Lung nodule detection and classification from Thorax CT-scan using RetinaNet with transfer learning
Lung malignancy is one of the most common causes of death in the world caused by malignant lung nodules which commonly diagnosed radiologically by radiologists. Unfortunately, the continuous flow of medical images in hospitals drives radiologists to prioritize quantity over quality. This work condit...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-03-01
|
| Series: | Journal of King Saud University: Computer and Information Sciences |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1319157820303335 |
| _version_ | 1819104145637900288 |
|---|---|
| author | Ivan William Harsono Suryadiputra Liawatimena Tjeng Wawan Cenggoro |
| author_facet | Ivan William Harsono Suryadiputra Liawatimena Tjeng Wawan Cenggoro |
| author_sort | Ivan William Harsono |
| collection | DOAJ |
| description | Lung malignancy is one of the most common causes of death in the world caused by malignant lung nodules which commonly diagnosed radiologically by radiologists. Unfortunately, the continuous flow of medical images in hospitals drives radiologists to prioritize quantity over quality. This work condition allows misinterpretation especially on ambiguous anatomical structures that resemble lung nodule for example enlarged lymph nodes and resulting in decreasing sensitivity and accuracy of malignant lung nodule detections and late diagnosis proven to be fatal to patients. To address the problem, this paper proposed a novel lung nodule detection and classification model using one stage detector called as “I3DR-Net.” The model was formed by combining pre-trained natural images weight of Inflated 3D ConvNet (I3D) backbone with feature pyramid network to multi-scale 3D Thorax Computed tomography scan (CT-scan) dataset. I3DR-Net able to produce remarkable results on lung nodule texture detection task with mAP 49.61% and 22.86%, and area under curve (AUC) 81.84% and 70.36% for public and private dataset. Additionally, I3DR-Net successfully outperform previous state-of-the-art Retina U-Net and U-FRCNN + mean average precision (mAP) by 7.9% and 7.2% (57.71% VS 49.8% VS 50.5%) for malignant nodule detection and classification task. |
| first_indexed | 2024-12-22T02:01:42Z |
| format | Article |
| id | doaj.art-e50da469f0354125b2db7cae393dc5df |
| institution | Directory Open Access Journal |
| issn | 1319-1578 |
| language | English |
| last_indexed | 2024-12-22T02:01:42Z |
| publishDate | 2022-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of King Saud University: Computer and Information Sciences |
| spelling | doaj.art-e50da469f0354125b2db7cae393dc5df2022-12-21T18:42:39ZengElsevierJournal of King Saud University: Computer and Information Sciences1319-15782022-03-01343567577Lung nodule detection and classification from Thorax CT-scan using RetinaNet with transfer learningIvan William Harsono0Suryadiputra Liawatimena1Tjeng Wawan Cenggoro2Computer Science Department, BINUS Graduate Program – Master of Computer Science, Bina Nusantara University, Jakarta 11480, Indonesi; Corresponding author.Computer Science Department, BINUS Graduate Program – Doctor of Computer Science, Bina Nusantara University, Jakarta 11480, Indonesia; Computer Engineering Department, Faculty of Engineering, Bina Nusantara University, Jakarta 11480, IndonesiaComputer Science Department, School of Computer Science, Bina Nusantara University, Jakarta 11480, Indonesia; Bioinformatics and Data Science Research Center, Bina Nusantara University, Jakarta 11480, IndonesiaLung malignancy is one of the most common causes of death in the world caused by malignant lung nodules which commonly diagnosed radiologically by radiologists. Unfortunately, the continuous flow of medical images in hospitals drives radiologists to prioritize quantity over quality. This work condition allows misinterpretation especially on ambiguous anatomical structures that resemble lung nodule for example enlarged lymph nodes and resulting in decreasing sensitivity and accuracy of malignant lung nodule detections and late diagnosis proven to be fatal to patients. To address the problem, this paper proposed a novel lung nodule detection and classification model using one stage detector called as “I3DR-Net.” The model was formed by combining pre-trained natural images weight of Inflated 3D ConvNet (I3D) backbone with feature pyramid network to multi-scale 3D Thorax Computed tomography scan (CT-scan) dataset. I3DR-Net able to produce remarkable results on lung nodule texture detection task with mAP 49.61% and 22.86%, and area under curve (AUC) 81.84% and 70.36% for public and private dataset. Additionally, I3DR-Net successfully outperform previous state-of-the-art Retina U-Net and U-FRCNN + mean average precision (mAP) by 7.9% and 7.2% (57.71% VS 49.8% VS 50.5%) for malignant nodule detection and classification task.http://www.sciencedirect.com/science/article/pii/S1319157820303335Computed tomographyLung noduleObject detectionPattern recognition and classificationTransfer learning |
| spellingShingle | Ivan William Harsono Suryadiputra Liawatimena Tjeng Wawan Cenggoro Lung nodule detection and classification from Thorax CT-scan using RetinaNet with transfer learning Journal of King Saud University: Computer and Information Sciences Computed tomography Lung nodule Object detection Pattern recognition and classification Transfer learning |
| title | Lung nodule detection and classification from Thorax CT-scan using RetinaNet with transfer learning |
| title_full | Lung nodule detection and classification from Thorax CT-scan using RetinaNet with transfer learning |
| title_fullStr | Lung nodule detection and classification from Thorax CT-scan using RetinaNet with transfer learning |
| title_full_unstemmed | Lung nodule detection and classification from Thorax CT-scan using RetinaNet with transfer learning |
| title_short | Lung nodule detection and classification from Thorax CT-scan using RetinaNet with transfer learning |
| title_sort | lung nodule detection and classification from thorax ct scan using retinanet with transfer learning |
| topic | Computed tomography Lung nodule Object detection Pattern recognition and classification Transfer learning |
| url | http://www.sciencedirect.com/science/article/pii/S1319157820303335 |
| work_keys_str_mv | AT ivanwilliamharsono lungnoduledetectionandclassificationfromthoraxctscanusingretinanetwithtransferlearning AT suryadiputraliawatimena lungnoduledetectionandclassificationfromthoraxctscanusingretinanetwithtransferlearning AT tjengwawancenggoro lungnoduledetectionandclassificationfromthoraxctscanusingretinanetwithtransferlearning |