Longitudinal changes in the carotid arteries of head and neck cancer patients following radiation therapy: Results from a prospective serial imaging biomarker characterization study.

INTRODUCTION: We prospectively evaluated morphologic and functional changes in the carotid arteries of patients treated with unilateral neck radiation therapy (RT) for head and neck cancer. METHODS: Bilateral carotid artery duplex studies were performed at 0, 3, 6, 12, 18 months and 2, 3, 4, and 5 years following RT. Intima media thickness (IMT); global and regional circumferential, as well as radial strain, arterial elasticity, stiffness, and distensibility were calculated. RESULTS: Thirty-eight patients were included. A significant difference in the IMT from baseline between irradiated and unirradiated carotid arteries was detected at 18 months (median, 0.073 mm vs -0.003 mm; P = 0.014), which increased at 3 and 4 years (0.128 mm vs 0.013 mm, P = 0.016, and 0.177 mm vs 0.023 mm, P = 0.0002, respectively). A significant transient change was noted in global circumferential strain between the irradiated and unirradiated arteries at 6 months (median difference, -0.89, P = 0.023), which did not persist. No significant differences were detected in the other measures of elasticity, stiffness, and distensibility. CONCLUSIONS: Functional and morphologic changes of the carotid arteries detected by carotid ultrasound, such as changes in global circumferential strain at 6 months and carotid IMT at 18 months, may be useful for the early detection of radiation-induced carotid artery injury, can guide future research aiming to mitigate carotid artery stenosis, and should be considered for clinical surveillance survivorship recommendations after head and neck RT.

Determining The Role Of Radiation Oncologist Demographic Factors On Segmentation Quality: Insights From A Crowd-Sourced Challenge Using Bayesian Estimation.

BACKGROUND: Medical image auto-segmentation is poised to revolutionize radiotherapy workflows. The quality of auto-segmentation training data, primarily derived from clinician observers, is of utmost importance. However, the factors influencing the quality of these clinician-derived segmentations have yet to be fully understood or quantified. Therefore, the purpose of this study was to determine the role of common observer demographic variables on quantitative segmentation performance. METHODS: Organ at risk (OAR) and tumor volume segmentations provided by radiation oncologist observers from the Contouring Collaborative for Consensus in Radiation Oncology public dataset were utilized for this study. Segmentations were derived from five separate disease sites comprised of one patient case each: breast, sarcoma, head and neck (H&N), gynecologic (GYN), and gastrointestinal (GI). Segmentation quality was determined on a structure-by-structure basis by comparing the observer segmentations with an expert-derived consensus gold standard primarily using the Dice Similarity Coefficient (DSC); surface DSC was investigated as a secondary metric. Metrics were stratified into binary groups based on previously established structure-specific expert-derived interobserver variability (IOV) cutoffs. Generalized linear mixed-effects models using Markov chain Monte Carlo Bayesian estimation were used to investigate the association between demographic variables and the binarized segmentation quality for each disease site separately. Variables with a highest density interval excluding zero - loosely analogous to frequentist significance - were considered to substantially impact the outcome measure. RESULTS: After filtering by practicing radiation oncologists, 574, 110, 452, 112, and 48 structure observations remained for the breast, sarcoma, H&N, GYN, and GI cases, respectively. The median percentage of observations that crossed the expert DSC IOV cutoff when stratified by structure type was 55% and 31% for OARs and tumor volumes, respectively. Bayesian regression analysis revealed tumor category had a substantial negative impact on binarized DSC for the breast (coefficient mean ± standard deviation: -0.97 ± 0.20), sarcoma (-1.04 ± 0.54), H&N (-1.00 ± 0.24), and GI (-2.95 ± 0.98) cases. There were no clear recurring relationships between segmentation quality and demographic variables across the cases, with most variables demonstrating large standard deviations and wide highest density intervals. CONCLUSION: Our study highlights substantial uncertainty surrounding conventionally presumed factors influencing segmentation quality. Future studies should investigate additional demographic variables, more patients and imaging modalities, and alternative metrics of segmentation acceptability.

Deep-learning CT reconstruction in clinical scans of the abdomen: a systematic review and meta-analysis.

OBJECTIVE: To perform a systematic literature review and meta-analysis of the two most common commercially available deep-learning algorithms for CT. METHODS: We used PubMed, Scopus, Embase, and Web of Science to conduct systematic searches for studies assessing the most common commercially available deep-learning CT reconstruction algorithms: True Fidelity (TF) and Advanced intelligent Clear-IQ Engine (AiCE) in the abdomen of human participants since only these two algorithms currently have adequate published data for robust systematic analysis. RESULTS: Forty-four articles fulfilled inclusion criteria. 32 studies evaluated TF and 12 studies assessed AiCE. DLR algorithms produced images with significantly less noise (22-57.3% less than IR) but preserved a desirable noise texture with increased contrast-to-noise ratios and improved lesion detectability on conventional CT. These improvements with DLR were similarly noted in dual-energy CT which was only assessed for a single vendor. Reported radiation reduction potential was 35.1-78.5%. Nine studies assessed observer performance with the two dedicated liver lesion studies being performed on the same vendor reconstruction (TF). These two studies indicate preserved low contrast liver lesion detection (> 5 mm) at CTDIvol 6.8 mGy (BMI 23.5 kg/m2) to 12.2 mGy (BMI 29 kg/m2). If smaller lesion detection and improved lesion characterization is needed, a CTDIvol of 13.6-34.9 mGy is needed in a normal weight to obese population. Mild signal loss and blurring have been reported at high DLR reconstruction strengths. CONCLUSION: Deep learning reconstructions significantly improve image quality in CT of the abdomen. Assessment of other dose levels and clinical indications is needed. Careful choice of radiation dose levels is necessary, particularly for small liver lesion assessment.

Diagnostic radiology training for medical students - a Brazilian multicenter survey

ABSTRACT Objective This study aimed to assess diagnostic radiology training and exposure during medical school, from the perspective of medical students in Brazil. Methods In this multicenter study approved by the Institutional Review Board, medical students from multiple universities in Brazil filled out an online questionnaire regarding their perception about diagnostic radiology training during medical school, including knowledge and use of the American College of Radiology Appropriateness Criteria and their confidence level in interpreting common radiological findings. Medical students from different regions of Brazil were sent invitations to participate in the anonymous survey through radiology group emails initiated by radiology professors and a group of ambassadors representing different institutions. Informed consent was obtained electronically at the beginning of the survey. Results The survey demonstrated diagnostic radiology is frequently included in preclinical exams; however, radiology training during medical school was considered inadequate from the medical students´ perspective. Overall, radiological imaging teaching was provided by radiologists for more than half of the survey respondents; however, radiological imaging is frequently shown to students by non-radiologist physicians during case discussion rounds. Moreover, few respondents had a mandatory radiology training rotation during medical school. Conclusion This Brazilian medical student survey demonstrated that from the medical students' perspective, diagnostic radiology is an important subject in clinical practice; however, their radiology training and exposure are overall heterogeneous.

Vascular Anomaly Syndromes in the ISSVA Classification System: Imaging Findings and Role of Interventional Radiology in Management.

Vascular anomalies encompass a spectrum of tumors and malformations that can cause significant morbidity and mortality in children and adults. Use of the International Society for the Study of Vascular Anomalies (ISSVA) classification system is strongly recommended for consistency. Vascular anomalies can occur in isolation or in association with clinical syndromes that involve complex multifocal lesions affecting different organ systems. Thus, it is critical to be familiar with the differences and similarities among vascular anomalies to guide selection of the appropriate imaging studies and possible interventions. Syndromes associated with simple vascular malformations include hereditary hemorrhagic telangiectasia, blue rubber bleb nevus syndrome, Gorham-Stout disease, and primary lymphedema. Syndromes categorized as vascular malformations associated with other anomalies include Klippel-Trenaunay-Weber syndrome, Parkes Weber syndrome, Servelle-Martorell syndrome, Maffucci syndrome, macrocephaly-capillary malformation, CLOVES (congenital lipomatous overgrowth, vascular malformations, epidermal nevi, and scoliosis, skeletal, and spinal anomalies) syndrome, Proteus syndrome, Bannayan-Riley-Ruvalcaba syndrome, and CLAPO (capillary malformations of the lower lip, lymphatic malformations of the face and neck, asymmetry of the face and limbs, and partial or generalized overgrowth) syndrome. With PHACES (posterior fossa malformations, hemangiomas, arterial anomalies, cardiac defects and/or coarctation of the aorta, eye abnormalities, and sternal clefting or supraumbilical raphe) syndrome, infantile hemangiomas associated with other lesions occur. Diagnostic and interventional radiologists have important roles in diagnosing these conditions and administering image-guided therapies-embolization and sclerotherapy, and different ablation procedures in particular. The key imaging features of vascular anomaly syndromes based on the 2018 ISSVA classification system and the role of interventional radiology in the management of these syndromes are reviewed. Online supplemental material is available for this article. ©RSNA, 2022.

Neurofibromatosis from Head to Toe: What the Radiologist Needs to Know.

Neurofibromatosis type 1 (NF1) and neurofibromatosis type 2 (NF2) are autosomal dominant inherited neurocutaneous disorders or phakomatoses secondary to mutations in the NF1 and NF2 tumor suppressor genes, respectively. Although they share a common name, NF1 and NF2 are distinct disorders with a wide range of multisystem manifestations that include benign and malignant tumors. Imaging plays an essential role in diagnosis, surveillance, and management of individuals with NF1 and NF2. Therefore, it is crucial for radiologists to be familiar with the imaging features of NF1 and NF2 to allow prompt diagnosis and appropriate management. Key manifestations of NF1 include café-au-lait macules, axillary or inguinal freckling, neurofibromas or plexiform neurofibromas, optic pathway gliomas, Lisch nodules, and osseous lesions such as sphenoid dysplasia, all of which are considered diagnostic features of NF1. Other manifestations include focal areas of signal intensity in the brain, low-grade gliomas, interstitial lung disease, various abdominopelvic neoplasms, scoliosis, and vascular dysplasia. The various NF1-associated abdominopelvic neoplasms can be categorized by their cellular origin: neurogenic neoplasms, interstitial cells of Cajal neoplasms, neuroendocrine neoplasms, and embryonal neoplasms. Malignant peripheral nerve sheath tumors and intracranial tumors are the leading contributors to mortality in NF1. Classic manifestations of NF2 include schwannomas, meningiomas, and ependymomas. However, NF2 may have shared cutaneous manifestations with NF1. Lifelong multidisciplinary management is critical for patients with either disease. The authors highlight the genetics and molecular pathogenesis, clinical and pathologic features, imaging manifestations, and multidisciplinary management and surveillance of NF1 and NF2. Online supplemental material is available for this article. ©RSNA, 2022.

Radiomics in Abdominopelvic Solid-Organ Oncologic Imaging: Current Status.

Radiomics is the process of extraction of high-throughput quantitative imaging features from medical images. These features represent noninvasive quantitative biomarkers that go beyond the traditional imaging features visible to the human eye. This article first reviews the steps of the radiomics pipeline, including image acquisition, ROI selection and image segmentation, image preprocessing, feature extraction, feature selection, and model development and application. Current evidence for the application of radiomics in abdominopelvic solid-organ cancers is then reviewed. Applications including diagnosis, subtype determination, treatment response assessment, and outcome prediction are explored within the context of hepatobiliary and pancreatic cancer, renal cell carcinoma, prostate cancer, gynecologic cancer, and adrenal masses. This literature review focuses on the strongest available evidence, including systematic reviews, meta-analyses, and large multicenter studies. Limitations of the available literature are highlighted, including marked heterogeneity in radiomics methodology, frequent use of small sample sizes with high risk of overfitting, and lack of prospective design, external validation, and standardized radiomics workflow. Thus, although studies have laid a foundation that supports continued investigation into radiomics models, stronger evidence is needed before clinical adoption.

Artificial Intelligence in Diagnostic Radiology: Where Do We Stand, Challenges, and Opportunities.

ABSTRACT: Artificial intelligence (AI) is the most revolutionizing development in the health care industry in the current decade, with diagnostic imaging having the greatest share in such development. Machine learning and deep learning (DL) are subclasses of AI that show breakthrough performance in image analysis. They have become the state of the art in the field of image classification and recognition. Machine learning deals with the extraction of the important characteristic features from images, whereas DL uses neural networks to solve such problems with better performance. In this review, we discuss the current applications of machine learning and DL in the field of diagnostic radiology.Deep learning applications can be divided into medical imaging analysis and applications beyond analysis. In the field of medical imaging analysis, deep convolutional neural networks are used for image classification, lesion detection, and segmentation. Also used are recurrent neural networks when extracting information from electronic medical records and to augment the use of convolutional neural networks in the field of image classification. Generative adversarial networks have been explicitly used in generating high-resolution computed tomography and magnetic resonance images and to map computed tomography images from the corresponding magnetic resonance imaging. Beyond image analysis, DL can be used for quality control, workflow organization, and reporting.In this article, we review the most current AI models used in medical imaging research, providing a brief explanation of the various models described in the literature within the past 5 years. Emphasis is placed on the various DL models, as they are the most state-of-art in imaging analysis.

Current Concepts in Multi-Modality Imaging of Solid Tumor Angiogenesis.

There have been rapid advancements in cancer treatment in recent years, including targeted molecular therapy and the emergence of anti-angiogenic agents, which necessitate the need to quickly and accurately assess treatment response. The ideal tool is robust and non-invasive so that the treatment can be rapidly adjusted or discontinued based on efficacy. Since targeted therapies primarily affect tumor angiogenesis, morphological assessment based on tumor size alone may be insufficient, and other imaging modalities and features may be more helpful in assessing response. This review aims to discuss the biological principles of tumor angiogenesis and the multi-modality imaging evaluation of anti-angiogenic therapeutic responses.