Purpose: To develop and evaluate the accuracy of a multi-view deep learning approach to the analysis of high-resolution synthetic mammograms from digital breast tomosynthesis screening cases, and to assess the effect on accuracy of image resolution and training set size. Materials and Methods: In a retrospective study, 21,264 screening digital breast tomosynthesis (DBT) exams obtained at our institution were collected along with associated radiology reports. The 2D synthetic mammographic images from these exams, with varying resolutions and data set sizes, were used to train a multi-view deep convolutional neural network (MV-CNN) to classify screening images into BI-RADS classes (BI-RADS 0, 1 and 2) before evaluation on a held-out set of exams.
Results: Area under the receiver operating characteristic curve (AUC) for BI-RADS 0 vs non-BI-RADS 0 class was 0.912 for the MV-CNN trained on the full dataset. The model obtained accuracy of 84.8%, recall of 95.9% and precision of 95.0%. This AUC value decreased when the same model was trained with 50% and 25% of images (AUC = 0.877, P=0.010 and 0.834, P=0.009 respectively). Also, the performance dropped when the same model was trained using images that were under-sampled by 1/2 and 1/4 (AUC = 0.870, P=0.011 and 0.813, P=0.009 respectively).
Conclusion: This deep learning model classified high-resolution synthetic mammography scans into normal vs needing further workup using tens of thousands of high-resolution images. Smaller training data sets and lower resolution images both caused significant decrease in performance.
Deep Learning Predicts Mammographic Breast Density in Clinical Breast Ultrasound Images
University of Hawai’i at Mānoa（夏威夷大学马诺阿分校）；University of Hawai’i Cancer Center（夏威夷大学癌症中心）；The Queen’s Medical Center（皇后医疗中心）；Hawaiʻi Pacific Health（夏威夷太平洋健康）；Hawaiʻi Diagnostic Radiology Services（夏威夷诊断放射服务中心）
Automatic Application Level Set Approach in Detection Calcifications in Mammographic Image
National School of Medicine of Sfax（斯法克斯医学院）；CES-Computer, Electronic And Smart engineering systems design Laboratory（计算机、电子与智能工程系统设计实验室）；National School of Engineers of Sfax（斯法克斯工程师学院）
乳腺癌被认为是一个主要的健康问题，是世界上妇女死亡的最大原因。因此，在这十年中，乳腺癌是第二种最常见的癌症，就出现频率而言，也是第五种最常见的癌症相关死亡原因。为了减少放射科医师的工作量，已经提出了各种CAD系统;计算机辅助诊断（CADi）和计算机辅助检测（CADe）。在本文中，我们感兴趣的CADe工具，以帮助放射科医生检测癌症。建议的CADe是基于一个三步的工作流程，即检测，分析和分类。研究了基于边缘和区域准则的水平集方法的感兴趣区域自动检测问题。这种方法从放射科医生那里得到了良好的视觉信息。在此基础上，采用基于纹理特征的图像特征提取和基于多层感知（MLP）和k-近邻（KNN）的图像矢量分类方法，对图像进行了ACR（American College of Radiology）分类。此外，我们使用数字数据库筛选乳腺X射线摄影（DDSM）的实验和这些结果的准确性在60%和70%之间变化是可以接受的，必须加以改进，以帮助放射科医生。
Study of Different Deep Learning Approach with Explainable AI for Screening Patients with COVID-19 Symptoms: Using CT Scan and Chest X-ray Image Dataset
University of Oklahoma（俄克拉荷马大学）；University of Memphis（孟菲斯大学）；Marquette University（马凯特大学）
Cluster-Induced Mask Transformers for Effective Opportunistic Gastric Cancer Screening on Non-contrast CT Scans
DAMO Academy, Alibaba Group（阿里巴巴集团达摩院）；Peking University（北京大学）；Hupan Lab（湖畔实验室）；Changsha Institute for Computing and Digital Economy（长沙计算与数字经济研究院）；Guangdong Province People’s Hospital（广东省人民医院）；The First Affiliated Hospital of Zhejiang University（浙江大学医学院附属第一医院）
Lung cancer screening with low-dose CT scans using a deep learning approach
Arkansas State University（阿肯色州立大学）；University of Michigan（密歇根大学）；Ann Arbor Algorithm（安阿伯算法公司）；University of Arkansas at Pine Bluff（阿肯色大学松崖分校）；University of Arkansas for Medical Sciences（阿肯色大学医学院）
DSU-Net: An Attention-Enhanced Dense Skip U-Net for Breast Lesion Segmentation in Mammographic Images
Department of Biomedical Engineering, University of Wisconsin-Milwaukee（威斯康星大学密尔沃基分校生物医学工程系）；Department of Mechanical Engineering, Northern Arizona University（北亚利桑那大学机械工程系）
DeepVox and SAVE-CT: a contrast- and dose-independent 3D deep learning approach for thoracic aorta segmentation and aneurysm prediction using computed tomography scans
Albert Einstein Israeli Institute of Education and Research（阿尔伯特·爱因斯坦以色列教育与研究学院）；Health Innovation Techcenter（健康创新技术中心）；Department of Radiology（放射科）；Department of Vascular and Endovascular Surgery（血管与血管外科）
本研究提出 DeepVox 分割模型和 SAVE-CT 分类模型，在独立于对比剂和剂量的 3D CT 扫描上实现全自动胸主动脉瘤筛查，分割和分类测试集表现优异，有望降低死亡率并优化放射科评估队列。
RADIFUSION: A multi-radiomics deep learning based breast cancer risk prediction model using sequential mammographic images with image attention and bilateral asymmetry refinement
Monash University Malaysia（莫纳什大学马来西亚分校）；The University of Oklahoma（俄克拉荷马大学）；Karolinska Institute（卡罗林斯卡学院）；Karolinska University Hospital（卡罗林斯卡大学医院）；University of Malaya（马来亚大学）；The University of Melbourne（墨尔本大学）
CERN（欧洲核子研究组织）；Masaryk University（马萨里克大学）；National & Kapodistrian University of Athens（雅典国立与卡波迪斯特里亚大学）；Democritus University of Thrace（德谟克利特色雷斯大学）；University Hospital Brno（布尔诺大学医院）
Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans
University of Cambridge（剑桥大学）；AstraZeneca（阿斯利康）；University of Manchester（曼彻斯特大学）；Royal Papworth Hospital（皇家帕普沃斯医院）；Cambridge University Hospitals NHS Trust（剑桥大学医院国民医疗服务信托）
Feature extraction with regularized siamese networks for outlier detection: application to lesion screening in medical imaging
Univ Lyon（里昂大学）；INSA‐Lyon（里昂国立应用科学学院）；Université Claude Bernard Lyon 1（里昂第一大学）；UJM-Saint Etienne（圣艾蒂安让·莫奈大学）；CNRS（法国国家科学研究中心）；Inserm（法国国家健康与医学研究院）；CREATIS UMR 5220（CREATIS联合研究单位5220）；U1206（U1206研究单位）
A 3D Probabilistic Deep Learning System for Detection and Diagnosis of Lung Cancer Using Low-Dose CT Scans
Draper（德雷珀实验室）；Google（谷歌）
我们介绍了一种新的计算机辅助检测和诊断系统，肺癌筛查与低剂量CT扫描，产生有意义的概率评估。我们的系统完全基于3D卷积神经网络，并在公开的LUNA16和Kaggle Data Science Bowl挑战赛上实现了肺结节检测和恶性肿瘤分类任务的最新性能。虽然结节检测系统通常是自行设计和优化的，但我们发现，考虑检测和诊断组件之间的耦合是很重要的。利用这种耦合使我们能够开发一种端到端系统，该系统具有更高和更强大的性能，并消除了对结节检测假阳性减少阶段的需要。此外，我们在我们的深度学习系统中描述了模型的不确定性，这是肺部CT分析的第一个，并表明我们可以使用它来为结节检测和患者恶性肿瘤诊断提供校准良好的分类概率。这些由模型不确定性提供信息的校准概率可用于后续基于风险的诊断干预或疾病治疗决策，正如我们使用基于概率的患者转诊策略进一步改善我们的结果所证明的那样。
Harmony-Search and Otsu based System for Coronavirus Disease (COVID-19) Detection using Lung CT Scan Images
St. Joseph’s College of Engineering（圣约瑟夫工程学院）；Techno India College of Technology（泰克诺印度技术学院）；Shantou University（汕头大学）；Cairo University（开罗大学）；University of South Dakota（南达科他大学）
A New Computer-Aided Diagnosis System with Modified Genetic Feature Selection for BI-RADS Classification of Breast Masses in Mammograms
Beijing Institute of Technology（北京理工大学）；Université 8 Mai 1945 Guelma（盖勒马8 Mai 1945大学）；National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College（中国医学科学院北京协和医学院肿瘤医院国家癌症中心国家肿瘤临床研究中心）
Mammographic Breast Positioning Assessment via Deep Learning
Istanbul Technical University（伊斯坦布尔理工大学）；Acibadem Healthcare Group（阿西巴德姆医疗集团）；Acibadem Mehmet Ali Aydinlar University（阿西巴德姆穆罕默德·阿里·艾丁拉尔大学）；Sisli Hamidiye Etfal Training and Research Hospital（锡什利哈米迪耶埃特法尔培训与研究医院）
Three Applications of Conformal Prediction for Rating Breast Density in Mammography
Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital（马萨诸塞州总医院阿西诺拉·A·马丁诺斯生物医学成像中心）；Memorial Sloan Kettering Cancer Center（纪念斯隆·凯特琳癌症中心）；University of Colorado School of Medicine（科罗拉多大学医学院）；Harvard Medical School（哈佛医学院）
Artificial Intelligence-Informed Handheld Breast Ultrasound for Screening: A Systematic Review of Diagnostic Test Accuracy
University of Hawai’i at Mānoa（夏威夷大学马诺阿分校）；University of Hawai’i Cancer Center（夏威夷大学癌症中心）；Karolinska Institutet（卡罗林斯卡学院）；Karolinska University Hospital（卡罗林斯卡大学医院）
Coronary artery calcification assessment in National Lung Screening Trial CT images (DeepCAC2)
Mass General Brigham, Harvard Medical School（马萨诸塞综合 Brigham，哈佛医学院）；Maastricht University（马斯特里赫特大学）；Dana-Farber Cancer Institute, Brigham and Women’s Hospital, Harvard Medical School（达纳-法伯癌症研究所，布莱姆妇女医院，哈佛医学院）；Goethe University（歌德大学）；Massachusetts General Hospital and Harvard Medical School（马萨诸塞州总医院和哈佛医学院）；Brigham and Women’s Hospital, Harvard Medical School（布莱姆妇女医院，哈佛医学院）
