PseudoCell: Hard Negative Mining as Pseudo Labeling for Deep Learning-Based Centroblast Cell Detection
<italic>Background:</italic> Deep learning models for patch classification in whole-slide images (WSIs) have shown promise in assisting follicular lymphoma grading. However, these models often require pathologists to identify centroblasts and manually provide refined labels for model opt...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Journal of Engineering in Medicine and Biology |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10542389/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583205376491520 |
|---|---|
| author | Narongrid Seesawad Piyalitt Ittichaiwong Thapanun Sudhawiyangkul Phattarapong Sawangjai Peti Thuwajit Paisarn Boonsakan Supasan Sripodok Kanyakorn Veerakanjana Komgrid Charngkaew Ananya Pongpaibul Napat Angkathunyakul Narit Hnoohom Sumeth Yuenyong Chanitra Thuwajit Theerawit Wilaiprasitporn |
| author_facet | Narongrid Seesawad Piyalitt Ittichaiwong Thapanun Sudhawiyangkul Phattarapong Sawangjai Peti Thuwajit Paisarn Boonsakan Supasan Sripodok Kanyakorn Veerakanjana Komgrid Charngkaew Ananya Pongpaibul Napat Angkathunyakul Narit Hnoohom Sumeth Yuenyong Chanitra Thuwajit Theerawit Wilaiprasitporn |
| author_sort | Narongrid Seesawad |
| collection | DOAJ |
| description | <italic>Background:</italic> Deep learning models for patch classification in whole-slide images (WSIs) have shown promise in assisting follicular lymphoma grading. However, these models often require pathologists to identify centroblasts and manually provide refined labels for model optimization. <italic>Objective:</italic> To address this limitation, we propose <italic>PseudoCell</italic>, an object detection framework for automated centroblast detection in WSI, eliminating the need for extensive pathologist's refined labels. <italic>Methods:</italic> <italic>PseudoCell</italic> leverages a combination of pathologist-provided centroblast labels and pseudo-negative labels generated from undersampled false-positive predictions based on cell morphology features. This approach reduces the reliance on time-consuming manual annotations. <italic>Results:</italic> Our framework significantly reduces the workload for pathologists by accurately identifying and narrowing down areas of interest containing centroblasts. Depending on the confidence threshold, <italic>PseudoCell</italic> can eliminate 58.18-99.35% of irrelevant tissue areas on WSI, streamlining the diagnostic process. <italic>Conclusion:</italic> This study presents <italic>PseudoCell</italic> as a practical and efficient prescreening method for centroblast detection, eliminating the need for refined labels from pathologists. The discussion section provides detailed guidance for implementing <italic>PseudoCell</italic> in clinical practice. |
| format | Article |
| id | doaj-art-0c94c612c58b45a59a13709fb48791dd |
| institution | Kabale University |
| issn | 2644-1276 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Engineering in Medicine and Biology |
| spelling | doaj-art-0c94c612c58b45a59a13709fb48791dd2025-01-29T00:01:30ZengIEEEIEEE Open Journal of Engineering in Medicine and Biology2644-12762024-01-01551452310.1109/OJEMB.2024.340735110542389PseudoCell: Hard Negative Mining as Pseudo Labeling for Deep Learning-Based Centroblast Cell DetectionNarongrid Seesawad0https://orcid.org/0009-0002-9426-7218Piyalitt Ittichaiwong1https://orcid.org/0000-0002-4950-7764Thapanun Sudhawiyangkul2https://orcid.org/0000-0003-2486-9280Phattarapong Sawangjai3Peti Thuwajit4https://orcid.org/0000-0003-0798-1510Paisarn Boonsakan5https://orcid.org/0000-0002-7199-7106Supasan Sripodok6https://orcid.org/0009-0008-3239-2674Kanyakorn Veerakanjana7Komgrid Charngkaew8Ananya Pongpaibul9https://orcid.org/0000-0002-1950-0385Napat Angkathunyakul10Narit Hnoohom11https://orcid.org/0000-0003-3714-8482Sumeth Yuenyong12Chanitra Thuwajit13https://orcid.org/0000-0001-9506-6405Theerawit Wilaiprasitporn14https://orcid.org/0000-0003-4941-4354Bio-inspired Robotics and Neural Engineering (BRAIN) Lab, School of Information Science and Technology (IST), Vidyasirimedhi Institute of Science & Technology (VISTEC), Rayong, ThailandSiriraj Informatics and Data Innovation Center, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, ThailandBio-inspired Robotics and Neural Engineering (BRAIN) Lab, School of Information Science and Technology (IST), Vidyasirimedhi Institute of Science & Technology (VISTEC), Rayong, ThailandBio-inspired Robotics and Neural Engineering (BRAIN) Lab, School of Information Science and Technology (IST), Vidyasirimedhi Institute of Science & Technology (VISTEC), Rayong, ThailandDepartment of Immunology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, ThailandDepartment of Pathology, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, ThailandDepartment of Pathology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, ThailandSiriraj Informatics and Data Innovation Center, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, ThailandDepartment of Pathology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, ThailandDepartment of Pathology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, ThailandDepartment of Pathology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, ThailandDepartment of Computer Engineering, Faculty of Engineering, Mahidol University, Nakhon Pathom, ThailandDepartment of Computer Engineering, Faculty of Engineering, Mahidol University, Nakhon Pathom, ThailandDepartment of Immunology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, ThailandBio-inspired Robotics and Neural Engineering (BRAIN) Lab, School of Information Science and Technology (IST), Vidyasirimedhi Institute of Science & Technology (VISTEC), Rayong, Thailand<italic>Background:</italic> Deep learning models for patch classification in whole-slide images (WSIs) have shown promise in assisting follicular lymphoma grading. However, these models often require pathologists to identify centroblasts and manually provide refined labels for model optimization. <italic>Objective:</italic> To address this limitation, we propose <italic>PseudoCell</italic>, an object detection framework for automated centroblast detection in WSI, eliminating the need for extensive pathologist's refined labels. <italic>Methods:</italic> <italic>PseudoCell</italic> leverages a combination of pathologist-provided centroblast labels and pseudo-negative labels generated from undersampled false-positive predictions based on cell morphology features. This approach reduces the reliance on time-consuming manual annotations. <italic>Results:</italic> Our framework significantly reduces the workload for pathologists by accurately identifying and narrowing down areas of interest containing centroblasts. Depending on the confidence threshold, <italic>PseudoCell</italic> can eliminate 58.18-99.35% of irrelevant tissue areas on WSI, streamlining the diagnostic process. <italic>Conclusion:</italic> This study presents <italic>PseudoCell</italic> as a practical and efficient prescreening method for centroblast detection, eliminating the need for refined labels from pathologists. The discussion section provides detailed guidance for implementing <italic>PseudoCell</italic> in clinical practice.https://ieeexplore.ieee.org/document/10542389/Centroblast cell detectiondeep convolutional neural networkfollicular lymphomahard negative miningmorphological features |
| spellingShingle | Narongrid Seesawad Piyalitt Ittichaiwong Thapanun Sudhawiyangkul Phattarapong Sawangjai Peti Thuwajit Paisarn Boonsakan Supasan Sripodok Kanyakorn Veerakanjana Komgrid Charngkaew Ananya Pongpaibul Napat Angkathunyakul Narit Hnoohom Sumeth Yuenyong Chanitra Thuwajit Theerawit Wilaiprasitporn PseudoCell: Hard Negative Mining as Pseudo Labeling for Deep Learning-Based Centroblast Cell Detection IEEE Open Journal of Engineering in Medicine and Biology Centroblast cell detection deep convolutional neural network follicular lymphoma hard negative mining morphological features |
| title | PseudoCell: Hard Negative Mining as Pseudo Labeling for Deep Learning-Based Centroblast Cell Detection |
| title_full | PseudoCell: Hard Negative Mining as Pseudo Labeling for Deep Learning-Based Centroblast Cell Detection |
| title_fullStr | PseudoCell: Hard Negative Mining as Pseudo Labeling for Deep Learning-Based Centroblast Cell Detection |
| title_full_unstemmed | PseudoCell: Hard Negative Mining as Pseudo Labeling for Deep Learning-Based Centroblast Cell Detection |
| title_short | PseudoCell: Hard Negative Mining as Pseudo Labeling for Deep Learning-Based Centroblast Cell Detection |
| title_sort | pseudocell hard negative mining as pseudo labeling for deep learning based centroblast cell detection |
| topic | Centroblast cell detection deep convolutional neural network follicular lymphoma hard negative mining morphological features |
| url | https://ieeexplore.ieee.org/document/10542389/ |
| work_keys_str_mv | AT narongridseesawad pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT piyalittittichaiwong pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT thapanunsudhawiyangkul pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT phattarapongsawangjai pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT petithuwajit pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT paisarnboonsakan pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT supasansripodok pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT kanyakornveerakanjana pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT komgridcharngkaew pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT ananyapongpaibul pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT napatangkathunyakul pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT narithnoohom pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT sumethyuenyong pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT chanitrathuwajit pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection AT theerawitwilaiprasitporn pseudocellhardnegativeminingaspseudolabelingfordeeplearningbasedcentroblastcelldetection |