File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Adaptive Region-Specific Loss for Improved Medical Image Segmentation

TitleAdaptive Region-Specific Loss for Improved Medical Image Segmentation
Authors
KeywordsBiomedical imaging
Deep learning
Deep learning
Image segmentation
loss function
medical image
neural network
Oncology
Optimization
segmentation
Task analysis
Training
Issue Date1-Jan-2023
PublisherInstitute of Electrical and Electronics Engineers
Citation
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023 How to Cite?
Abstract

Defining the loss function is an important part of neural network design and critically determines the success of deep learning modeling. A significant shortcoming of the conventional loss functions is that they weight all regions in the input image volume equally, despite the fact that the system is known to be heterogeneous (i.e., some regions can achieve high prediction performance more easily than others). Here, we introduce a region-specific loss to lift the implicit assumption of homogeneous weighting for better learning. We divide the entire volume into multiple sub-regions, each with an individualized loss constructed for optimal local performance. Effectively, this scheme imposes higher weightings on the sub-regions that are more difficult to segment, and vice versa. Furthermore, the regional false positive and false negative errors are computed for each input image during a training step and the regional penalty is adjusted accordingly to enhance the overall accuracy of the prediction. Using different public and in-house medical image datasets, we demonstrate that the proposed regionally adaptive loss paradigm outperforms conventional methods in the multi-organ segmentations, without any modification to the neural network architecture or additional data preparation.


Persistent Identifierhttp://hdl.handle.net/10722/331455
ISSN
2023 Impact Factor: 20.8
2023 SCImago Journal Rankings: 6.158
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorChen, Y-
dc.contributor.authorYu, L-
dc.contributor.authorWang, J-
dc.contributor.authorPanjwani, N-
dc.contributor.authorObeid, J-
dc.contributor.authorLiu, W-
dc.contributor.authorLiu, L-
dc.contributor.authorKovalchuk, N-
dc.contributor.authorGensheimer, MF-
dc.contributor.authorVitzthum, LK-
dc.contributor.authorBeadle, BM-
dc.contributor.authorChang, DT-
dc.contributor.authorLe, Q-
dc.contributor.authorHan, B-
dc.contributor.authorXing, L-
dc.date.accessioned2023-09-21T06:55:54Z-
dc.date.available2023-09-21T06:55:54Z-
dc.date.issued2023-01-01-
dc.identifier.citationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2023-
dc.identifier.issn0162-8828-
dc.identifier.urihttp://hdl.handle.net/10722/331455-
dc.description.abstract<p>Defining the loss function is an important part of neural network design and critically determines the success of deep learning modeling. A significant shortcoming of the conventional loss functions is that they weight all regions in the input image volume equally, despite the fact that the system is known to be heterogeneous (i.e., some regions can achieve high prediction performance more easily than others). Here, we introduce a region-specific loss to lift the implicit assumption of homogeneous weighting for better learning. We divide the entire volume into multiple sub-regions, each with an individualized loss constructed for optimal local performance. Effectively, this scheme imposes higher weightings on the sub-regions that are more difficult to segment, and vice versa. Furthermore, the regional false positive and false negative errors are computed for each input image during a training step and the regional penalty is adjusted accordingly to enhance the overall accuracy of the prediction. Using different public and in-house medical image datasets, we demonstrate that the proposed regionally adaptive loss paradigm outperforms conventional methods in the multi-organ segmentations, without any modification to the neural network architecture or additional data preparation.<br></p>-
dc.languageeng-
dc.publisherInstitute of Electrical and Electronics Engineers-
dc.relation.ispartofIEEE Transactions on Pattern Analysis and Machine Intelligence-
dc.rightsThis work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.-
dc.subjectBiomedical imaging-
dc.subjectDeep learning-
dc.subjectDeep learning-
dc.subjectImage segmentation-
dc.subjectloss function-
dc.subjectmedical image-
dc.subjectneural network-
dc.subjectOncology-
dc.subjectOptimization-
dc.subjectsegmentation-
dc.subjectTask analysis-
dc.subjectTraining-
dc.titleAdaptive Region-Specific Loss for Improved Medical Image Segmentation-
dc.typeArticle-
dc.identifier.doi10.1109/TPAMI.2023.3289667-
dc.identifier.scopuseid_2-s2.0-85163532381-
dc.identifier.eissn1939-3539-
dc.identifier.isiWOS:001085050900036-
dc.identifier.issnl0162-8828-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats