The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
데이터 증대(Data Augmentation) 방법은 딥러닝의 지도 학습을 위해 적은 수의 이미지로 구성된 데이터 세트에서 많은 수의 이미지로 구성된 데이터 세트를 생성하는 데 유용한 기술로 알려져 있습니다. 그러나 최근 인공지능(AI)에 대한 연구에서는 영상인식에 대한 낮은 타당도 증강 방법이 보고됐다. 본 연구에서는 백혈구(WBC) 인식을 위한 딥러닝 모델 생성에서 최적의 데이터 증대 방법을 명확히 하는 것을 목표로 했습니다. 연구 디자인 : 우리는 감독 훈련을 통해 생성된 WBC 인식을 위한 각 AI 모델을 사용하여 원본 WBC 이미지에 대해 세 가지 다른 데이터 확대 방법(회전, 크기 조정 및 왜곡)을 수행했습니다. 임상평가 대상자는 건강한 사람 51명이었다. 말초 혈액으로부터 박층 혈액 도말을 준비하고 May-Grünwald-Giemsa 염색을 실시했습니다. 결과 : WBC 인식을 위한 AI 모델 중 유일하게 유의미하게 효과적인 기술은 회전을 통한 데이터 확대였습니다. 대조적으로, 이미지 왜곡과 이미지 스케일링의 효율성은 모두 낮았으며 향상된 정확도는 특정 WBC 하위 범주로 제한되었습니다. 결론 : 지도 학습을 통해 AI 생성에서 높은 정확도를 달성하기 위해 데이터 증강 방법이 자주 사용되지만, 의료 영상의 특성을 기반으로 의료용 AI 생성을 위한 최적의 데이터 증강 방법을 선택하는 것이 필요하다고 생각합니다.
Hiroyuki NOZAKA
Hirosaki University
Kosuke KAMATA
Hirosaki University
Kazufumi YAMAGATA
Hirosaki University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
부
Hiroyuki NOZAKA, Kosuke KAMATA, Kazufumi YAMAGATA, "The Effectiveness of Data Augmentation for Mature White Blood Cell Image Classification in Deep Learning — Selection of an Optimal Technique for Hematological Morphology Recognition —" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 5, pp. 707-714, May 2023, doi: 10.1587/transinf.2022DLP0066.
Abstract: The data augmentation method is known as a helpful technique to generate a dataset with a large number of images from one with a small number of images for supervised training in deep learning. However, a low validity augmentation method for image recognition was reported in a recent study on artificial intelligence (AI). This study aimed to clarify the optimal data augmentation method in deep learning model generation for the recognition of white blood cells (WBCs). Study Design: We conducted three different data augmentation methods (rotation, scaling, and distortion) on original WBC images, with each AI model for WBC recognition generated by supervised training. The subjects of the clinical assessment were 51 healthy persons. Thin-layer blood smears were prepared from peripheral blood and subjected to May-Grünwald-Giemsa staining. Results: The only significantly effective technique among the AI models for WBC recognition was data augmentation with rotation. By contrast, the effectiveness of both image distortion and image scaling was poor, and improved accuracy was limited to a specific WBC subcategory. Conclusion: Although data augmentation methods are often used for achieving high accuracy in AI generation with supervised training, we consider that it is necessary to select the optimal data augmentation method for medical AI generation based on the characteristics of medical images.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022DLP0066/_p
부
@ARTICLE{e106-d_5_707,
author={Hiroyuki NOZAKA, Kosuke KAMATA, Kazufumi YAMAGATA, },
journal={IEICE TRANSACTIONS on Information},
title={The Effectiveness of Data Augmentation for Mature White Blood Cell Image Classification in Deep Learning — Selection of an Optimal Technique for Hematological Morphology Recognition —},
year={2023},
volume={E106-D},
number={5},
pages={707-714},
abstract={The data augmentation method is known as a helpful technique to generate a dataset with a large number of images from one with a small number of images for supervised training in deep learning. However, a low validity augmentation method for image recognition was reported in a recent study on artificial intelligence (AI). This study aimed to clarify the optimal data augmentation method in deep learning model generation for the recognition of white blood cells (WBCs). Study Design: We conducted three different data augmentation methods (rotation, scaling, and distortion) on original WBC images, with each AI model for WBC recognition generated by supervised training. The subjects of the clinical assessment were 51 healthy persons. Thin-layer blood smears were prepared from peripheral blood and subjected to May-Grünwald-Giemsa staining. Results: The only significantly effective technique among the AI models for WBC recognition was data augmentation with rotation. By contrast, the effectiveness of both image distortion and image scaling was poor, and improved accuracy was limited to a specific WBC subcategory. Conclusion: Although data augmentation methods are often used for achieving high accuracy in AI generation with supervised training, we consider that it is necessary to select the optimal data augmentation method for medical AI generation based on the characteristics of medical images.},
keywords={},
doi={10.1587/transinf.2022DLP0066},
ISSN={1745-1361},
month={May},}
부
TY - JOUR
TI - The Effectiveness of Data Augmentation for Mature White Blood Cell Image Classification in Deep Learning — Selection of an Optimal Technique for Hematological Morphology Recognition —
T2 - IEICE TRANSACTIONS on Information
SP - 707
EP - 714
AU - Hiroyuki NOZAKA
AU - Kosuke KAMATA
AU - Kazufumi YAMAGATA
PY - 2023
DO - 10.1587/transinf.2022DLP0066
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2023
AB - The data augmentation method is known as a helpful technique to generate a dataset with a large number of images from one with a small number of images for supervised training in deep learning. However, a low validity augmentation method for image recognition was reported in a recent study on artificial intelligence (AI). This study aimed to clarify the optimal data augmentation method in deep learning model generation for the recognition of white blood cells (WBCs). Study Design: We conducted three different data augmentation methods (rotation, scaling, and distortion) on original WBC images, with each AI model for WBC recognition generated by supervised training. The subjects of the clinical assessment were 51 healthy persons. Thin-layer blood smears were prepared from peripheral blood and subjected to May-Grünwald-Giemsa staining. Results: The only significantly effective technique among the AI models for WBC recognition was data augmentation with rotation. By contrast, the effectiveness of both image distortion and image scaling was poor, and improved accuracy was limited to a specific WBC subcategory. Conclusion: Although data augmentation methods are often used for achieving high accuracy in AI generation with supervised training, we consider that it is necessary to select the optimal data augmentation method for medical AI generation based on the characteristics of medical images.
ER -