Cargando…
A Robust Training Method for Pathological Cellular Detector via Spatial Loss Calibration
Computer-aided diagnosis of pathological images usually requires detecting and examining all positive cells for accurate diagnosis. However, cellular datasets tend to be sparsely annotated due to the challenge of annotating all the cells. However, training detectors on sparse annotations may be misl...
Autores principales: | , , , , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8712578/ https://www.ncbi.nlm.nih.gov/pubmed/34970560 http://dx.doi.org/10.3389/fmed.2021.767625 |
_version_ | 1784623585948598272 |
---|---|
author | Li, Hansheng Kang, Yuxin Yang, Wentao Wu, Zhuoyue Shi, Xiaoshuang Liu, Feihong Liu, Jianye Hu, Lingyu Ma, Qian Cui, Lei Feng, Jun Yang, Lin |
author_facet | Li, Hansheng Kang, Yuxin Yang, Wentao Wu, Zhuoyue Shi, Xiaoshuang Liu, Feihong Liu, Jianye Hu, Lingyu Ma, Qian Cui, Lei Feng, Jun Yang, Lin |
author_sort | Li, Hansheng |
collection | PubMed |
description | Computer-aided diagnosis of pathological images usually requires detecting and examining all positive cells for accurate diagnosis. However, cellular datasets tend to be sparsely annotated due to the challenge of annotating all the cells. However, training detectors on sparse annotations may be misled by miscalculated losses, limiting the detection performance. Thus, efficient and reliable methods for training cellular detectors on sparse annotations are in higher demand than ever. In this study, we propose a training method that utilizes regression boxes' spatial information to conduct loss calibration to reduce the miscalculated loss. Extensive experimental results show that our method can significantly boost detectors' performance trained on datasets with varying degrees of sparse annotations. Even if 90% of the annotations are missing, the performance of our method is barely affected. Furthermore, we find that the middle layers of the detector are closely related to the generalization performance. More generally, this study could elucidate the link between layers and generalization performance, provide enlightenment for future research, such as designing and applying constraint rules to specific layers according to gradient analysis to achieve “scalpel-level” model training. |
format | Online Article Text |
id | pubmed-8712578 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | Frontiers Media S.A. |
record_format | MEDLINE/PubMed |
spelling | pubmed-87125782021-12-29 A Robust Training Method for Pathological Cellular Detector via Spatial Loss Calibration Li, Hansheng Kang, Yuxin Yang, Wentao Wu, Zhuoyue Shi, Xiaoshuang Liu, Feihong Liu, Jianye Hu, Lingyu Ma, Qian Cui, Lei Feng, Jun Yang, Lin Front Med (Lausanne) Medicine Computer-aided diagnosis of pathological images usually requires detecting and examining all positive cells for accurate diagnosis. However, cellular datasets tend to be sparsely annotated due to the challenge of annotating all the cells. However, training detectors on sparse annotations may be misled by miscalculated losses, limiting the detection performance. Thus, efficient and reliable methods for training cellular detectors on sparse annotations are in higher demand than ever. In this study, we propose a training method that utilizes regression boxes' spatial information to conduct loss calibration to reduce the miscalculated loss. Extensive experimental results show that our method can significantly boost detectors' performance trained on datasets with varying degrees of sparse annotations. Even if 90% of the annotations are missing, the performance of our method is barely affected. Furthermore, we find that the middle layers of the detector are closely related to the generalization performance. More generally, this study could elucidate the link between layers and generalization performance, provide enlightenment for future research, such as designing and applying constraint rules to specific layers according to gradient analysis to achieve “scalpel-level” model training. Frontiers Media S.A. 2021-12-14 /pmc/articles/PMC8712578/ /pubmed/34970560 http://dx.doi.org/10.3389/fmed.2021.767625 Text en Copyright © 2021 Li, Kang, Yang, Wu, Shi, Liu, Liu, Hu, Ma, Cui, Feng and Yang. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. |
spellingShingle | Medicine Li, Hansheng Kang, Yuxin Yang, Wentao Wu, Zhuoyue Shi, Xiaoshuang Liu, Feihong Liu, Jianye Hu, Lingyu Ma, Qian Cui, Lei Feng, Jun Yang, Lin A Robust Training Method for Pathological Cellular Detector via Spatial Loss Calibration |
title | A Robust Training Method for Pathological Cellular Detector via Spatial Loss Calibration |
title_full | A Robust Training Method for Pathological Cellular Detector via Spatial Loss Calibration |
title_fullStr | A Robust Training Method for Pathological Cellular Detector via Spatial Loss Calibration |
title_full_unstemmed | A Robust Training Method for Pathological Cellular Detector via Spatial Loss Calibration |
title_short | A Robust Training Method for Pathological Cellular Detector via Spatial Loss Calibration |
title_sort | robust training method for pathological cellular detector via spatial loss calibration |
topic | Medicine |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8712578/ https://www.ncbi.nlm.nih.gov/pubmed/34970560 http://dx.doi.org/10.3389/fmed.2021.767625 |
work_keys_str_mv | AT lihansheng arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT kangyuxin arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT yangwentao arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT wuzhuoyue arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT shixiaoshuang arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT liufeihong arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT liujianye arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT hulingyu arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT maqian arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT cuilei arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT fengjun arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT yanglin arobusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT lihansheng robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT kangyuxin robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT yangwentao robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT wuzhuoyue robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT shixiaoshuang robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT liufeihong robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT liujianye robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT hulingyu robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT maqian robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT cuilei robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT fengjun robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration AT yanglin robusttrainingmethodforpathologicalcellulardetectorviaspatiallosscalibration |