计算机科学与探索 ›› 2022, Vol. 16 ›› Issue (2): 413-427.DOI: 10.3778/j.issn.1673-9418.2008028
李广丽1, 袁天1, 李传秀1, 邬任重2, 卓建武1, 张红斌2,+()
收稿日期:
2020-08-10
修回日期:
2020-10-21
出版日期:
2022-02-01
发布日期:
2020-11-06
通讯作者:
+ E-mail: zhanghongbin@whu.edu.cn作者简介:
李广丽(1977—),女,广西博白人,硕士,副教授,CCF会员,主要研究方向为医学图像分析、跨媒体检索、推荐系统。基金资助:
LI Guangli1, YUAN Tian1, LI Chuanxiu1, WU Renzhong2, ZHUO Jianwu1, ZHANG Hongbin2,+()
Received:
2020-08-10
Revised:
2020-10-21
Online:
2022-02-01
Published:
2020-11-06
About author:
LI Guangli, born in 1977, M.S., associate professor, member of CCF. Her research interests include medical image analysis, cross-media retrieval and recommendation system.Supported by:
摘要:
乳腺癌是女性中最常见的癌症,乳腺肿块识别模型能有效地辅助医生的临床诊断工作。然而,医学图像样本稀缺使识别模型易过拟合。提出融入深层病理信息挖掘的乳腺肿块识别模型:构建样本精选策略,跨越不同乳腺造影图像数据集筛选优质样本,从数据增强角度应对医学图像样本稀缺;由浅入深挖掘有限标注样本中蕴含的病理信息,从特征优选角度应对医学图像样本稀缺。设计多视角有效区域基因优选(MvERGS)算法,以精化原始图像特征,提升特征判别性并压缩特征维度,更好地匹配样本数量;对精化的新特征执行判别相关分析(DCA),深入挖掘异构特征间的跨模态相关性,即深层病理信息,以准确刻画乳腺肿块病灶区域。基于深层病理信息与传统分类器训练出高效的乳腺肿块识别模型,完成乳腺造影图像分类。实验表明:识别模型的关键技术指标,包括Accuracy和AUC,均优于主流基线,样本稀缺导致的过拟合问题得到缓解。
中图分类号:
李广丽, 袁天, 李传秀, 邬任重, 卓建武, 张红斌. 融入深层病理信息挖掘的乳腺肿块识别模型[J]. 计算机科学与探索, 2022, 16(2): 413-427.
LI Guangli, YUAN Tian, LI Chuanxiu, WU Renzhong, ZHUO Jianwu, ZHANG Hongbin. Breast Mass Recognition Model via Deep-Level Pathological Information Mining[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(2): 413-427.
数据集 | 预处理后尺寸/像素 | 阴性样本数/幅 | 阳性样本数/幅 | 训练-测试比 |
---|---|---|---|---|
CBIS-DDSM | 1 152×896 | 1 434 | 1 347 | 7:3 |
INbreast | 2 500×3 300 | 287 | 100 |
表1 CBIS-DDSM和INbreast数据集的相关信息
Table 1 Detailed information about CBIS-DDSM and INbreast datasets
数据集 | 预处理后尺寸/像素 | 阴性样本数/幅 | 阳性样本数/幅 | 训练-测试比 |
---|---|---|---|---|
CBIS-DDSM | 1 152×896 | 1 434 | 1 347 | 7:3 |
INbreast | 2 500×3 300 | 287 | 100 |
设置 | 模型 | Acc/% | AUC/% | 设置 | 模型 | Acc/% | AUC/% |
---|---|---|---|---|---|---|---|
训练-测试:85-15 图像:Whole | RESNET-RESNET[ | — | 87.00 | 训练-测试:70-30 图像:Whole | GS-XGBoost[ | 78.71 | 85.94 |
RESNET-VGG[ | — | 88.00 | VGG16[ | 50.46 | 51.21 | ||
VGG-VGG[ | — | 86.00 | ResNet152[ | 50.46 | 53.22 | ||
VGG-RESNET[ | — | 88.00 | DenseNet121[ | 50.69 | 54.69 | ||
Model Averaging[ | — | 91.00 | Fisher Score[ | 80.50 | 90.65 | ||
GS-XGBoost[ | 76.99 | 72.81 | ERGS[ | 79.31 | 93.73 | ||
M-SVM | 77.51 | 85.17 | PSO[ | 74.40 | 83.36 | ||
M-LR | 77.03 | 86.36 | HGSCCA[ | 53.23 | 50.00 | ||
MD-XGBoost | 77.75 | 77.98 | M-RF | 83.49 | 97.98 | ||
MD-LR | 77.27 | 85.16 | M-SVM | 81.34 | 98.60 | ||
RMD-XGBoost | 77.65 | 77.59 | MD-GBDT | 91.39 | 96.03 | ||
RMD-NB | 77.19 | 84.84 | MD-NB | 81.10 | 98.24 | ||
图像:ROI | Tsochatzidis[ | 74.90 | 80.40 | RMD-NB | 82.66 | 93.56 | |
Rampun[ | 80.40 | 84.00 | RMD-RF | 67.58 | 94.21 |
表2 在CBIS-DDSM数据集上RMD与基线的对比
Table 2 Comparison of RMD and baselines on CBIS-DDSM dataset
设置 | 模型 | Acc/% | AUC/% | 设置 | 模型 | Acc/% | AUC/% |
---|---|---|---|---|---|---|---|
训练-测试:85-15 图像:Whole | RESNET-RESNET[ | — | 87.00 | 训练-测试:70-30 图像:Whole | GS-XGBoost[ | 78.71 | 85.94 |
RESNET-VGG[ | — | 88.00 | VGG16[ | 50.46 | 51.21 | ||
VGG-VGG[ | — | 86.00 | ResNet152[ | 50.46 | 53.22 | ||
VGG-RESNET[ | — | 88.00 | DenseNet121[ | 50.69 | 54.69 | ||
Model Averaging[ | — | 91.00 | Fisher Score[ | 80.50 | 90.65 | ||
GS-XGBoost[ | 76.99 | 72.81 | ERGS[ | 79.31 | 93.73 | ||
M-SVM | 77.51 | 85.17 | PSO[ | 74.40 | 83.36 | ||
M-LR | 77.03 | 86.36 | HGSCCA[ | 53.23 | 50.00 | ||
MD-XGBoost | 77.75 | 77.98 | M-RF | 83.49 | 97.98 | ||
MD-LR | 77.27 | 85.16 | M-SVM | 81.34 | 98.60 | ||
RMD-XGBoost | 77.65 | 77.59 | MD-GBDT | 91.39 | 96.03 | ||
RMD-NB | 77.19 | 84.84 | MD-NB | 81.10 | 98.24 | ||
图像:ROI | Tsochatzidis[ | 74.90 | 80.40 | RMD-NB | 82.66 | 93.56 | |
Rampun[ | 80.40 | 84.00 | RMD-RF | 67.58 | 94.21 |
设置 | 模型 | Acc/% | AUC/% | 设置 | 模型 | Acc/% | AUC/% |
---|---|---|---|---|---|---|---|
训练-测试:70-30 图像:Whole | RESNET-RESNET[ | — | 95.00 | 训练-测试:70-30 图像:Whole | PSO[ | 79.31 | 80.43 |
RESNET-VGG[ | — | 95.00 | HGSCCA[ | 78.56 | 50.00 | ||
VGG-VGG[ | — | 95.00 | M-GBDT | 85.34 | 83.08 | ||
VGG-RESNET[ | — | 95.00 | M-LR | 83.62 | 86.24 | ||
Model Averaging[ | — | 98.00 | MD-LR | 89.66 | 91.48 | ||
GS-XGBoost[ | 85.35 | 82.84 | MD-GBDT | 87.07 | 93.00 | ||
VGG16[ | 75.86 | 52.52 | RMD-NB | 77.59 | 81.94 | ||
ResNet152[ | 75.86 | 54.14 | RMD-SVM | 75.86 | 82.79 | ||
DenseNet121[ | 75.86 | 62.01 | 图像:ROI | CNN[ | 84.00±4 | 69.00±10 | |
Fisher Score[ | 81.03 | 88.51 | SMIL[ | 90.00±2 | 89.00±3 | ||
ERGS[ | 81.89 | 78.41 | Carneiro[ | — | 86.00 |
表3 在INbreast数据集上RMD与基线的对比
Table 3 Comparison of RMD and baselines on INbreast dataset
设置 | 模型 | Acc/% | AUC/% | 设置 | 模型 | Acc/% | AUC/% |
---|---|---|---|---|---|---|---|
训练-测试:70-30 图像:Whole | RESNET-RESNET[ | — | 95.00 | 训练-测试:70-30 图像:Whole | PSO[ | 79.31 | 80.43 |
RESNET-VGG[ | — | 95.00 | HGSCCA[ | 78.56 | 50.00 | ||
VGG-VGG[ | — | 95.00 | M-GBDT | 85.34 | 83.08 | ||
VGG-RESNET[ | — | 95.00 | M-LR | 83.62 | 86.24 | ||
Model Averaging[ | — | 98.00 | MD-LR | 89.66 | 91.48 | ||
GS-XGBoost[ | 85.35 | 82.84 | MD-GBDT | 87.07 | 93.00 | ||
VGG16[ | 75.86 | 52.52 | RMD-NB | 77.59 | 81.94 | ||
ResNet152[ | 75.86 | 54.14 | RMD-SVM | 75.86 | 82.79 | ||
DenseNet121[ | 75.86 | 62.01 | 图像:ROI | CNN[ | 84.00±4 | 69.00±10 | |
Fisher Score[ | 81.03 | 88.51 | SMIL[ | 90.00±2 | 89.00±3 | ||
ERGS[ | 81.89 | 78.41 | Carneiro[ | — | 86.00 |
Dataset | Feature | Acc/% | PreNeg/% | PrePos/% | Sen/% | Spe/% | AUC/% | TP | TN | FP | FN |
---|---|---|---|---|---|---|---|---|---|---|---|
CBIS-DDSM | S | 81.34 | 73.91 | 97.70 | 62.96 | 98.61 | 95.24 | 255 | 425 | 6 | 150 |
G | 67.34 | 65.80 | 69.64 | 57.78 | 76.33 | 67.06 | 234 | 329 | 102 | 171 | |
H | 63.64 | 63.60 | 63.69 | 58.02 | 68.91 | 68.05 | 235 | 297 | 134 | 170 | |
L | 55.50 | 57.58 | 53.69 | 59.26 | 51.97 | 58.76 | 240 | 224 | 207 | 165 | |
D | 52.27 | 53.21 | 50.89 | 42.47 | 61.48 | 51.98 | 172 | 265 | 166 | 233 | |
R | 58.97 | 59.52 | 58.29 | 53.83 | 63.81 | 62.32 | 218 | 275 | 156 | 187 | |
V | 54.31 | 54.97 | 53.35 | 45.19 | 62.88 | 55.70 | 183 | 271 | 160 | 222 | |
INbreast | S | 77.59 | 77.68 | 75.00 | 10.71 | 98.86 | 54.79 | 3 | 87 | 1 | 25 |
G | 75.86 | 75.86 | 0 | 0 | 100.00 | 58.81 | 0 | 88 | 0 | 28 | |
H | 75.86 | 76.32 | 50.00 | 3.57 | 98.86 | 48.25 | 1 | 87 | 1 | 27 | |
L | 75.86 | 75.86 | 0 | 0 | 100.00 | 50.00 | 0 | 88 | 0 | 28 | |
D | 78.45 | 78.38 | 80.00 | 14.29 | 98.86 | 61.63 | 4 | 87 | 1 | 24 | |
R | 75.86 | 75.86 | 0 | 0 | 100.00 | 53.08 | 0 | 88 | 0 | 28 | |
V | 78.45 | 77.88 | 100.00 | 10.71 | 100.00 | 55.36 | 3 | 88 | 0 | 25 |
表4 原图像特征的识别性能
Table 4 Recognition performance of original image features
Dataset | Feature | Acc/% | PreNeg/% | PrePos/% | Sen/% | Spe/% | AUC/% | TP | TN | FP | FN |
---|---|---|---|---|---|---|---|---|---|---|---|
CBIS-DDSM | S | 81.34 | 73.91 | 97.70 | 62.96 | 98.61 | 95.24 | 255 | 425 | 6 | 150 |
G | 67.34 | 65.80 | 69.64 | 57.78 | 76.33 | 67.06 | 234 | 329 | 102 | 171 | |
H | 63.64 | 63.60 | 63.69 | 58.02 | 68.91 | 68.05 | 235 | 297 | 134 | 170 | |
L | 55.50 | 57.58 | 53.69 | 59.26 | 51.97 | 58.76 | 240 | 224 | 207 | 165 | |
D | 52.27 | 53.21 | 50.89 | 42.47 | 61.48 | 51.98 | 172 | 265 | 166 | 233 | |
R | 58.97 | 59.52 | 58.29 | 53.83 | 63.81 | 62.32 | 218 | 275 | 156 | 187 | |
V | 54.31 | 54.97 | 53.35 | 45.19 | 62.88 | 55.70 | 183 | 271 | 160 | 222 | |
INbreast | S | 77.59 | 77.68 | 75.00 | 10.71 | 98.86 | 54.79 | 3 | 87 | 1 | 25 |
G | 75.86 | 75.86 | 0 | 0 | 100.00 | 58.81 | 0 | 88 | 0 | 28 | |
H | 75.86 | 76.32 | 50.00 | 3.57 | 98.86 | 48.25 | 1 | 87 | 1 | 27 | |
L | 75.86 | 75.86 | 0 | 0 | 100.00 | 50.00 | 0 | 88 | 0 | 28 | |
D | 78.45 | 78.38 | 80.00 | 14.29 | 98.86 | 61.63 | 4 | 87 | 1 | 24 | |
R | 75.86 | 75.86 | 0 | 0 | 100.00 | 53.08 | 0 | 88 | 0 | 28 | |
V | 78.45 | 77.88 | 100.00 | 10.71 | 100.00 | 55.36 | 3 | 88 | 0 | 25 |
Dataset | Feature | Acc/% | PreNeg/% | PrePos/% | Sen/% | Spe/% | AUC/% | TP | TN | FP | FN |
---|---|---|---|---|---|---|---|---|---|---|---|
CBIS- DDSM | | 82.89↑ | 75.44↑ | 98.52↑ | 65.68↑ | 99.07↑ | 97.71↑ | 266↑ | 427↑ | 4↑ | 139↑ |
| 64.59 | 64.15 | 65.18 | 57.78 | 71.00 | 70.78↑ | 234 | 306 | 125 | 171 | |
| 63.76↑ | 63.62↑ | 63.93↑ | 57.78 | 69.37↑ | 63.58 | 234 | 299↑ | 132↑ | 171 | |
| 66.03↑ | 67.13↑ | 64.86↑ | 65.19↑ | 66.82↑ | 72.43↑ | 264↑ | 288↑ | 143↑ | 141↑ | |
| 55.38↑ | 55.56↑ | 55.10↑ | 42.72↑ | 67.29↑ | 55.00↑ | 173↑ | 290↑ | 141↑ | 232↑ | |
| 61.00↑ | 63.29↑ | 58.96↑ | 64.20↑ | 58.00 | 62.86↑ | 260↑ | 250 | 181 | 145↑ | |
| 57.30↑ | 57.09↑ | 57.64↑ | 44.69 | 69.14↑ | 57.88↑ | 181 | 298↑ | 133↑ | 224 | |
Avg1 | 61.91 | 61.23 | 63.89 | 54.22 | 69.14 | 65.59 | 220 | 298 | 133 | 185 | |
Avg2 | 64.42↑ | 63.75↑ | 66.31↑ | 56.86↑ | 71.53↑ | 68.60↑ | 230↑ | 308↑ | 123↑ | 175↑ | |
INbreast | | 81.03↑ | 83.00↑ | 68.75 | 39.29↑ | 94.32 | 61.16↑ | 11↑ | 83 | 5 | 17↑ |
| 76.72↑ | 81.44↑ | 52.63↑ | 35.71↑ | 89.77 | 65.50↑ | 10↑ | 79 | 9 | 18↑ | |
| 76.72↑ | 76.99↑ | 66.67↑ | 7.14↑ | 98.86 | 53.00↑ | 2↑ | 87 | 1 | 26↑ | |
| 77.59↑ | 78.70↑ | 62.50↑ | 17.86↑ | 96.59 | 57.22↑ | 5↑ | 85 | 3 | 23↑ | |
| 76.72 | 77.98 | 57.14 | 14.29↑ | 96.59 | 55.44 | 4 | 85 | 3 | 24 | |
| 80.17↑ | 80.95↑ | 72.73↑ | 28.57↑ | 96.59 | 75.57↑ | 8↑ | 85 | 3 | 20↑ | |
| 77.59 | 78.18↑ | 66.67 | 14.29↑ | 97.73 | 64.98↑ | 4↑ | 86 | 2 | 24↑ | |
Avg3 | 76.85 | 76.83 | 43.57 | 5.61 | 99.51 | 54.79 | 2 | 88 | 1 | 26 | |
Avg4 | 78.08↑ | 79.61↑ | 63.87↑ | 22.45↑ | 95.78 | 61.84↑ | 6↑ | 84 | 4 | 22↑ |
表5 基于MvERGS算法的识别性能
Table 5 Recognition performance based on MvERGS algorithm
Dataset | Feature | Acc/% | PreNeg/% | PrePos/% | Sen/% | Spe/% | AUC/% | TP | TN | FP | FN |
---|---|---|---|---|---|---|---|---|---|---|---|
CBIS- DDSM | | 82.89↑ | 75.44↑ | 98.52↑ | 65.68↑ | 99.07↑ | 97.71↑ | 266↑ | 427↑ | 4↑ | 139↑ |
| 64.59 | 64.15 | 65.18 | 57.78 | 71.00 | 70.78↑ | 234 | 306 | 125 | 171 | |
| 63.76↑ | 63.62↑ | 63.93↑ | 57.78 | 69.37↑ | 63.58 | 234 | 299↑ | 132↑ | 171 | |
| 66.03↑ | 67.13↑ | 64.86↑ | 65.19↑ | 66.82↑ | 72.43↑ | 264↑ | 288↑ | 143↑ | 141↑ | |
| 55.38↑ | 55.56↑ | 55.10↑ | 42.72↑ | 67.29↑ | 55.00↑ | 173↑ | 290↑ | 141↑ | 232↑ | |
| 61.00↑ | 63.29↑ | 58.96↑ | 64.20↑ | 58.00 | 62.86↑ | 260↑ | 250 | 181 | 145↑ | |
| 57.30↑ | 57.09↑ | 57.64↑ | 44.69 | 69.14↑ | 57.88↑ | 181 | 298↑ | 133↑ | 224 | |
Avg1 | 61.91 | 61.23 | 63.89 | 54.22 | 69.14 | 65.59 | 220 | 298 | 133 | 185 | |
Avg2 | 64.42↑ | 63.75↑ | 66.31↑ | 56.86↑ | 71.53↑ | 68.60↑ | 230↑ | 308↑ | 123↑ | 175↑ | |
INbreast | | 81.03↑ | 83.00↑ | 68.75 | 39.29↑ | 94.32 | 61.16↑ | 11↑ | 83 | 5 | 17↑ |
| 76.72↑ | 81.44↑ | 52.63↑ | 35.71↑ | 89.77 | 65.50↑ | 10↑ | 79 | 9 | 18↑ | |
| 76.72↑ | 76.99↑ | 66.67↑ | 7.14↑ | 98.86 | 53.00↑ | 2↑ | 87 | 1 | 26↑ | |
| 77.59↑ | 78.70↑ | 62.50↑ | 17.86↑ | 96.59 | 57.22↑ | 5↑ | 85 | 3 | 23↑ | |
| 76.72 | 77.98 | 57.14 | 14.29↑ | 96.59 | 55.44 | 4 | 85 | 3 | 24 | |
| 80.17↑ | 80.95↑ | 72.73↑ | 28.57↑ | 96.59 | 75.57↑ | 8↑ | 85 | 3 | 20↑ | |
| 77.59 | 78.18↑ | 66.67 | 14.29↑ | 97.73 | 64.98↑ | 4↑ | 86 | 2 | 24↑ | |
Avg3 | 76.85 | 76.83 | 43.57 | 5.61 | 99.51 | 54.79 | 2 | 88 | 1 | 26 | |
Avg4 | 78.08↑ | 79.61↑ | 63.87↑ | 22.45↑ | 95.78 | 61.84↑ | 6↑ | 84 | 4 | 22↑ |
[1] | PASHOUTAN S, SHOKOUHI S B, PASHOUTAN M. Automatic breast tumor classification using a level set method and feature extraction in mammography[C]//Proceedings of the 2017 24th National and 2nd International Iranian Conference on Biomedical Engineering, Tehran, Nov 30-Dec 1,2017. Piscataway: IEEE, 2017: 1-6. |
[2] | WU E, WU K, COX D D, et al. Conditional infilling GANs for data augmentation in mammogram classification[C]//LNCS 11040: Proceedings of the 3rd International Workshop on Image Analysis for Moving Organ, Breast, and Thoracic Images, Granada, Sep 16-20, 2018. Cham: Springer, 2018: 98-106. |
[3] | RADFORD A, METZ L, CHINTALA S. Unsupervised representation learning with deep convolutional generative adversarial networks[J]. arXiv:1511.06434, 2015. |
[4] | LI H Y, CHEN D D, NAILON W H, et al. A deep dual-path network for improved mammogram image processing[C]//Proceedings of the 2019 International Conference on Acoustics, Speech and Signal Processing, Brighton, May 12-17, 2019. Piscataway: IEEE, 2019: 1224-1228. |
[5] |
RIBLI D, HORVÁTH A, UNGER Z, et al. Detecting and classifying lesions in mammograms with deep learning[J]. Scientific Reports, 2018, 8(1):4165.
DOI URL |
[6] |
HAGHIGHAT M, ABDEL-MOTTALEB M, ALHALABI W. Discriminant correlation analysis: real-time feature level fusion for multimodal biometric recognition[J]. IEEE Transactions on Information Forensics and Security, 2016, 11(9):1984-1996.
DOI URL |
[7] |
LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2):91-110.
DOI URL |
[8] |
O’ROURKE S M, HERSKOWITZ I, O’SHEA E K. Yeast go the whole HOG for the hyperosmotic response[J]. Trends in Genetics, 2002, 18(8):405-412.
DOI URL |
[9] |
LI Y, CHEN H, WEI X, et al. Mass classification in mam- mograms based on two-concentric masks and discriminating texton[J]. Pattern Recognition, 2016, 60:648-656.
DOI URL |
[10] |
LIU X, TANG J. Mass classification in mammograms using selected geometry and texture features, and a new SVM-based feature selection method[J]. IEEE Systems Journal, 2013, 8(3):910-920.
DOI URL |
[11] |
GUO Z, ZHANG L, ZHANG D. A completed modeling of local binary pattern operator for texture classification[J]. IEEE Transactions on Image Processing, 2010, 19(6):1657-1663.
DOI URL |
[12] | HARALICK R M, SHANMUGAM K. Textural features for image classification[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1973, 6:610-621. |
[13] | JI H K, GUO K W, YANG L Z, et al. Research on feature selection and classification algorithm of medical optical tomography images[C]//Proceedings of the 2019 IEEE/CIC International Conference on Communications in China,Changchun, Aug 11-13, 2019. Piscataway: IEEE, 2019: 561-566. |
[14] | VEERAMUTHU A, MEENAKSHI S, KAMESHWRAN A. A plug-in feature extraction and feature subset selection algorithm for classification of medicinal brain image data[C]//Proceedings of the 2014 International Conference on Communication and Signal Processing, Melmaruvathur, Apr 3-5, 2014. Piscataway: IEEE, 2014: 1545-1551. |
[15] |
KUMAR S U, INBARANI H H. PSO-based feature selection and neighborhood rough set-based classification for BCI multiclass motor imagery task[J]. Neural Computing and Applications, 2017, 28(11):3239-3258.
DOI URL |
[16] |
SUDHA M N, SELVARAJAN S, SUGANTHI M. Feature selection using improved lion optimisation algorithm for breast cancer classification[J]. International Journal of Bio-Inspired Computation, 2019, 14(4):237-246.
DOI URL |
[17] |
ZHU X, SUK H I, LEE S W, et al. Subspace regularized sparse multitask learning for multiclass neurodegenerative disease identification[J]. IEEE Transactions on Biomedical Engineering, 2015, 63(3):607-618.
DOI URL |
[18] |
ZHANG D, SHEN D. Multi-modal multi-task learning for joint prediction of multiple regression and classification variables in Alzheimer’s disease[J]. NeuroImage, 2012, 59(2):895-907.
DOI URL |
[19] |
ZHOU T, LIU M, THUNG K H, et al. Latent representation learning for Alzheimer’s disease diagnosis with incomplete multi-modality neuroimaging and genetic data[J]. IEEE Transactions on Medical Imaging, 2019, 38(10):2411-2422.
DOI URL |
[20] | ZHENG X, SHI J, LI Y, et al. Multi-modality stacked deep polynomial network-based feature learning for Alzheimer’s disease diagnosis[C]//Proceedings of the 13th IEEE International Symposium on Biomedical Imaging, Prague, Apr 13-16, 2016. Piscataway: IEEE, 2016: 851-854. |
[21] |
SHAO W, XIANG S N, ZHANG Z Y, et al. Hyper-graph based sparse canonical correlation analysis for the diagnosis of Alzheimer’s disease from multi-dimensional genomic data[J]. Methods, 2021, 189:86-94.
DOI URL |
[22] |
LIU M, GAO Y, YAP P T, et al. Multi-hypergraph learning for incomplete multimodality data[J]. IEEE Journal of Biomedical and Health Informatics, 2017, 22(4):1197-1208.
DOI URL |
[23] | AGARWAL R, DIAZ O, LLADÓ X, et al. Automatic mass detection in mammograms using deep convolutional neural networks[J]. Journal of Medical Imaging, 2019, 6(3):031409. |
[24] |
TSOCHATZIDIS L, COSTARIDOU L, PRATIKAKIS I. Deep learning for breast cancer diagnosis from mammograms—a comparative study[J]. Journal of Imaging, 2019, 5(3):37.
DOI URL |
[25] |
RAGAB D A, SHARKAS M, MARSHALL S, et al. Breast cancer detection using deep convolutional neural networks and support vector machines[J]. PeerJ, 2019, 7:e6201.
DOI URL |
[26] | DHUNGEL N, CARNEIRO G, BRADLEY A P. Fully automated classification of mammograms using deep residual neural networks[C]//Proceedings of the 14th IEEE International Symposium on Biomedical Imaging, Melbourne, Apr 18-21, 2017. Piscataway: IEEE, 2017: 310-314. |
[27] | RAMPUN A, SCOTNEY B W, MORROW P J, et al. Breast mass classification in mammograms using ensemble convolutional neural networks[C]//Proceedings of the 20th IEEE International Conference on e-Health Networking, Applications and Services, Ostrava, Sep 17-20, 2018. Piscataway: IEEE, 2018: 1-6. |
[28] |
KHAN S U, ISLAM N, JAN Z, et al. A novel deep learning based framework for the detection and classification of breast cancer using transfer learning[J]. Pattern Recognition Letters, 2019, 125:1-6.
DOI URL |
[29] | SHEN L, MARGOLIES L R, ROTHSTEIN J H, et al. Deep learning to improve breast cancer detection on screening mammography[J]. Scientific Reports, 2019, 9(1):1-12. |
[30] |
BYRA M, GALPERIN M, OJEDA-FOURNIER H, et al. Breast mass classification in sonography with transfer learning using a deep convolutional neural network and color conversion[J]. Medical Physics, 2019, 46(2):746-755.
DOI URL |
[31] | ANSAR W, SHAHID A R, RAZA B, et al. Breast cancer detection and localization using MobileNet based transfer learning for mammograms[C]//Proceedings of the 2020 International Symposium on Intelligent Computing Systems, Sharjah, Mar 18-19, 2020. Cham: Springer, 2020: 11-21. |
[32] |
WANG K, PATEL B K, WANG L, et al. A dual-mode deep transfer learning (D2TL) system for breast cancer detection using contrast enhanced digital mammograms[J]. IISE Transactions on Healthcare Systems Engineering, 2019, 9(4):357-370.
DOI URL |
[33] | HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, Jul 21-26, 2017. Washington: IEEE Computer Society, 2017: 2261-2269. |
[34] | HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, Jun 27-30, 2016. Washington: IEEE Computer Society, 2016: 770-778. |
[35] | SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[J]. arXiv:1409. 1556, 2014. |
[36] | 余绍德. 卷积神经网络和迁移学习在癌症影像分析中的研究[D]. 深圳: 中国科学院大学(中国科学院深圳先进技术研究院), 2018. |
YU S D. Research on convolutional neural network and transfer learning in cancer image analysis[D]. Shenzhen: University of Chinese Academy of Sciences (Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences), 2018. | |
[37] |
ZHANG H, QIU D, WU R, et al. Novel framework for image attribute annotation with gene selection XGBoost algorithm and relative attribute model[J]. Applied Soft Computing, 2019, 80:57-79.
DOI URL |
[38] | LIN X H, SONG H H, FAN M, et al. The feature selection algorithm based on feature overlapping and group overlapping[C]//Proceedings of the 2016 IEEE International Conference on Bioinformatics and Biomedicine, Shenzhen, Dec 15-18, 2016. Washington: IEEE Computer Society, 2016: 619-624. |
[39] | LI J D, CHENG K W, WANG S H, et al. Feature selection: a data perspective[J]. ACM Computing Surveys, 2017, 50(6):1-45. |
[40] | ZHU W T, LOU Q, VANG Y S, et al. Deep multi-instance networks with sparse label assignment for whole mammogram classification[C]//LNCS 10435: Proceedings of the 20th International Conference on Medical Image Computing and Computer Assisted Intervention, Quebec City, Sep 11-13, 2017. Cham: Springer, 2017: 603-611. |
[41] | CARNEIRO G, NASCIMENTO J, BRADLEY A P. Deep learning models for classifying mammogram exams containing unregistered multi-view images and segmentation maps of lesions[J]. Deep Learning for Medical Image Analysis, 2017: 321-339. |
[42] | DHUNGEL N, CARNEIRO G, BRADLEY A P. The automated learning of deep features for breast mass classification from mammograms[C]//LNCS 9901: Proceedings of the 19th International Conference on Medical Image Computing and Computer-Assisted Intervention, Athens, Oct 17-21, 2016. Cham: Springer, 2016: 106-114. |
[43] | WANG X L, ROSS G, ABHINAV G, et al. Non-local neural networks[C]//Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, Jun 18-22, 2018. Washington: IEEE Computer Society, 2018: 7794-7803. |
[44] |
SHABAN W M, RABIE A H, SALEH A I, et al. A new COVID-19 patients detection strategy (CPDS) based on hybrid feature selection and enhanced KNN classifier[J]. Knowledge-Based Systems, 2020, 205:106270.
DOI URL |
[1] | 杨知桥, 张莹, 王新杰, 张东波, 王玉. 改进U型网络在视网膜病变检测中的应用研究[J]. 计算机科学与探索, 2022, 16(8): 1877-1884. |
[2] | 何丽, 张红艳, 房婉琳. 融合多尺度边界特征的显著实例分割[J]. 计算机科学与探索, 2022, 16(8): 1865-1876. |
[3] | 黄浩, 葛洪伟. 强化类间区分的深度残差表情识别网络[J]. 计算机科学与探索, 2022, 16(8): 1842-1849. |
[4] | 于慧琳, 陈炜, 王琪, 高建伟, 万怀宇. 使用子图推理实现知识图谱关系预测[J]. 计算机科学与探索, 2022, 16(8): 1800-1808. |
[5] | 潘玉, 陈晓红, 李舜酩, 李纪永. 块增量典型相关分析[J]. 计算机科学与探索, 2022, 16(8): 1809-1818. |
[6] | 叶廷宇, 叶军, 王晖, 王磊. 结合人工蜂群优化的粗糙K-means聚类算法[J]. 计算机科学与探索, 2022, 16(8): 1923-1932. |
[7] | 曾凡智, 许露倩, 周燕, 周月霞, 廖俊玮. 面向智慧教育的知识追踪模型研究综述[J]. 计算机科学与探索, 2022, 16(8): 1742-1763. |
[8] | 洪惠群, 沈贵萍, 黄风华. 表情识别技术综述[J]. 计算机科学与探索, 2022, 16(8): 1764-1778. |
[9] | 刘艺, 李蒙蒙, 郑奇斌, 秦伟, 任小广. 视频目标跟踪算法综述[J]. 计算机科学与探索, 2022, 16(7): 1504-1515. |
[10] | 陈江美, 张文德. 基于位置社交网络的兴趣点推荐系统研究综述[J]. 计算机科学与探索, 2022, 16(7): 1462-1478. |
[11] | 赵小明, 杨轶娇, 张石清. 面向深度学习的多模态情感识别研究进展[J]. 计算机科学与探索, 2022, 16(7): 1479-1503. |
[12] | 夏鸿斌, 肖奕飞, 刘渊. 融合自注意力机制的长文本生成对抗网络模型[J]. 计算机科学与探索, 2022, 16(7): 1603-1610. |
[13] | 王雪纯, 吕晟凯, 吴浩, 何鹏, 曾诚. 多网络混合嵌入学习的服务推荐方法研究[J]. 计算机科学与探索, 2022, 16(7): 1529-1542. |
[14] | 李玉轩, 洪学海, 汪洋, 唐正正, 班艳. 引入激活加权策略的分组排序学习方法[J]. 计算机科学与探索, 2022, 16(7): 1594-1602. |
[15] | 彭豪, 李晓明. 多尺度选择金字塔网络的小样本目标检测算法[J]. 计算机科学与探索, 2022, 16(7): 1649-1660. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||