Active Learning in Brain Tumor Segmentation with Uncertainty Sampling, Annotation Redundancy Restriction, and Data Initialization
Warren Alpert Medical School of Brown University(布朗大学沃伦·阿尔珀特医学院);Rhode Island Hospital(罗德岛医院);Perelman School of Medicine at the University of Pennsylvania(宾夕法尼亚大学佩雷尔曼医学院);Second Xiangya Hospital of Central South University(中南大学湘雅二医院);University of Virginia(弗吉尼亚大学);Johns Hopkins University(约翰霍普金斯大学);Central South University(中南大学);Brown University(布朗大学)
Daniel D Kim, Rajat S Chandra, Jian Peng, Jing Wu, Xue Feng, Michael Atalay, Chetan Bettegowda, Craig Jones, Haris Sair, Wei-hua Liao, Chengzhang Zhu, Beiji Zou… 展开作者
Daniel D Kim, Rajat S Chandra, Jian Peng, Jing Wu, Xue Feng, Michael Atalay, Chetan Bettegowda, Craig Jones, Haris Sair, Wei-hua Liao, Chengzhang Zhu, Beiji Zou, Li Yang, Anahita Fathi Kazerooni, Ali Nabavizadeh, Harrison X Bai, Zhicheng Jiao
备注:
22 pages, 3 figures, 3 tables, 1 supplementary data document. Submitted to Medical Physics in Jan 2023
Deep learning models have demonstrated great potential in medical 3D imaging, but their development is limited by the expensive, large volume of annotated data required. Active learning (AL) addresses this by training a model on a subset of the most informative data samples without compromising performance. We compared different AL strategies and propose a framework that minimizes the amount of data needed for state-of-the-art performance. 638 multi-institutional brain tumor MRI images were used to train a 3D U-net model and compare AL strategies. We investigated uncertainty sampling, annotation redundancy restriction, and initial dataset selection techniques. Uncertainty estimation techniques including Bayesian estimation with dropout, bootstrapping, and margins sampling were compared to random query. Strategies to avoid annotation redundancy by removing similar images within the to-be-annotated subset were considered as well. We determined the minimum amount of data necessary to achieve similar performance to the model trained on the full dataset (α = 0.1). A variance-based selection strategy using radiomics to identify the initial training dataset is also proposed. Bayesian approximation with dropout at training and testing showed similar results to that of the full data model with less than 20% of the training data (p=0.293) compared to random query achieving similar performance at 56.5% of the training data (p=0.814). Annotation redundancy restriction techniques achieved state-of-the-art performance at approximately 40%-50% of the training data. Radiomics dataset initialization had higher Dice with initial dataset sizes of 20 and 80 images, but improvements were not significant. In conclusion, we investigated various AL strategies with dropout uncertainty estimation achieving state-of-the-art performance with the least annotated data.
Improving Deep Learning Models for Pediatric Low-Grade Glioma Tumors Molecular Subtype Identification Using 3D Probability Distributions of Tumor Location
The Hospital for Sick Children (SickKids)(SickKids 病童医院);SickKids Research Institute(SickKids 研究所);University of Toronto(多伦多大学);Institute of Medical Science, University of Toronto(多伦多大学医学科学研究所);Vector Institute(矢量研究院);NVIDIA Deep Learning Institute(英伟达深度学习研究院)
Complementary Information Mutual Learning for Multimodality Medical Image Segmentation
School of Computer Science and Technology(计算机科学与技术学院);East China Normal University(华东师范大学);School of Data Science(数据科学学院);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳));Shenzhen Institute of Artificial Intelligence and Robotics for Society(深圳市人工智能与机器人研究院);Huashan Hospital(华山医院);Fudan University(复旦大学);Shanghai AI Laboratory(上海人工智能实验室);School of Software Engineering(软件工程学院);Shanghai Research Institute for Intelligent Autonomous Systems(上海智能自主系统研究院);Tongji University(同济大学)
Frequency-Aware Ensemble Learning for BraTS 2025 Pediatric Brain Tumor Segmentation
Institute of Natural Sciences, School of Mathematical Sciences, MOE-LSC, Shanghai Jiao Tong University(自然科学院、数学科学学院、MOE-LSC、上海交通大学);Department of Radiology, Huashan Hospital, Fudan University(放射科、华山医院、复旦大学);Department of Neurosurgery, Huashan Hospital, Fudan University(神经外科、华山医院、复旦大学)
Slot Attention with Re-Initialization and Self-Distillation
Department of Electrical Engineering and Automation(电气工程与自动化系);Aalto University(阿尔托大学);Department of Computer Science(计算机科学系);Center for Machine Vision and Signal Analysis(机器视觉与信号分析中心);University of Oulu(奥卢大学)
本文提出 DIAS 方法,通过槽重初始化和自蒸馏减少冗余槽,提升对象为中心学习的对象发现与识别性能。
Trustworthy Multi-phase Liver Tumor Segmentation via Evidence-based Uncertainty
Southeast University(东南大学);the First Affiliated Hospital, Zhejiang University School of Medicine(浙江大学医学院附属第一医院);Zhongda Hospital, School of Medicine, Southeast University(东南大学医学院附属中大医院)
Towards Reliable Pediatric Brain Tumor Segmentation: Task-Specific nnU-Net Enhancements
Institute of Natural Sciences, School of Mathematical Sciences, MOE-LSC, Shanghai Jiao Tong University(自然科学院,数学科学学院,MOE-LSC,上海交通大学);Department of Radiology, Huashan Hospital, Fudan University(放射科,华山医院,复旦大学);Department of Neurosurgery, Huashan Hospital, Fudan University(神经外科,华山医院,复旦大学)
PathMoE: Interpretable Multimodal Interaction Experts for Pediatric Brain Tumor Classification
University of Texas(德克萨斯大学);Dell Children’s Medical Center(德尔儿童医学中心);University of North Carolina at Chapel Hill(北卡罗来纳大学教堂山分校);University of Nevada, Reno(内华达大学里诺分校)
2D-3D Deformable Image Registration of Histology Slide and Micro-CT with ML-based Initialization
ImFusion GmbH(ImFusion有限公司);Technical University of Munich(慕尼黑工业大学);Institute of Pathology, Heinrich Heine University and University Hospital of Düsseldorf(海因里希·海涅大学与杜塞尔多夫大学医院病理研究所);Fraunhofer Institute for High-Speed Dynamics, Ernst-Mach-Institut(弗劳恩霍夫高速动力学研究所(恩斯特·马赫研究所))
CRISP -- Clustering-Based Redundancy-Reduced Instance Sampling for Pathology Case Representation and Retrieval
Kimia Lab, Department of Artificial Intelligence & Informatics, Mayo Clinic, Rochester, MN, USA(Kimia实验室,人工智能与信息学系,梅奥诊所,罗切斯特,明尼苏达州,美国);DICE Lab, Department of Electrical and Computer Engineering, University of Illinois Chicago, IL, USA(DICE实验室,电气与计算机工程系,伊利诺伊大学芝加哥分校,伊利诺伊州,美国);MD Kimia Lab, Department of Artificial Intelligence & Informatics, Mayo Clinic, Rochester, MN, USA(MD Kimia实验室,人工智能与信息学系,梅奥诊所,罗切斯特,明尼苏达州,美国);PhD Kimia Lab, Department of Artificial Intelligence & Informatics, Mayo Clinic, Rochester, MN, USA(PhD Kimia实验室,人工智能与信息学系,梅奥诊所,罗切斯特,明尼苏达州,美国);Division of Computational Pathology and Informatics, Mayo Clinic, Rochester, MN, USA(计算病理学与信息学部,梅奥诊所,罗切斯特,明尼苏达州,美国);Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA(实验室医学与病理学系,梅奥诊所,罗切斯特,明尼苏达州,美国);Department of Breast and Melanoma Surgical Oncology, Comprehensive Cancer Center, Mayo Clinic, Rochester, MN, USA(乳腺和黑色素瘤外科肿瘤学系,综合癌症中心,梅奥诊所,罗切斯特,明尼苏达州,美国);Department of Oncology, Comprehensive Cancer Center, Mayo Clinic, Rochester, MN, USA(肿瘤学系,综合癌症中心,梅奥诊所,罗切斯特,明尼苏达州,美国);PhD H.R. Tizhoosh
E2ENet: Dynamic Sparse Feature Fusion for Accurate and Efficient 3D Medical Image Segmentation
University of Twente(特温特大学);University of Luxembourg(卢森堡大学);Eindhoven University of Technology(埃因霍温理工大学);University of Oxford(牛津大学);University of Surrey(萨里大学)