Advances in Breast Imaging - AI based CAD Online Training

Continue Continue Welcome 7 AI based CAD Online Training This training module covers the evidence for using artificial intelligence in breast screening and diagnostics and its clinical implementation. Completion time: 45 min. Screening Studies with AI 1 Exam Score 3 How Can AI be Used? 2 Master Template HOOD05162003052540 | Effective Date: 26-Nov-2019 Welcome Srechertext: Welcome to our Web-based training “Advances in Breast imaging.” The seventh and last module “AI-based CAD” covers the role of artificial intelligence in breast screening and diagnostics. It reviews the capabilities of breast AI systems, their differences from traditional CAD systems, and the evidence from scientific studies conducted using breast AI systems. Case discussions illustrate the use of AI in the clinical workflow. Introduction Prof. Dr. Nico Karssemeijer Screenpoint Medical Radboud University Medical Centre Nijmegen, The Netherlands Introduction 07-2_Screening-AI_01-Intro-to-lecture_V02.mp4 Screening Studies with AI Screening Studies with AI Prof. Dr. Nico Karssemeijer Screenpoint Medical Radboud University Medical Centre Nijmegen, The Netherlands Screening Studies with AI 07-2_Screening-AI_02-Studies_V02.mp4 How Can AI be Used? How Can AI be Used? Prof. Dr. Nico Karssemeijer Screenpoint Medical Radboud University Medical Centre Nijmegen, The Netherlands How Can AI be Used? 07-2_Screening-AI_03-How-can-AI-be-used_V02.mp4 Studies indicate that current AI systems can match human performance in reading mammograms. The use of AI can improve reading performance and reduce variability in breast screening. AI systems make the radiologist redundant. Which of the following statements about artificial intelligence (AI) is wrong? Knowledge Check Select the best answer. AI may help overcome problems with high workload involved in DBT reading. Knowledge Check Correct Incorrect The solution is: AI systems make the radiologist redundant. Incorrect The solution is: AI systems make the radiologist redundant. Incorrect The solution is: AI systems make the radiologist redundant. Explanation of the Correct Answer Continue AI systems have to be used in conjunction with a radiologist’s evaluation. They can improve reading performance and help overcome problems with high workload, but in the foreseeable future they will not be able to make the radiologist redundant. Explanation of the Correct Answer Exam Score Exam Score Prof. Dr. Nico Karssemeijer Screenpoint Medical Radboud University Medical Centre Nijmegen, The Netherlands Exam Score 07-2_Screening-AI_04-Exam-Score_V02.mp4 Case Discussions: Screening Cases Henny Rijken Previously Training and QA Director at LRCB - Dutch Expert Centre for Screening Clinical Expert, ScreenPoint Medical, Nijmegen, The Netherlands Screening cases assessed with the exam score and markings Case Discussions: Screening Cases transpara-screening-cases_rijken.mp4 Case Discussions: Mammography-Tomosynthesis Cases Mammography-Tomosynthesis cases assessed with the exam score and markings Henny Rijken Previously Training and QA Director at LRCB - Dutch Expert Centre for Screening Clinical Expert, ScreenPoint Medical, Nijmegen, The Netherlands Case Discussions: Mammography-Tomosynthesis Cases transpara-mammo-tomo-cases_rijken.mp4 Course Review Exam Score How Can AI be Used? Screening Studies with AI Course Review Here is a brief summary of the content covered in this training. Please select a topic to access the related information. 1 1 1 2 2 2 3 3 3 Course Review Exam Score Ranks exams in deciles from low to high risk of cancer prevalence Potential for patient triaging Independent of breast density Potential to reduce workload Potential to go from double to single reading for exams that are most likely normal How Can AI be Used? Traditional CAD approach: Limited scope: Reduce oversight errors Use after reading Should not be used to influence decisions Increases reading time Tends to increase recall rate Effectiveness is questionable AI alternatives: Support detection and interpretation Regions are marked with confidence level Concurrent use Triaging Exam score indicates level of suspicion Automated reading Breast cancer risk assessment Screening Studies with AI Studies indicate that current AI systems can match human performance in reading mammograms. Using AI can improve reading performance and reduce variability in breast screening. AI may help overcome problems with the high workload involved in DBT reading. AI systems must be used in conjunction with a radiologist: They don’t make radiologists redundant. Following the market introduction in 2009, several clinical studies on Digital Breast Tomosynthesis have proven the value of this technique for breast imaging. For a list of Siemens Healthineers-related reference studies shown in this WBT, please download the PDF file XP_Tomosynthesis_Bibliography.pdf under “Resources” in the upper right corner on this page. The questions and course reviews depicted in this WBT also refer to additional reference studies that are subject of the clinical expert presentations. Reference Studies Reference Studies The herein illustrated statements made by Siemens Healthineers customers and physicians are based on their own and discrete opinion. The speaker is responsible for obtaining permission to use any previously published figures or tables. The speaker is also responsible for obtaining permission to reproduce any photograph showing recognizable persons. The statements by Siemens Healthineers customers described herein are based on results that were achieved in the customer's unique setting. Since there is no "typical" setting and many variables exist there can be no guarantee that other customers will achieve the same results. Some products/features (here mentioned) are not necessarily commercially available in all countries. Due to regulatory reasons their availability cannot be guaranteed. Please contact your local Siemens Healthineers organization for further details. Copyright © Siemens Healthcare GmbH 2021 Siemens Healthineers Headquarters | Siemens Healthcare GmbH | Henkestr. 127 | 91052 Erlangen, Germany Telephone: +49 9131 84-0 | siemens-healthineers.com Disclaimer Disclaimer Assessment Assessment This assessment will test your retention of the presented content. Two out of three questions must be correct to complete the course and earn your certificate. You may repeat the assessment as many times as needed. Start Assessment Studies indicate that current AI systems can match human performance in reading mammograms. The use of AI can improve reading performance and reduce variability in breast screening. AI may help overcome problems with the high workloads involved in DBT reading. AI systems make the radiologist redundant. Which of the following statements about artificial intelligence (AI) is wrong? Question 1 of 3 Select the best answer. Question 1 of 3 Correct Incorrect You did not select the correct answer. Incorrect You did not select the correct answer. Incorrect You did not select the correct answer. Increases reading time. Tends to increase recall rate. Supports detection but not interpretation. Regions are marked with confidence level. Which of the following statements about artificial intelligence (AI) is correct? Question 2 of 3 Select the best answer. Should not be used to influence decisions. Question 2 of 3 Incorrect You did not select the correct answer. Correct Incorrect You did not select the correct answer. Incorrect You did not select the correct answer. Incorrect You did not select the correct answer. The Exam Score ranks every exam individually from low to high risk, and so a reduction in workload can’t be expected. The Exam Score depends on breast density. In triaging, Exam Score indicates the level of suspicion. Which of the following statements is true regarding Exam Score? Question 3 of 3 Select the best answer. Question 3 of 3 Correct Incorrect You did not select the correct answer. Incorrect You did not select the correct answer. Review Review Retry Assessment Results %Results.ScorePercent%% %Results.PassPercent%% Continue YOUR SCORE: PASSING SCORE: Results Slide You did not pass the course. Take time to review the assessment then select Retry to continue. Congratulations. You passed the course. Exit To access your Certificate of Completion, select the Launch button drop down on the course overview page. You can also access the certificate from your PEPconnect transcript. You have completed the training module no. 7 from training series “Advances in Breast Imaging”. Completion Sprechertext: Thank you for your attention! You have completed the training no. 7. With this last module our training series “Advances in Breast Imaging” ends. We hope you have taken a lot of helpful information with you for your daily work. We wish you all the best, much success and stay healthy. HOOD05162003171396 | Effective Date: 20-APR-2021 XP_Tomosynthesis_Bibliography.pdf 2D Two-dimensional 3D Three-dimensional AI Artificial Intelligence AUC Area Under the Curve BPE Background Parenchymal Enhancement CAD Computer Aided Detection CC Cranio-caudal CEM Contrast Enhanced Mammography CESM Contrast Enhanced Spectral Mammography CNB Core Needle Biopsy DBT Digital Breast Tomosynthesis DCIS Ductal carcinoma in situ DM Digital Mammography FFDM Full Field Digital Mammography FN False Negative FP False Positive i.m. intramuscular IDC Invasive ductal carcinoma ILC Invasive lobular carcinoma IN Insight = Sythetic Mammogram for wide-angle Tomosynthesis IT Information Technology MAG Magnification View MB Megabyte MG Mammogramm ML Medio-lateral MLO Medio-lateral-oblique MP Megapixel MRI Magnetic Resonance Imaging NCB Needle Core Biopsy NPV Negative Predictive Value PPV Positive Predictive Value pRCT Prospective, Randomized, Controlled Trial ROC Receiver Operating Characteristic STX Stereotaxis TiCEM Titanium Contrast Enhanced Mammography TN True Negative TP True Positive US Ultrasound wa-DBT wide-angle Digital Breast Tomosynthesis WBT Web-based Training 1.1 Welcome 1.2 Introduction 2.1 Screening Studies with AI 3.1 How Can AI be Used? 3.2 Knowledge Check 4.1 Exam Score 4.2 Case Discussions: Screening Cases 4.3 Case Discussions: Mammography-Tomosynthesis Cases 5.1 Course Review 5.2 Reference Studies 5.3 Disclaimer 6.1 Assessment 6.6 Completion