Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes最新文献

{"title":"Applications of Artificial Intelligence in Musculoskeletal Imaging: From the Request to the Report.","authors":"Natalia Gorelik,&nbsp;Soterios Gyftopoulos","doi":"10.1177/0846537120947148","DOIUrl":"https://doi.org/10.1177/0846537120947148","url":null,"abstract":"<p><p>Artificial intelligence (AI) will transform every step in the imaging value chain, including interpretive and noninterpretive components. Radiologists should familiarize themselves with AI developments to become leaders in their clinical implementation. This article explores the impact of AI through the entire imaging cycle of musculoskeletal radiology, from the placement of the requisition to the generation of the report, with an added Canadian perspective. Noninterpretive tasks which may be assisted by AI include the ordering of appropriate imaging tests, automatic exam protocoling, optimized scheduling, shorter magnetic resonance imaging acquisition time, computed tomography imaging with reduced artifact and radiation dose, and new methods of generation and utilization of radiology reports. Applications of AI for image interpretation consist of the determination of bone age, body composition measurements, screening for osteoporosis, identification of fractures, evaluation of segmental spine pathology, detection and temporal monitoring of osseous metastases, diagnosis of primary bone and soft tissue tumors, and grading of osteoarthritis.</p>","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"45-59"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537120947148","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"38276528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Imaging Database Preparation for Machine Learning.","authors":"Christian B van der Pol,&nbsp;An Tang","doi":"10.1177/0846537120967720","DOIUrl":"https://doi.org/10.1177/0846537120967720","url":null,"abstract":"Machine learning will revolutionize the practice of radiology in many ways; however, the change will be unpredictable and incremental. There are several challenges to the development of machine learning models for interpretation of medical imaging and perhaps none greater than obtaining large and welllabeled data sets. High-quality and consistently labeled data sets are needed for training if a machine is expected to accomplish tasks traditionally performed by radiologists. Such data sets are generally not readily available due to the time, cost, and expertise to collect, clean, and annotate large medical imaging data sets that include a spectrum of disease acquired with a variety of technical parameters on scanners from different manufacturers. At the onset, when considering a machine learning project, it is important to ensure there is a sufficient sample size to address the research question. A large data set with varied feature combinations is desirable. The optimal sample size depends on the task (ie, classification, detection, or segmentation). A rule of thumb is to have at least 10 times as many samples as the total number of features (or events) being assessed. For example, if a machine will be trained using magnetic resonance imaging (MRI) exams to identify the presence of the Liver Imaging Reporting and Data System (LI-RADS) features of arterial phase hyperenhancement and nonperipheral ‘‘washout,’’ given 2 features are being assessed, at least 20 MRI exams demonstrating these features are required when using classic machine learning. Some classes of machine learning methods such as deep learning require larger sample sizes to perform better than classic machine learning techniques. This is necessary to minimize the risk of ‘‘overfitting,’’ which occurs when a model does not generalize well beyond training data. Using a large data set is the ideal way to reduce the risk of overfitting. Otherwise, data augmentation techniques can be used if a larger data set is needed with many techniques available including, for example, data synthesis using generative adversarial networks. There are several statistical techniques that can also be used to minimize the risk of overfitting including cross-validation, early stopping, regularization, and ensembling. Using data from a variety of sources, rather than data obtained from one machine at one site, can improve algorithm generalizability by reducing sampling bias. Many open source data sets are now available which provide another option for obtaining images. Once the scope of a machine learning project has been established and concerns regarding sufficient data set size are addressed, the next step is data preparation or data curation. This involves addressing incomplete data, duplicate data, and other data set formatting issues and can be a very timeconsuming step, especially for large data sets. A team of radiologists may need to annotate imaging exams independently and in consensus to establish t","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"9-10"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537120967720","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"38495088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Canadian Association of Radiologists White Paper on De-identification of Medical Imaging: Part 2, Practical Considerations.","authors":"William Parker,&nbsp;Jacob L Jaremko,&nbsp;Mark Cicero,&nbsp;Marleine Azar,&nbsp;Khaled El-Emam,&nbsp;Bruce G Gray,&nbsp;Casey Hurrell,&nbsp;Flavie Lavoie-Cardinal,&nbsp;Benoit Desjardins,&nbsp;Andrea Lum,&nbsp;Lori Sheremeta,&nbsp;Emil Lee,&nbsp;Caroline Reinhold,&nbsp;An Tang,&nbsp;Rebecca Bromwich","doi":"10.1177/0846537120967345","DOIUrl":"https://doi.org/10.1177/0846537120967345","url":null,"abstract":"<p><p>The application of big data, radiomics, machine learning, and artificial intelligence (AI) algorithms in radiology requires access to large data sets containing personal health information. Because machine learning projects often require collaboration between different sites or data transfer to a third party, precautions are required to safeguard patient privacy. Safety measures are required to prevent inadvertent access to and transfer of identifiable information. The Canadian Association of Radiologists (CAR) is the national voice of radiology committed to promoting the highest standards in patient-centered imaging, lifelong learning, and research. The CAR has created an AI Ethical and Legal standing committee with the mandate to guide the medical imaging community in terms of best practices in data management, access to health care data, de-identification, and accountability practices. Part 2 of this article will inform CAR members on the practical aspects of medical imaging de-identification, strengths and limitations of de-identification approaches, list of de-identification software and tools available, and perspectives on future directions.</p>","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"25-34"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537120967345","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"38559290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial Intelligence Solutions for Analysis of X-ray Images.","authors":"Scott J Adams,&nbsp;Robert D E Henderson,&nbsp;Xin Yi,&nbsp;Paul Babyn","doi":"10.1177/0846537120941671","DOIUrl":"https://doi.org/10.1177/0846537120941671","url":null,"abstract":"<p><p>Artificial intelligence (AI) presents a key opportunity for radiologists to improve quality of care and enhance the value of radiology in patient care and population health. The potential opportunity of AI to aid in triage and interpretation of conventional radiographs (X-ray images) is particularly significant, as radiographs are the most common imaging examinations performed in most radiology departments. Substantial progress has been made in the past few years in the development of AI algorithms for analysis of chest and musculoskeletal (MSK) radiographs, with deep learning now the dominant approach for image analysis. Large public and proprietary image data sets have been compiled and have aided the development of AI algorithms for analysis of radiographs, many of which demonstrate accuracy equivalent to radiologists for specific, focused tasks. This article describes (1) the basis for the development of AI solutions for radiograph analysis, (2) current AI solutions to aid in the triage and interpretation of chest radiographs and MSK radiographs, (3) opportunities for AI to aid in noninterpretive tasks related to radiographs, and (4) considerations for radiology practices selecting AI solutions for radiograph analysis and integrating them into existing IT systems. Although comprehensive AI solutions across modalities have yet to be developed, institutions can begin to select and integrate focused solutions which increase efficiency, increase quality and patient safety, and add value for their patients.</p>","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"60-72"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537120941671","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"38232643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial Intelligence and Deep Learning in Neuroradiology: Exploring the New Frontier.","authors":"Hussam Kaka,&nbsp;Euan Zhang,&nbsp;Nazir Khan","doi":"10.1177/0846537120954293","DOIUrl":"https://doi.org/10.1177/0846537120954293","url":null,"abstract":"<p><p>There have been many recently published studies exploring machine learning (ML) and deep learning applications within neuroradiology. The improvement in performance of these techniques has resulted in an ever-increasing number of commercially available tools for the neuroradiologist. In this narrative review, recent publications exploring ML in neuroradiology are assessed with a focus on several key clinical domains. In particular, major advances are reviewed in the context of: (1) intracranial hemorrhage detection, (2) stroke imaging, (3) intracranial aneurysm screening, (4) multiple sclerosis imaging, (5) neuro-oncology, (6) head and tumor imaging, and (7) spine imaging.</p>","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"35-44"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537120954293","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"38396410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiology Artificial Intelligence: Bringing Theory to Clinical Practice.","authors":"Jaron Chong,&nbsp;Michael N Patlas","doi":"10.1177/0846537120959875","DOIUrl":"https://doi.org/10.1177/0846537120959875","url":null,"abstract":"","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"6"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537120959875","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"38408046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiogenomic Models Using Machine Learning Techniques to Predict EGFR Mutations in Non-Small Cell Lung Cancer.","authors":"Jay Kumar Raghavan Nair,&nbsp;Umar Abid Saeed,&nbsp;Connor C McDougall,&nbsp;Ali Sabri,&nbsp;Bojan Kovacina,&nbsp;B V S Raidu,&nbsp;Riaz Ahmed Khokhar,&nbsp;Stephan Probst,&nbsp;Vera Hirsh,&nbsp;Jeffrey Chankowsky,&nbsp;Léon C Van Kempen,&nbsp;Jana Taylor","doi":"10.1177/0846537119899526","DOIUrl":"https://doi.org/10.1177/0846537119899526","url":null,"abstract":"<p><strong>Background: </strong>The purpose of this study was to build radiogenomics models from texture signatures derived from computed tomography (CT) and ¹⁸F-FDG PET-CT (FDG PET-CT) images of non-small cell lung cancer (NSCLC) with and without epidermal growth factor receptor (<i>EGFR</i>) mutations.</p><p><strong>Methods: </strong>Fifty patients diagnosed with NSCLC between 2011 and 2015 and with known <i>EGFR</i> mutation status were retrospectively identified. Texture features extracted from pretreatment CT and FDG PET-CT images by manual contouring of the primary tumor were used to develop multivariate logistic regression (LR) models to predict <i>EGFR</i> mutations in exon 19 and exon 20.</p><p><strong>Results: </strong>An LR model evaluating FDG PET-texture features was able to differentiate <i>EGFR</i> mutant from wild type with an area under the curve (AUC), sensitivity, specificity, and accuracy of 0.87, 0.76, 0.66, and 0.71, respectively. The model derived from CT texture features had an AUC, sensitivity, specificity, and accuracy of 0.83, 0.84, 0.73, and 0.78, respectively. FDG PET-texture features that could discriminate between mutations in <i>EGFR</i> exon 19 and 21 demonstrated AUC, sensitivity, specificity, and accuracy of 0.86, 0.84, 0.73, and 0.78, respectively. Based on CT texture features, the AUC, sensitivity, specificity, and accuracy were 0.75, 0.81, 0.69, and 0.75, respectively.</p><p><strong>Conclusion: </strong>Non-small cell lung cancer texture analysis using FGD-PET and CT images can identify tumors with mutations in <i>EGFR</i>. Imaging signatures could be valuable for pretreatment assessment and prognosis in precision therapy.</p>","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"109-119"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537119899526","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"37648724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring the Role of Artificial Intelligence in an Emergency and Trauma Radiology Department.","authors":"Sabeena Jalal,&nbsp;William Parker,&nbsp;Duncan Ferguson,&nbsp;Savvas Nicolaou","doi":"10.1177/0846537120918338","DOIUrl":"https://doi.org/10.1177/0846537120918338","url":null,"abstract":"<p><p>Emergency and trauma radiologists, emergency department's physicians and nurses, researchers, departmental leaders, and health policymakers have attempted to discover efficient approaches to enhance the provision of quality patient care. There are increasing expectations for radiology practices to deliver a dedicated emergency radiology service providing 24/7/365 on-site attending radiologist coverage. Emergency radiologists (ERs) are pressed to meet the demand of increased imaging volume, provide accurate reports, maintain a lower proportion of discrepancy rate, and with a rapid report turnaround time of finalized reports. Thus, rendering the radiologists overburdened. The demand for an increased efficiency in providing quality care to acute patients has led to the emergence of artificial intelligence (AI) in the field. AI can be used to assist emergency and trauma radiologists deal with the ever-increasing imaging volume and workload, as AI methods have typically demonstrated a variety of applications in medical image analysis and interpretation, albeit most programs are in a training or validation phase. This article aims to offer an evidence-based discourse about the evolving role of artificial intelligence in assisting the imaging pathway in an emergency and trauma radiology department. We hope to generate a multidisciplinary discourse that addresses the technical processes, the challenges in the labour-intensive process of training, validation and testing of an algorithm, the need for emphasis on ethics, and how an emergency radiologist's role is pivotal in the execution of AI-guided systems within the context of an emergency and trauma radiology department. This exploratory narrative serves the present-day health leadership's information needs by proposing an AI supported and radiologist centered framework depicting the work flow within a department. It is suspected that the use of such a framework, if efficacious, could provide considerable benefits for patient safety and quality of care provided. Additionally, alleviating radiologist burnout and decreasing healthcare costs over time.</p>","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"167-174"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537120918338","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"37850937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Utilizing Artificial Intelligence for Head and Neck Cancer Outcomes Prediction From Imaging.","authors":"Tricia Chinnery,&nbsp;Andrew Arifin,&nbsp;Keng Yeow Tay,&nbsp;Andrew Leung,&nbsp;Anthony C Nichols,&nbsp;David A Palma,&nbsp;Sarah A Mattonen,&nbsp;Pencilla Lang","doi":"10.1177/0846537120942134","DOIUrl":"https://doi.org/10.1177/0846537120942134","url":null,"abstract":"<p><p>Artificial intelligence (AI)-based models have become a growing area of interest in predictive medicine and have the potential to aid physician decision-making to improve patient outcomes. Imaging and radiomics play an increasingly important role in these models. This review summarizes recent developments in the field of radiomics for AI in head and neck cancer. Prediction models for oncologic outcomes, treatment toxicity, and pathological findings have all been created. Exploratory studies are promising; however, validation studies that demonstrate consistency, reproducibility, and prognostic impact remain uncommon. Prospective clinical trials with standardized procedures are required for clinical translation.</p>","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"73-85"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537120942134","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"38212122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effects of Deep Learning Reconstruction Technique in High-Resolution Non-contrast Magnetic Resonance Coronary Angiography at a 3-Tesla Machine.","authors":"Yasuhiro Yokota,&nbsp;Chika Takeda,&nbsp;Masafumi Kidoh,&nbsp;Seitaro Oda,&nbsp;Ryo Aoki,&nbsp;Kenichi Ito,&nbsp;Kosuke Morita,&nbsp;Kentaro Haraoka,&nbsp;Yuichi Yamashita,&nbsp;Hitoshi Iizuka,&nbsp;Shingo Kato,&nbsp;Kenichi Tsujita,&nbsp;Osamu Ikeda,&nbsp;Yasuyuki Yamashita,&nbsp;Daisuke Utsunomiya","doi":"10.1177/0846537119900469","DOIUrl":"https://doi.org/10.1177/0846537119900469","url":null,"abstract":"<p><strong>Purpose: </strong>To evaluate the effects of deep learning reconstruction (DLR) in qualitative and quantitative image quality of non-contrast magnetic resonance coronary angiography (MRCA).</p><p><strong>Methods: </strong>Ten healthy volunteers underwent conventional MRCA (C-MRCA) and high-resolution (HR) MRCA on a 3T magnetic resonance imaging with a voxel size of 1.8 × 1.1 × 1.7 mm³ and 1.8 × 0.6 × 1.0 mm³, respectively, for C-MRCA and HR-MRCA. High-resolution magnetic resonance coronary angiography was also reconstructed with the DLR technique (DLR-HR-MRCA). We compared the contrast-to-noise ratio (CNR) and visual evaluation scores for vessel sharpness and traceability of proximal and distal coronary vessels on a 4-point scale among 3 image series.</p><p><strong>Results: </strong>The vascular CNR value on the C-MRCA and the DLR-HR-MRCA was significantly higher than that on the HR-MRCA in the proximal and distal coronary arteries (13.9 ± 6.4, 11.3 ± 4.4, and 7.8 ± 2.6 for C-MRCA, DLR-HR-MRCA, and HR-MRCA, <i>P</i> < .05, respectively). Mean visual evaluation scores for the vessel sharpness and traceability of proximal and distal coronary vessels were significantly higher on the HR-DLR-MRCA than the C-MRCA (<i>P</i> < .05, respectively).</p><p><strong>Conclusion: </strong>Deep learning reconstruction significantly improved the CNR of coronary arteries on HR-MRCA, resulting in both higher visual image quality and better vessel traceability compared with C-MRCA.</p>","PeriodicalId":444006,"journal":{"name":"Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes","volume":" ","pages":"120-127"},"PeriodicalIF":3.1,"publicationDate":"2021-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1177/0846537119900469","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"37654578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}