Pub Date : 2020-10-04DOI: 10.1007/978-3-030-61166-8_12
Colin B. Hansen, V. Nath, Riqiang Gao, Camilo Bermúdez, Yuankai Huo, K. Sandler, P. Massion, J. Blume, T. Lasko, B. Landman
{"title":"Semi-supervised Machine Learning with MixMatch and Equivalence Classes","authors":"Colin B. Hansen, V. Nath, Riqiang Gao, Camilo Bermúdez, Yuankai Huo, K. Sandler, P. Massion, J. Blume, T. Lasko, B. Landman","doi":"10.1007/978-3-030-61166-8_12","DOIUrl":"https://doi.org/10.1007/978-3-030-61166-8_12","url":null,"abstract":"","PeriodicalId":93329,"journal":{"name":"Lecture notes-monograph series","volume":"12446 1","pages":"112-121"},"PeriodicalIF":0.0,"publicationDate":"2020-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45661582","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}
Pub Date : 2020-10-01Epub Date: 2020-09-26DOI: 10.1007/978-3-030-60548-3_17
Samuel W Remedios, John A Butman, Bennett A Landman, Dzung L Pham
Multi-site training methods for artificial neural networks are of particular interest to the medical machine learning community primarily due to the difficulty of data sharing between institutions. However, contemporary multi-site techniques such as weight averaging and cyclic weight transfer make theoretical sacrifices to simplify implementation. In this paper, we implement federated gradient averaging (FGA), a variant of federated learning without data transfer that is mathematically equivalent to single site training with centralized data. We evaluate two scenarios: a simulated multi-site dataset for handwritten digit classification with MNIST and a real multi-site dataset with head CT hemorrhage segmentation. We compare federated gradient averaging to single site training, federated weight averaging (FWA), and cyclic weight transfer. In the MNIST task, we show that training with FGA results in a weight set equivalent to centralized single site training. In the hemorrhage segmentation task, we show that FGA achieves on average superior results to both FWA and cyclic weight transfer due to its ability to leverage momentum-based optimization.
{"title":"Federated Gradient Averaging for Multi-Site Training with Momentum-Based Optimizers.","authors":"Samuel W Remedios, John A Butman, Bennett A Landman, Dzung L Pham","doi":"10.1007/978-3-030-60548-3_17","DOIUrl":"https://doi.org/10.1007/978-3-030-60548-3_17","url":null,"abstract":"<p><p>Multi-site training methods for artificial neural networks are of particular interest to the medical machine learning community primarily due to the difficulty of data sharing between institutions. However, contemporary multi-site techniques such as weight averaging and cyclic weight transfer make theoretical sacrifices to simplify implementation. In this paper, we implement federated gradient averaging (FGA), a variant of federated learning without data transfer that is mathematically equivalent to single site training with centralized data. We evaluate two scenarios: a simulated multi-site dataset for handwritten digit classification with MNIST and a real multi-site dataset with head CT hemorrhage segmentation. We compare federated gradient averaging to single site training, federated weight averaging (FWA), and cyclic weight transfer. In the MNIST task, we show that training with FGA results in a weight set equivalent to centralized single site training. In the hemorrhage segmentation task, we show that FGA achieves on average superior results to both FWA and cyclic weight transfer due to its ability to leverage momentum-based optimization.</p>","PeriodicalId":93329,"journal":{"name":"Lecture notes-monograph series","volume":"12444 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8442829/pdf/nihms-1687698.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"39423625","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Colin B Hansen, Vishwesh Nath, Riqiang Gao, Camilo Bermudez, Yuankai Huo, Kim L Sandler, Pierre P Massion, Jeffrey D Blume, Thomas A Lasko, Bennett A Landman
Semi-supervised methods have an increasing impact on computer vision tasks to make use of scarce labels on large datasets, yet these approaches have not been well translated to medical imaging. Of particular interest, the MixMatch method achieves significant performance improvement over popular semi-supervised learning methods with scarce labels in the CIFAR-10 dataset. In a complementary approach, Nullspace Tuning on equivalence classes offers the potential to leverage multiple subject scans when the ground truth for the subject is unknown. This work is the first to (1) explore MixMatch with Nullspace Tuning in the context of medical imaging and (2) characterize the impacts of the methods with diminishing labels. We consider two distinct medical imaging domains: skin lesion diagnosis and lung cancer prediction. In both cases we evaluate models trained with diminishing labeled data using supervised, MixMatch, and Nullspace Tuning methods as well as MixMatch with Nullspace Tuning together. MixMatch with Nullspace Tuning together is able to achieve an AUC of 0.755 in lung cancer diagnosis with only 200 labeled subjects on the National Lung Screening Trial and a balanced multi-class accuracy of 77% with only 779 labeled examples on HAM10000. This performance is similar to that of the fully supervised methods when all labels are available. In advancing data driven methods in medical imaging, it is important to consider the use of current state-of-the-art semi-supervised learning methods from the greater machine learning community and their impact on the limitations of data acquisition and annotation.
{"title":"Semi-supervised Machine Learning with MixMatch and Equivalence Classes.","authors":"Colin B Hansen, Vishwesh Nath, Riqiang Gao, Camilo Bermudez, Yuankai Huo, Kim L Sandler, Pierre P Massion, Jeffrey D Blume, Thomas A Lasko, Bennett A Landman","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Semi-supervised methods have an increasing impact on computer vision tasks to make use of scarce labels on large datasets, yet these approaches have not been well translated to medical imaging. Of particular interest, the MixMatch method achieves significant performance improvement over popular semi-supervised learning methods with scarce labels in the CIFAR-10 dataset. In a complementary approach, Nullspace Tuning on equivalence classes offers the potential to leverage multiple subject scans when the ground truth for the subject is unknown. This work is the first to (1) explore MixMatch with Nullspace Tuning in the context of medical imaging and (2) characterize the impacts of the methods with diminishing labels. We consider two distinct medical imaging domains: skin lesion diagnosis and lung cancer prediction. In both cases we evaluate models trained with diminishing labeled data using supervised, MixMatch, and Nullspace Tuning methods as well as MixMatch with Nullspace Tuning together. MixMatch with Nullspace Tuning together is able to achieve an AUC of 0.755 in lung cancer diagnosis with only 200 labeled subjects on the National Lung Screening Trial and a balanced multi-class accuracy of 77% with only 779 labeled examples on HAM10000. This performance is similar to that of the fully supervised methods when all labels are available. In advancing data driven methods in medical imaging, it is important to consider the use of current state-of-the-art semi-supervised learning methods from the greater machine learning community and their impact on the limitations of data acquisition and annotation.</p>","PeriodicalId":93329,"journal":{"name":"Lecture notes-monograph series","volume":"12446 ","pages":"112-121"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8388309/pdf/nihms-1647070.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"39366463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-11DOI: 10.1007/978-3-030-32692-0_36
Riqiang Gao, Yuankai Huo, S. Bao, Yucheng Tang, S. Antic, Emily S. Epstein, A. Balar, S. Deppen, Alexis B. Paulson, K. Sandler, P. Massion, B. Landman
{"title":"Distanced LSTM: Time-Distanced Gates in Long Short-Term Memory Models for Lung Cancer Detection","authors":"Riqiang Gao, Yuankai Huo, S. Bao, Yucheng Tang, S. Antic, Emily S. Epstein, A. Balar, S. Deppen, Alexis B. Paulson, K. Sandler, P. Massion, B. Landman","doi":"10.1007/978-3-030-32692-0_36","DOIUrl":"https://doi.org/10.1007/978-3-030-32692-0_36","url":null,"abstract":"","PeriodicalId":93329,"journal":{"name":"Lecture notes-monograph series","volume":"11861 1","pages":"310-318"},"PeriodicalIF":0.0,"publicationDate":"2019-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47255262","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}
Pub Date : 2007-03-01DOI: 10.1142/9789812708298_0025
T. Zheng, S. Lo
Assessing the reproducibility of research studies can be difficult, especially when the number of objects involved is large. In such situations, there is only a small set of those objects that are truly relevant to the scientific questions. For example, in microarray analysis, despite data sets containing expression levels for tens of thousands of genes, it is expected that only a small fraction of these genes are regulated by the treatment in a single experiment. In such cases, it is acknowledged that reproducibility of two studies is high only for objects with real signals. One way to assess reproducibility is to measure the associations between the two sets of data. The traditional association methods suffered from the lack of adequate power to detect the real signals, however. We propose in this article the use of a modified Kendall rank-order test of association, based on truncated ranks. Simulation results show that the proposed procedure increases the capacity to detect the real signals considerably.
{"title":"A Modified Kendall Rank-Order Association Test For Evaluating The Repeatability Of Two Studies With A Large Number Of Objects.","authors":"T. Zheng, S. Lo","doi":"10.1142/9789812708298_0025","DOIUrl":"https://doi.org/10.1142/9789812708298_0025","url":null,"abstract":"Assessing the reproducibility of research studies can be difficult, especially when the number of objects involved is large. In such situations, there is only a small set of those objects that are truly relevant to the scientific questions. For example, in microarray analysis, despite data sets containing expression levels for tens of thousands of genes, it is expected that only a small fraction of these genes are regulated by the treatment in a single experiment. In such cases, it is acknowledged that reproducibility of two studies is high only for objects with real signals. One way to assess reproducibility is to measure the associations between the two sets of data. The traditional association methods suffered from the lack of adequate power to detect the real signals, however. We propose in this article the use of a modified Kendall rank-order test of association, based on truncated ranks. Simulation results show that the proposed procedure increases the capacity to detect the real signals considerably.","PeriodicalId":93329,"journal":{"name":"Lecture notes-monograph series","volume":"3 1","pages":"515-528"},"PeriodicalIF":0.0,"publicationDate":"2007-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1142/9789812708298_0025","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"64022274","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号