Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (9): 2975-2982.DOI: 10.11772/j.issn.1001-9081.2023091273
• Frontier and comprehensive applications • Previous Articles
Rui ZHANG(), Pengyun ZHANG, Meirong GAO
Received:
2023-09-18
Revised:
2024-01-29
Accepted:
2024-02-21
Online:
2024-03-19
Published:
2024-09-10
Contact:
Rui ZHANG
About author:
ZHANG Pengyun, born in 1999, M. S. candidate. His research interests include intelligent information processing.Supported by:
通讯作者:
张睿
作者简介:
张鹏云(1999—),男,山西太原人,硕士研究生,主要研究方向:智能信息处理基金资助:
CLC Number:
Rui ZHANG, Pengyun ZHANG, Meirong GAO. Self-optimized dual-modal multi-channel non-deep vestibular schwannoma recognition model[J]. Journal of Computer Applications, 2024, 44(9): 2975-2982.
张睿, 张鹏云, 高美蓉. 自优化双模态多通路非深度前庭神经鞘瘤识别模型[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2975-2982.
Add to citation manager EndNote|Ris|BibTeX
URL: https://www.joca.cn/EN/10.11772/j.issn.1001-9081.2023091273
超参数 | 搜索范围 | 超参数 | 搜索范围 |
---|---|---|---|
Op | Adam,SGD, Adamx,ASGD | Act | ReLU,Sigmoid,Tanh, LReLU,Softsign |
L_rate | [0.000 1,0.1] | C_k_x | [ |
Batch | [ | Parallel | [ |
Tab. 1 Search scope of each hyperparameter of DFMP model
超参数 | 搜索范围 | 超参数 | 搜索范围 |
---|---|---|---|
Op | Adam,SGD, Adamx,ASGD | Act | ReLU,Sigmoid,Tanh, LReLU,Softsign |
L_rate | [0.000 1,0.1] | C_k_x | [ |
Batch | [ | Parallel | [ |
基准函数 | SSA | DA | GWO | |||
---|---|---|---|---|---|---|
平均值 | 标准差 | 平均值 | 标准差 | 平均值 | 标准差 | |
F1 | 3.472E-35 | 1.654E-34 | 1.200E-02 | 3.200E-02 | 2.061E-34 | 3.192E-33 |
F2 | 1.153E-20 | 4.559E-20 | 1.818E+00 | 1.497E+00 | 1.155E-20 | 3.784E-19 |
F3 | 1.648E-26 | 8.040E-26 | 2.137E+02 | 4.291E+02 | 1.569E-08 | 1.515E-08 |
F4 | 2.652E-14 | 1.221E-13 | 1.895E+00 | 1.012E+00 | 2.024E-08 | 1.173E-08 |
F5 | 8.360E-02 | 1.988E-06 | 2.197E+00 | 9.501E-01 | 2.645E+01 | 1.376E-04 |
F6 | 1.580E-03 | 1.541E-10 | 1.167E+00 | 1.408E+00 | 2.223E-01 | 1.848E-01 |
F7 | 5.110E-04 | 1.660E-04 | 1.790E-03 | 1.051E-01 | 8.752E-04 | 4.663E-04 |
基准函数 | SFO | EO | GT-PSSA | |||
平均值 | 标准差 | 平均值 | 标准差 | 平均值 | 标准差 | |
F1 | 1.251E-12 | 2.408E+04 | 1.933E-81 | 5.938E-22 | 0.000 | 0.000 |
F2 | 2.077E-05 | 7.150E+00 | 1.683E-23 | 4.727E-13 | 0.000 | 0.000 |
F3 | 2.064E-13 | 7.743E+00 | 9.556E-23 | 0.064E+00 | 0.000 | 0.000 |
F4 | 1.053E-08 | 9.865E+04 | 1.022E-20 | 1.085E+00 | 0.000 | 0.000 |
F5 | 2.702E-06 | 1.465E+01 | 2.441E+01 | 3.543E+01 | 2.433E-08 | 1.074E-07 |
F6 | 8.550E-03 | 1.736E+01 | 9.455E-07 | 6.384E-04 | 1.245E-07 | 1.331E-12 |
F7 | 5.125E-05 | 5.536E-05 | 1.473E-03 | 2.641E-03 | 1.437E-05 | 1.157E-05 |
Tab. 2 Experimental results of seven single-peak benchmark functions by different algorithms
基准函数 | SSA | DA | GWO | |||
---|---|---|---|---|---|---|
平均值 | 标准差 | 平均值 | 标准差 | 平均值 | 标准差 | |
F1 | 3.472E-35 | 1.654E-34 | 1.200E-02 | 3.200E-02 | 2.061E-34 | 3.192E-33 |
F2 | 1.153E-20 | 4.559E-20 | 1.818E+00 | 1.497E+00 | 1.155E-20 | 3.784E-19 |
F3 | 1.648E-26 | 8.040E-26 | 2.137E+02 | 4.291E+02 | 1.569E-08 | 1.515E-08 |
F4 | 2.652E-14 | 1.221E-13 | 1.895E+00 | 1.012E+00 | 2.024E-08 | 1.173E-08 |
F5 | 8.360E-02 | 1.988E-06 | 2.197E+00 | 9.501E-01 | 2.645E+01 | 1.376E-04 |
F6 | 1.580E-03 | 1.541E-10 | 1.167E+00 | 1.408E+00 | 2.223E-01 | 1.848E-01 |
F7 | 5.110E-04 | 1.660E-04 | 1.790E-03 | 1.051E-01 | 8.752E-04 | 4.663E-04 |
基准函数 | SFO | EO | GT-PSSA | |||
平均值 | 标准差 | 平均值 | 标准差 | 平均值 | 标准差 | |
F1 | 1.251E-12 | 2.408E+04 | 1.933E-81 | 5.938E-22 | 0.000 | 0.000 |
F2 | 2.077E-05 | 7.150E+00 | 1.683E-23 | 4.727E-13 | 0.000 | 0.000 |
F3 | 2.064E-13 | 7.743E+00 | 9.556E-23 | 0.064E+00 | 0.000 | 0.000 |
F4 | 1.053E-08 | 9.865E+04 | 1.022E-20 | 1.085E+00 | 0.000 | 0.000 |
F5 | 2.702E-06 | 1.465E+01 | 2.441E+01 | 3.543E+01 | 2.433E-08 | 1.074E-07 |
F6 | 8.550E-03 | 1.736E+01 | 9.455E-07 | 6.384E-04 | 1.245E-07 | 1.331E-12 |
F7 | 5.125E-05 | 5.536E-05 | 1.473E-03 | 2.641E-03 | 1.437E-05 | 1.157E-05 |
基准函数 | SSA | DA | GWO | |||
---|---|---|---|---|---|---|
平均值 | 标准差 | 平均值 | 标准差 | 平均值 | 标准差 | |
F8 | -7.953E+03 | 7.204E+02 | -1.990E+03 | 1.701E+03 | -1.702E+03 | 1.473E+02 |
F9 | 0.000 | 0.000 | 1.218E+01 | 5.908E+00 | 3.047E-00 | 3.716E+00 |
F10 | 8.882E-16 | 0.000 | 2.204E+00 | 2.167E+00 | 7.021E-00 | 2.751E-14 |
F11 | 0.000 | 0.000 | 3.381E-01 | 1.243E-02 | 1.325E+01 | 4.741E-03 |
F12 | 3.352E-12 | 2.853E-11 | 5.014E-01 | 3.313E-01 | 1.826E-01 | 1.422E-04 |
基准函数 | SFO | EO | GT-PSSA | |||
平均值 | 标准差 | 平均值 | 标准差 | 平均值 | 标准差 | |
F8 | -1.250E+04 | 1.677E+00 | -8.398E+03 | 5.017E-01 | -1.255E+04 | 4.464E-02 |
F9 | 2.273E-13 | 1.216E-04 | 0.000 | 0.000 | 0.000 | 0.000 |
F10 | 4.702E-06 | 5.017E-01 | 3.996E-15 | 0.267E+00 | 4.440E-16 | 0.000 |
F11 | 2.386E-14 | 1.182E-01 | 0.000 | 0.000 | 0.000 | 0.000 |
F12 | 1.535E-02 | 4.443E-06 | 4.849E-07 | 4.502E-01 | 1.507E-07 | 9.679E-10 |
Tab. 3 Experimental results of five multi-peak benchmark functions by different algorithms
基准函数 | SSA | DA | GWO | |||
---|---|---|---|---|---|---|
平均值 | 标准差 | 平均值 | 标准差 | 平均值 | 标准差 | |
F8 | -7.953E+03 | 7.204E+02 | -1.990E+03 | 1.701E+03 | -1.702E+03 | 1.473E+02 |
F9 | 0.000 | 0.000 | 1.218E+01 | 5.908E+00 | 3.047E-00 | 3.716E+00 |
F10 | 8.882E-16 | 0.000 | 2.204E+00 | 2.167E+00 | 7.021E-00 | 2.751E-14 |
F11 | 0.000 | 0.000 | 3.381E-01 | 1.243E-02 | 1.325E+01 | 4.741E-03 |
F12 | 3.352E-12 | 2.853E-11 | 5.014E-01 | 3.313E-01 | 1.826E-01 | 1.422E-04 |
基准函数 | SFO | EO | GT-PSSA | |||
平均值 | 标准差 | 平均值 | 标准差 | 平均值 | 标准差 | |
F8 | -1.250E+04 | 1.677E+00 | -8.398E+03 | 5.017E-01 | -1.255E+04 | 4.464E-02 |
F9 | 2.273E-13 | 1.216E-04 | 0.000 | 0.000 | 0.000 | 0.000 |
F10 | 4.702E-06 | 5.017E-01 | 3.996E-15 | 0.267E+00 | 4.440E-16 | 0.000 |
F11 | 2.386E-14 | 1.182E-01 | 0.000 | 0.000 | 0.000 | 0.000 |
F12 | 1.535E-02 | 4.443E-06 | 4.849E-07 | 4.502E-01 | 1.507E-07 | 9.679E-10 |
基准函数 | 算法A | 算法B | 算法C | GT-PSSA | ||||
---|---|---|---|---|---|---|---|---|
平均值 | 运行时间/s | 平均值 | 运行时间/s | 平均值 | 运行时间/s | 平均值 | 运行时间/s | |
F1 | 3.472E-35 | 6.35 | 3.413E-77 | 6.57 | 7.202E-101 | 8.57 | 0.000 | 6.21 |
F2 | 1.153E-20 | 7.00 | 6.043E-121 | 7.14 | 6.078E-139 | 9.34 | 0.000 | 7.33 |
F3 | 1.648E-26 | 20.05 | 3.865E-41 | 25.68 | 2.754E-133 | 32.08 | 0.000 | 15.15 |
F4 | 2.652E-14 | 6.43 | 2.353E-57 | 6.71 | 2.395E-65 | 8.91 | 0.000 | 7.04 |
F5 | 8.360E-02 | 7.12 | 6.334E-02 | 8.02 | 6.144E-02 | 9.97 | 2.433E-08 | 7.13 |
F6 | 1.580E-03 | 6.50 | 5.047E-03 | 6.88 | 3.224E-04 | 9.01 | 1.245E-07 | 6.43 |
F7 | 5.110E-04 | 7.63 | 4.674E-04 | 8.11 | 4.340E-04 | 10.52 | 1.437E-05 | 7.61 |
F8 | -7.953E+03 | 6.51 | -8.878E+03 | 6.96 | -8.896E+03 | 7.67 | -1.255E+04 | 6.49 |
F9 | 0.000 | 6.40 | 0.000 | 7.16 | 0.000 | 8.97 | 0.000 | 6.38 |
F10 | 8.882E-16 | 7.02 | 4.440E-16 | 7.94 | 4.440E-16 | 9.29 | 4.440E-16 | 7.00 |
F11 | 0.000 | 7.28 | 0.000 | 8.23 | 0.000 | 9.98 | 0.000 | 7.19 |
F12 | 3.352E-12 | 8.50 | 2.908E-03 | 10.07 | 2.112E-03 | 12.97 | 1.507E-07 | 8.05 |
Tab. 4 Ablation experiment results of GT-PSSA
基准函数 | 算法A | 算法B | 算法C | GT-PSSA | ||||
---|---|---|---|---|---|---|---|---|
平均值 | 运行时间/s | 平均值 | 运行时间/s | 平均值 | 运行时间/s | 平均值 | 运行时间/s | |
F1 | 3.472E-35 | 6.35 | 3.413E-77 | 6.57 | 7.202E-101 | 8.57 | 0.000 | 6.21 |
F2 | 1.153E-20 | 7.00 | 6.043E-121 | 7.14 | 6.078E-139 | 9.34 | 0.000 | 7.33 |
F3 | 1.648E-26 | 20.05 | 3.865E-41 | 25.68 | 2.754E-133 | 32.08 | 0.000 | 15.15 |
F4 | 2.652E-14 | 6.43 | 2.353E-57 | 6.71 | 2.395E-65 | 8.91 | 0.000 | 7.04 |
F5 | 8.360E-02 | 7.12 | 6.334E-02 | 8.02 | 6.144E-02 | 9.97 | 2.433E-08 | 7.13 |
F6 | 1.580E-03 | 6.50 | 5.047E-03 | 6.88 | 3.224E-04 | 9.01 | 1.245E-07 | 6.43 |
F7 | 5.110E-04 | 7.63 | 4.674E-04 | 8.11 | 4.340E-04 | 10.52 | 1.437E-05 | 7.61 |
F8 | -7.953E+03 | 6.51 | -8.878E+03 | 6.96 | -8.896E+03 | 7.67 | -1.255E+04 | 6.49 |
F9 | 0.000 | 6.40 | 0.000 | 7.16 | 0.000 | 8.97 | 0.000 | 6.38 |
F10 | 8.882E-16 | 7.02 | 4.440E-16 | 7.94 | 4.440E-16 | 9.29 | 4.440E-16 | 7.00 |
F11 | 0.000 | 7.28 | 0.000 | 8.23 | 0.000 | 9.98 | 0.000 | 7.19 |
F12 | 3.352E-12 | 8.50 | 2.908E-03 | 10.07 | 2.112E-03 | 12.97 | 1.507E-07 | 8.05 |
超参数 | 搜索范围 | 超参数 | 搜索范围 |
---|---|---|---|
Op | Adam | Act | ReLU |
L_rate | 4×10-3 | C_k_x | 7 |
Batch | 36 | Parallel | 5 |
Tab. 5 Optimal DMFP model hyperparameters searched by GT-PSSA
超参数 | 搜索范围 | 超参数 | 搜索范围 |
---|---|---|---|
Op | Adam | Act | ReLU |
L_rate | 4×10-3 | C_k_x | 7 |
Batch | 36 | Parallel | 5 |
模型 | 训练时间/h | 准确率/% | 参数量/106 |
---|---|---|---|
文献[ | 2.08 | 87.01 | 4.3 |
CNN-2 | 1.06 | 88.13 | 9.9 |
ResNet-TL | 3.22 | 88.52 | 4.6 |
U-Net-DL | 2.23 | 87.87 | 7.9 |
TriA | 2.46 | 87.07 | 3.6 |
WDD | 0.98 | 87.98 | 7.3 |
DMFP-A | 0.67 | 88.65 | 3.3 |
DMFP | 0.66 | 91.20 | 3.1 |
Tab. 6 Performance comparison between DMFP and other models
模型 | 训练时间/h | 准确率/% | 参数量/106 |
---|---|---|---|
文献[ | 2.08 | 87.01 | 4.3 |
CNN-2 | 1.06 | 88.13 | 9.9 |
ResNet-TL | 3.22 | 88.52 | 4.6 |
U-Net-DL | 2.23 | 87.87 | 7.9 |
TriA | 2.46 | 87.07 | 3.6 |
WDD | 0.98 | 87.98 | 7.3 |
DMFP-A | 0.67 | 88.65 | 3.3 |
DMFP | 0.66 | 91.20 | 3.1 |
融合方法 | 训练时间/h | 准确率/% |
---|---|---|
Add | 1.08 | 87.27 |
Concat | 1.17 | 87.03 |
DFFF | 0.66 | 91.20 |
Tab. 7 Performance comparison results of three different fusion methods
融合方法 | 训练时间/h | 准确率/% |
---|---|---|
Add | 1.08 | 87.27 |
Concat | 1.17 | 87.03 |
DFFF | 0.66 | 91.20 |
1 | MULÉ S, REIZINE E, BLANC-DURAND P, et al. Whole-body functional MRI and PET/MRI in multiple myeloma [J]. Cancers, 2020, 12(11): 3155. |
2 | WU C-C, GUO W-Y, W-Y CHUNG, et al. Magnetic resonance imaging characteristics and the prediction of outcome of vestibular schwannomas following Gamma Knife radiosurgery [J]. Journal of Neurosurgery, 2017, 127(6): 1384-1391. |
3 | CHEN Z, MO X, CHEN R, et al. A reparametrized CNN model to distinguish Alzheimer’s disease applying multiple morphological metrics and deep semantic features from structural MRI [J]. Frontiers in Aging Neuroscience, 2022, 14: 856391. |
4 | GOYAL A, BOCHKOVSKIY A, DENG J, et al. Non-deep networks [C]// Proceedings of the 36th Conference on Neural Information Processing Systems. Red Hook: Curran Associates Inc., 2022: 6789-6801. |
5 | TAMMINA S. Transfer learning using VGG-16 with deep convolutional neural network for classifying images[J]. International Journal of Scientific and Research Publications, 2019, 9(10): 143-150. |
6 | JAEGER P F, KOHL S, BICKELHAUPT S, et al. Retina U-Net: embarrassingly simple exploitation of segmentation supervision for medical object detection [C]// Proceedings of the 2020 Machine Learning for Health NeurIPS Workshop. New York: PMLR, 2020, 116: 171-183. |
7 | STRAUSS S, STERN S, LANTOS J E, et al. High-resolution T2-weighted imaging for surveillance in postoperative vestibular schwannoma: equivalence with contrast-enhanced T1WI for measurement and surveillance of residual tumor [J]. American Journal of Neuroradiology, 2022, 43(12): 1792-1796. |
8 | WANG G, SHAPEY J, LI W, et al. Automatic segmentation of vestibular schwannoma from T2-weighted MRI by deep spatial attention with hardness-weighted loss [C]// Proceedings of the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention. Cham: Springer, 2019: 264-272. |
9 | SHAPEY J, WANG G, DORENT R, et al. An artificial intelligence framework for automatic segmentation and volumetry of vestibular schwannomas from contrast-enhanced T1-weighted and high-resolution T2-weighted MRI [J]. Journal of Neurosurgery, 2019, 134(1): 171-179. |
10 | LEE W-K, WU C-C, LEE C-C, et al. Combining analysis of multi-parametric MR images into a convolutional neural network: precise target delineation for vestibular schwannoma treatment planning [J]. Artificial Intelligence in Medicine, 2020, 107: 101911. |
11 | CHEN P, CHEN M, WANG S, et al. Real-time defect detection of TFT-LCD displays using a lightweight network architecture [J]. Journal of Intelligent Manufacturing, 2024, 35: 1337-1352. |
12 | XUE J, SHEN B. A novel swarm intelligence optimization approach: sparrow search algorithm [J]. Systems Science and Control Engineering, 2020, 8(1): 22-34. |
13 | LIU G, SHU C, LIANG Z, et al. A modified sparrow search algorithm with application in 3D route planning for UAV [J]. Sensors, 2021, 21(4): 1224. |
14 | YUAN J, ZHAO Z, LIU Y, et al. DMPPT control of photovoltaic microgrid based on improved sparrow search algorithm [J]. IEEE Access, 2021, 9: 16623-16629. |
15 | YANG J, GAO S, ZHAO X, et al. Enhanced sparrow search algorithm based on improved game predatory mechanism and its application [J]. Digital Signal Processing, 2024, 145: 104310. |
16 | DENG X, WEN J, WANG Z. Zero-sum polymatrix games with link uncertainty: a Dempster-Shafer theory solution [J]. Applied Mathematics and Computation, 2019, 340: 101-112. |
17 | LI C, FENG B, LI S, et al. Dynamic analysis of digital chaotic maps via state-mapping networks [J]. IEEE Transactions on Circuits and Systems Ⅰ: Regular Papers, 2019, 66(6): 2322-2335. |
18 | BOVEIRI H R, ELHOSENY M. A-COA: an adaptive cuckoo optimization algorithm for continuous and combinatorial optimization[J]. Neural Computing and Applications, 2020, 32: 681-705. |
19 | SHAPEY J, KUJAWA A, DORENT R, et al. Segmentation of vestibular schwannoma from MRI: an open annotated dataset and baseline algorithm [J]. Scientific Data, 2021, 8: 286. |
20 | WANG L, SHI R, DONG J. A hybridization of dragonfly algorithm optimization and angle modulation mechanism for 0-1 knapsack problems [J]. Entropy, 2021, 23(5): 598. |
21 | MIRJALILI S, MIRJALILI S M, LEWIS A. Grey wolf optimizer[J]. Advances in Engineering Software, 2014, 69: 46-61. |
22 | GOMES G F, CUNHA S S DA, Jr, ANCELOTTI A C. A sunflower optimization (SFO) algorithm applied to damage identification on laminated composite plates [J]. Engineering with Computers, 2019, 35(2): 619-626. |
23 | FARAMARZI A, HEIDARINEJAD M, STEPHENS B, et al. Equilibrium optimizer: a novel optimization algorithm [J]. Knowledge-Based Systems, 2020, 191: 105190. |
24 | Y-A CHOI, S-J PARK, J-A JUN, et al. Deep learning-based stroke disease prediction system using real-time bio signals [J]. Sensors, 2021, 21(13): 4269. |
25 | CHEN Y-T, CHEN Y-L, CHEN Y-Y, et al. Deep learning-based brain computed tomography image classification with hyperparameter optimization through transfer learning for stroke[J]. Diagnostics, 2022, 12(4): 807. |
26 | RAO B N, MOHANTY S, SEN K, et al. Deep transfer learning for automatic prediction of hemorrhagic stroke on CT images [J]. Computational and Mathematical Methods in Medicine, 2022, 2022: 3560507. |
27 | LI L, WEI M, LIU B, et al. Deep learning for hemorrhagic lesion detection and segmentation on brain CT images [J]. IEEE Journal of Biomedical and Health Informatics, 2020, 25(5): 1646-1659. |
28 | ALZAHRANI S, QAHTANI A. Knowledge distillation in Transformers with tripartite attention: multiclass brain tumor detection in highly augmented MRIs [J]. Journal of King Saud University — Computer and Information Sciences, 2024, 36(1): 101907. |
29 | ZHANG R, GAO M-R, ZHANG P-Y, et al. Research on an ultrasonic detection method for weld defects based on neural network architecture search [J]. Measurement, 2023, 211: 113483. |
[1] | Rui ZHANG, Siqi SONG, Jing HU, Yongmei ZHANG, Yanfeng CHAI. Performance evaluation of industry-university-research based on statistics and adaptive ParNet [J]. Journal of Computer Applications, 2024, 44(2): 628-637. |
[2] | Yang WANG, Shijie JIANG, Yucong CAO, Chuanwen LI. Parallel pivoted subgraph matching with multiple coding trees on GPU [J]. Journal of Computer Applications, 2022, 42(1): 132-139. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||