基于先验知识优化的医学图像候选区域生成方法

doi:10.13229/j.cnki.jdxbgxb.20230508

Abstract

Abstract:

Considering the small size of calcified plaques and the difficulty in distinguishing them from non-plaque areas， a medical prior knowledge-guided optimization method for proposal generation was proposed. This method is based on the object detection network Faster R-CNN. Proposal generation is optimized through the location and shape of anchor. Calcified plaque definition is used to generate a guidance mask image to direct the location generation of proposal. Anchor shape prediction branch is employed to optimize the shape generation of proposal. A multi-scale guidance mask pyramid architecture is proposed for feature maps of different scales in FPN. The experimental results of detecting calcified plaques in CCTA images show that the proposed method improved AP and Recall by 12.8% and 25.7% respectively compared with Faster R-CNN with standard RPN； When the average number of false positives per image is 2， Recall of the proposed method is 86.05%.

Key words: computer application, calcified plaque detection, prior knowledge, deep learning, medical image

CLC Number:

TP391

Meng-xue ZHAO,Xiang-jiu CHE,Huan XU,Quan-le LIU. A method for generating proposals of medical image based on prior knowledge optimization[J].Journal of Jilin University(Engineering and Technology Edition), 2025, 55(2): 722-730.

Figures/Tables 10

Fig.1

Fig.2

Fig.3

Pseudo-code of the mask pyramid acquisition process"

Algorithm 1 Mask pyramid acquisition process

1： function get_mask_pyramid（img $I$ ， int $n$ ）：

2： # input： $I$ is the CT image file $(S * S)$

3： # input： $n$ is the serial number of the scale $n$ >2

4： # output： $M_p r ?$ is a mask image set $n$

5： # getting mask image $M$

6： $M$ = torch.zeros（ $[S, S]$ ）

7： while CT value at $I (i, j)$ is greater than or equal to 130 do

8： $M (i, j)$ = 1

9： end while

10： # getting mask pyramid $M_p r$

11： conv1 = MaxPool（kernel_size=4， stride=4， padding=0）

12： conv2 = MaxPool（kernel_size=2， stride=2， padding=0）

13： $M 1$ = conv1（ $M$ ）

14： $M_p r$ = $[M 1]$

15： for $t$ = 2 to $n$ do

16： $M t$ = conv2（ $M (t - 1)$ ）

17： $M_p r$ = $M_p r + [M t]$

18： end for

19： return $M_p r$

Fig.3

Fig.4

Fig.5

Table 1

Fig.6

Fig.7

Fig.8

Table 2

References 40

1	Naghavi M, Wang H, Lozano R, et al. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the global burden of disease study 2013[J]. The Lancet, 2015, 385(9963): 117-171.
2	Hutcheson J D, Goettsch C, Bertazzo S, et al. Genesis and growth of extracellular-vesicle-derived microcalcification in atherosclerotic plaques[J]. Nature Materials, 2016, 15(3): 335-343.
3	Išgum I, Van G B, Viergever M A. Automatic detection of calcifications in the aorta from abdominal CT scans[J]. International Congress Series, 2003, 1256(C): 1037-1042.
4	Isgum I, Prokop M, Niemeijer M, et al. Automatic coronary calcium scoring in low-dose chest computed tomography[J]. IEEE Transactions on Medical Imaging, 2012, 31(12): 2322-2334.
5	Wolterink J M, Leiner T, Takx R A P, et al. Automatic coronary calcium scoring in non-contrast-enhanced ECG-triggered cardiac CT with ambiguity detection[J]. IEEE Transactions on Medical Imaging, 2015, 34(9): 1867-1878.
6	Wolterink J M, Leiner T, Devos B D, et al. Automatic coronary artery calcium scoring in cardiac CT angiography using paired convolutional neural networks[J]. Medical Image Analysis, 2016, 34: 123-136.
7	Tourassi G D, Armato S G, Lessmann N, et al. Deep convolutional neural networks for automatic coronary calcium scoring in a screening study with low-dose chest CT[J].Medical Imaging 2016:Computer-Aided Diagnosis,2016, 9785: 255-260.
8	Liu J, Lu L, Yao J, et al. Pelvic artery calcification detection on CT scans using convolutional neural networks[J]. Conference on Computer-Aided Diagnosis,2017, 10134: 319-325.
9	Mohri M, Rostamizadeh A, Talwalkar A. Foundations of Machine Learning[M]. London: MIT Press, 2018.
10	Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[J]. Advances in Neural Information Processing Systems, 2012, 25: 1097-1105.
11	Lin T Y, Maire M, Belongie S, et al. Microsoft coco: common objects in context[J]. Computer Vision, 2014, 8693: 740-755.
12	Acharya U R, Meiburger K M, Koh J E W, et al. Automated detection of calcified plaque using higher-order spectra cumulant technique in computer tomography angiography images[J]. International Journal of Imaging Systems and Technology, 2020, 30: 285-297.
13	Li Y C, Shen T Y, Chen C C, et al. Automatic detection of atherosclerotic plaque and calcification from intravascular ultrasound images by using deep convolutional neural networks[J]. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 2021, 68(5): 1762-1772.
14	Chen C, Biffi C, Tarroni G, et al. Learning shape priors for robust cardiac mr segmentation from multi-view images[J]. Lecture Notes in Computer Science, 2019, 11765: 523-531.
15	Yue Q, Luo X, Ye Q, et al. Cardiac segmentation from LGE MRI using deep neural network incorporating shape and spatial priors[J]. Lecture Notes in Computer Science, 2019, 11765: 559-567.
16	Mirikharaji Z, Hamarneh G. Star shape prior in fully convolutional networks for skin lesion segmentation[C]∥Medical Image Computing and Computer Assisted Intervention, Granada, Spain, 201: 737-745.
17	Liao W, Rosenhahn B, Shuai L, et al. Natural language guided visual relationship detection[J]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, 2019, 6: 444-453.
18	Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]∥Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Columbus, USA, 2014: 580-587.
19	Yang G, Chen Y, Ning X, et al. Automatic coronary calcium scoring using noncontrast and contrast CT images[J]. Medical Physics, 2016, 43(5): 2174-2186.
20	Durlak F, Wels M, Schwemmer C, et al. Growing a random forest with fuzzy spatial features for fully automatic artery-specific coronary calcium scoring[J]. Lecture Notes in Computer Science, 2017, 10541: 27-35.
21	Zhang Y, Wang S, Zhao H, et al. CT image classification based on convolutional neural network[J]. Neural Computing and Applications, 2020, 33(14): 8191-8200.
22	Gao K, Su J, Jiang Z, et al. Dual-branch combination network (DCN): towards accurate diagnosis and lesion segmentation of COVID-19 using CT images[J]. Medical Image Analysis, 2021, 67: 101836.
23	Liang X, Li N, Zhang Z, et al. Incorporating the hybrid deformable model for improving the performance of abdominal CT segmentation via multi-scale feature fusion network[J]. Medical Image Analysis, 2021, 73: 102156.
24	Fan J, Cao X, Yap P T, et al. BIRNet: brain image registration using dual-supervised fully convolutional networks[J]. Medical Image Analysis, 2019, 54: 193-206.
25	Prasoon A, Petersen K, Igel C, et al. Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network[J]. Lecture Notes in Computer Science, 2013, 8150: 246-253.
26	Chellamuthu K, Liu J, Yao J, et al. Atherosclerotic vascular calcification detection and segmentation on low dose computed tomography scans using convolutional neural networks[C]∥IEEE 14th International Symposium on Biomedical Imaging, Melbourne, Australia, 2017: 388-391.
27	Gernaat S A M, Van S G M, Koh V, et al. Automatic quantification of calcifications in the coronary arteries and thoracic aorta on radiotherapy planning CT scans of Western and Asian breast cancer patients[J]. Radiotherapy and Oncology, 2018, 127(3): 487-492.
28	Lessmann N, Van Ginneken B, Zreik M, et al. Automatic calcium scoring in low-dose chest CT using deep neural networks with dilated convolutions[J]. IEEE Transactions on Medical Imaging, 2018, 37(2): 615-625.
29	Zreik M, Van H R W, Wolterink J M, et al. A recurrent CNN for automatic detection and classification of coronary artery plaque and stenosis in coronary CT angiography[J]. IEEE Transactions on Medical Imaging, 2019, 38(7): 1588-1598.
30	Fischer A M, Eid M, Dececco C N, et al. Accuracy of an artificial intelligence deep learning algorithm implementing a recurrent neural network with long short-term memory for the automated detection of calcified plaques from coronary computed tomography angiography[J]. Journal of Thoracic Imaging, 2020, 35: 49-57.
31	Ren S, He K, Girshick R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. Advances in Neural Information Processing Systems, 2015, 28: 91-99.
32	Lin T Y, Dollár P, Girshick R, et al. Feature pyramid networks for object detection[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, USA, 2017: 936-944.
33	Agatston A S, Janowitz W R, Hildner F J, et al. Quantification of coronary artery calcium using ultrafast computed tomography[J]. Journal of the American College of Cardiology, 1990, 15(4): 827-832.
34	Wang J, Chen K, Yang S, et al. Region proposal by guided anchoring[C]∥IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, USA, 2019: 2960-2969.
35	Lin T Y, Goyal P, Girshick R, et al. Focal loss for dense object detection[C]∥IEEE International Conference on Computer Vision, Venice, Italy, 2017: 2999-3007.
36	Wolterink J M, Leiner T, Devos B D, et al. An evaluation of automatic coronary artery calcium scoring methods with cardiac CT using the orCaScore framework[J]. Medical Physics, 2016, 43(5): 2361-2373.
37	Zhao M, Che X, Liu H, et al. Medical prior knowledge guided automatic detection of coronary arteries calcified plaque with cardiac CT[J]. Electronics (Switzerland), 2020, 9(12): 1-14.
38	Chen K, Wang J, Pang J, et al. MMdetection: open mmlab detection toolbox and benchmark[J]. Arxiv Preprint, 2019, 6: 190607155.
39	Sun P, Zhang R, Jiang Y, et al. Sparse R-CNN: end-to-end object detection with learnable proposals[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, USA, 2021: 14454-14463.
40	Ge Z, Liu S, Wang F, et al. Yolox: exceeding yolo series in 2021[J]. Arxiv Preprint, 2017, 7: 210708430.

Related Articles 15

[1]	Xiang-jiu CHE,Yu-ning WU,Quan-le LIU. A weighted isomorphic graph classification algorithm based on causal feature learning [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(2): 681-686.
[2]	Xiao-dong CAI,Qing-song ZHOU,Yan-yan ZHANG,Yun XUE. Social recommendation based on global capture of dynamic， static and relational features [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(2): 700-708.
[3]	Hui-zhi XU,Shi-sen JIANG,Xiu-qing WANG,Shuang CHEN. Vehicle target detection and ranging in vehicle image based on deep learning [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(1): 185-197.
[4]	Xiao-ran GUO,Tie-jun WANG,Yue YAN. Entity relationship extraction method based on local attention and local remote supervision [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(1): 307-315.
[5]	Hao WANG,Bin ZHAO,Guo-hua LIU. Temporal and motion enhancement for video action recognition [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(1): 339-346.
[6]	Yuan-ning LIU,Zi-nan ZANG,Hao ZHANG,Zhen LIU. Deep learning-based method for ribonucleic acid secondary structure prediction [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(1): 297-306.
[7]	Lei ZHANG,Jing JIAO,Bo-xin LI,Yan-jie ZHOU. Large capacity semi structured data extraction algorithm combining machine learning and deep learning [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(9): 2631-2637.
[8]	Lu Li,Jun-qi Song,Ming Zhu,He-qun Tan,Yu-fan Zhou,Chao-qi Sun,Cheng-yu Zhou. Object extraction of yellow catfish based on RGHS image enhancement and improved YOLOv5 network [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(9): 2638-2645.
[9]	Hong-wei ZHAO,Hong WU,Ke MA,Hai LI. Image classification framework based on knowledge distillation [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(8): 2307-2312.
[10]	Yun-zuo ZHANG,Yu-xin ZHENG,Cun-yu WU,Tian ZHANG. Accurate lane detection of complex environment based on double feature extraction network [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(7): 1894-1902.
[11]	Bai-you QIAO,Tong WU,Lu YANG,You-wen JIANG. A text sentiment analysis method based on BiGRU and capsule network [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(7): 2026-2037.
[12]	Xin-gang GUO,Ying-chen HE,Chao CHENG. Noise-resistant multistep image super resolution network [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(7): 2063-2071.
[13]	Li-ping ZHANG,Bin-yu LIU,Song LI,Zhong-xiao HAO. Trajectory k nearest neighbor query method based on sparse multi-head attention [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(6): 1756-1766.
[14]	Ming-hui SUN,Hao XUE,Yu-bo JIN,Wei-dong QU,Gui-he QIN. Video saliency prediction with collective spatio-temporal attention [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(6): 1767-1776.
[15]	Yan-feng LI,Ming-yang LIU,Jia-ming HU,Hua-dong SUN,Jie-yu MENG,Ao-ying WANG,Han-yue ZHANG,Hua-min YANG,Kai-xu HAN. Infrared and visible image fusion based on gradient transfer and auto-encoder [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(6): 1777-1787.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

Comments

Recommended 0

No Suggested Reading articles found!

方法	Recall	AP	参数量/MB	FLOPs/GB	推理时间/ms
Faster R-CNN^［31］	0.744	0.554	41.14	63.67	34.56
文献［26］	0.756	0.567	82.28	127.34	—
GA-Faster-RCNN^［34］	0.814	0.724	41.71	63.19	51.49
Sparse R-CNN^［39］	0.680	0.500	105.94	44.16	55.44
YOLOX^［40］	0.797	0.655	8.94	8.52	28.32
本文	0.872	0.811	41.71	63.18	65.63

方法	位置优化	形状优化	Recall	AP
Faster R-CNN^［31］			0.744	0.554
模型 A	√		0.773	0.663
模型 B		√	0.837	0.792
本文	√	√	0.872	0.811

A method for generating proposals of medical image based on prior knowledge optimization

RICH HTML

PDF (PC)