecialties. Our goals were as follows: (1) Standardize imaging protocols; (2) adapt the imaging protocols to specific modalities and available equipment; and (3) disseminate this knowledge across all of the sites of care. Our approach involved three components: (1) facilitation of imaging protocol definition across subspecialty radiologist teams; (2) creation of a database which links the clinical imaging protocols to the scanner/machine specific acquisition protocols; and (3) delivery of a protocol library and updates to all users regardless of location. We successfully instituted a process for the development, implementation, and delivery of standardized imaging protocols in a complex, multi-institutional healthcare system. Key elements for success include (1) a Project Champion who is able to articulate the importance of protocol standardization in improving the quality of patient care, (2) strong, effective modality-specific operational committees, (3) a Project Lead to manage the process efficiently, and (4) an electronic publishing of the protocol database to facilitate ease of access and use.

Autonomous magnetic resonance imaging

Access to Magnetic Resonance Imaging (MRI) across developing countries ranges from being prohibitive to scarcely available. For example, eleven countries in Africa have no scanners. One critical limitation is the absence of skilled manpower required for MRI usage. Some of these challenges can be mitigated using autonomous MRI (AMRI) operation. In this work, we demonstrate AMRI to simplify MRI workflow by separating the required intelligence and user interaction from the acquisition hardware. AMRI consists of three components: user node, cloud and scanner. The user node voice interacts with the user and presents the image reconstructions at the end of the AMRI exam. The cloud generates pulse sequences and performs image reconstructions while the scanner acquires the raw data. An AMRI exam is a custom brain screen protocol comprising of one T₁-, T₂- and T₂*-weighted exams. A neural network is trained to incorporate Intelligent Slice Planning (ISP) at the start of the AMRI exam. A Look Up Table was designed to perform intelligent protocolling by optimizing for contrast value while satisfying signal to noise ratio and acquisition time constraints. Data were acquired from four healthy volunteers for three experiments with different acquisition time constraints to demonstrate standard and self-administered AMRI. The source code is available online. AMRI achieved an average SNR of 22.86 ± 0.89 dB across all experiments with similar contrast. Experiment #3 (33.66% shorter table time than experiment #1) yielded a SNR of 21.84 ± 6.36 dB compared to 23.48 ± 7.95 dB for experiment #1. AMRI can potentially enable multiple scenarios to facilitate rapid prototyping and research and streamline radiological workflow. We believe we have demonstrated the first Autonomous MRI of the brain.