Top

Journal of Cancer Research and Clinical Oncology

Published in:

Open Access 20-11-2023 | Breast Cancer | Research

Graph neural network-based breast cancer diagnosis using ultrasound images with optimized graph construction integrating the medically significant features

Authors: Sadia Sultana Chowa, Sami Azam, Sidratul Montaha, Israt Jahan Payel, Md Rahad Islam Bhuiyan, Md. Zahid Hasan, Mirjam Jonkman

Published in: Journal of Cancer Research and Clinical Oncology | Issue 20/2023

Abstract

Purpose

An automated computerized approach can aid radiologists in the early diagnosis of breast cancer. In this study, a novel method is proposed for classifying breast tumors into benign and malignant, based on the ultrasound images through a Graph Neural Network (GNN) model utilizing clinically significant features.

Method

Ten informative features are extracted from the region of interest (ROI), based on the radiologists’ diagnosis markers. The significance of the features is evaluated using density plot and T test statistical analysis method. A feature table is generated where each row represents individual image, considered as node, and the edges between the nodes are denoted by calculating the Spearman correlation coefficient. A graph dataset is generated and fed into the GNN model. The model is configured through ablation study and Bayesian optimization. The optimized model is then evaluated with different correlation thresholds for getting the highest performance with a shallow graph. The performance consistency is validated with k-fold cross validation. The impact of utilizing ROIs and handcrafted features for breast tumor classification is evaluated by comparing the model’s performance with Histogram of Oriented Gradients (HOG) descriptor features from the entire ultrasound image. Lastly, a clustering-based analysis is performed to generate a new filtered graph, considering weak and strong relationships of the nodes, based on the similarities.

Results

The results indicate that with a threshold value of 0.95, the GNN model achieves the highest test accuracy of 99.48%, precision and recall of 100%, and F1 score of 99.28%, reducing the number of edges by 85.5%. The GNN model’s performance is 86.91%, considering no threshold value for the graph generated from HOG descriptor features. Different threshold values for the Spearman’s correlation score are experimented with and the performance is compared. No significant differences are observed between the previous graph and the filtered graph.

Conclusion

The proposed approach might aid the radiologists in effective diagnosing and learning tumor pattern of breast cancer.

Abadal S, Jain A, Guirado R, López-Alonso J, Alarcón E (2022) Computing graph neural networks: a survey from algorithms to accelerators. ACM Comput Surveys. https://doi.org/10.1145/3477141CrossRef

Ahila A, Poongodi P, Bourouis S, Band SS, Mosavi A, Agrawal S, Hamdi M (2022) Meta-Heuristic algorithm-tuned neural network for breast cancer diagnosis using ultrasound images. Front Oncol. https://doi.org/10.3389/fonc.2022.834028CrossRef

Al-Dhabyani W, Gomaa M, Khaled H, Fahmy A (2020) Dataset of breast ultrasound images. Data Brief. https://doi.org/10.1016/j.dib.2019.104863CrossRefPubMed

Asif NA, Sarker Y, Chakrabortty RK, Ryan MJ, Hafiz Ahamed M, Saha DK, Badal FR et al (2021) Graph neural network: a comprehensive review on non-euclidean space. IEEE Access. Institute of Electrical and Electronics Engineers Inc. https://doi.org/10.1109/ACCESS.2021.3071274

Aswiga RV, Aishwarya R, Shanthi AP (2021) Augmenting transfer learning with feature extraction techniques for limited breast imaging datasets. J Digit Imaging 34(3):618–629. https://doi.org/10.1007/s10278-021-00456-zCrossRef

Atrey K, Singh BK, Bodhey NK (2023) Multimodal classification of breast cancer using feature level fusion of mammogram and ultrasound images in machine learning paradigm. Multimedia Tools Appl. https://doi.org/10.1007/s11042-023-16414-6CrossRef

Ayana G, Dese K, Raj H, Krishnamoorthy J, Kwa T (2022) De-speckling breast cancer ultrasound images using a rotationally invariant block matching based non-local means (RIBM-NLM) method. Diagnostics. https://doi.org/10.3390/diagnostics12040862CrossRefPubMedPubMedCentral

Berbar MA (2018) Hybrid methods for feature extraction for breast masses classification. Egypt Inform J 19(1):63–73. https://doi.org/10.1016/j.eij.2017.08.001CrossRef

Boulenger A, Luo Y, Zhang C, Zhao C, Gao Y, Xiao M, Zhu Q, Tang J (2023) Deep learning-based system for automatic prediction of triple-negative breast cancer from ultrasound images. Med Biol Eng Comput 61(2):567–578. https://doi.org/10.1007/s11517-022-02728-4CrossRefPubMed

Brunetti N, Calabrese M, Martinoli C, Tagliafico AS (2023) Artificial intelligence in breast ultrasound: from diagnosis to prognosis—a rapid review. Diagnostics. https://doi.org/10.3390/diagnostics13010058CrossRefPubMedPubMedCentral

Byra M (2021) Breast mass classification with transfer learning based on scaling of deep representations. Biomed Signal Proces Control. https://doi.org/10.1016/j.bspc.2021.102828CrossRef

Cao Z, Duan L, Yang G, Yue T, Chen Q (2019) An experimental study on breast lesion detection and classification from ultrasound images using deep learning architectures. BMC Med Imaging. https://doi.org/10.1186/s12880-019-0349-xCrossRefPubMedPubMedCentral

Chen G, Dai Y, Zhang J (2023) RRCNet: refinement residual convolutional network for breast ultrasound images segmentation. Eng Appl Artif Intell. https://doi.org/10.1016/j.engappai.2022.105601CrossRefPubMedPubMedCentral

Cruz-Ramos C, García-Avila O, Almaraz-Damian J-A, Ponomaryov V, Reyes-Reyes R, Sadovnychiy S (2023) Benign and malignant breast tumor classification in ultrasound and mammography images via fusion of deep learning and handcraft features. Entropy 25(7):991. https://doi.org/10.3390/e25070991CrossRefPubMedPubMedCentral

Daoud MI, Abdel-Rahman S, Bdair TM, Al-Najar MS, Al-Hawari FH, Alazrai R (2020) Breast tumor classification in ultrasound images using combined deep and handcrafted features. Sensors (switz) 20(23):1–20. https://doi.org/10.3390/s20236838CrossRef

Deb SD, Jha RK (2023) Breast ultrasound image classification using fuzzy-rank-based ensemble network. Biomed Signal Proces Control. https://doi.org/10.1016/j.bspc.2023.104871CrossRef

Dhillon A, Singh A, Bhalla VK (2023) Biomarker identification and cancer survival prediction using random spatial local best cat swarm and Bayesian optimized DNN. Appl Soft Comput. https://doi.org/10.1016/j.asoc.2023.110649CrossRef

Effrosynidis D, Arampatzis A (2021) An evaluation of feature selection methods for environmental data. Ecol Inform. https://doi.org/10.1016/j.ecoinf.2021.101224CrossRef

Gayathri JL, Abraham B, Sujarani MS, Nair MS (2022) A computer-aided diagnosis system for the classification of COVID-19 and Non-COVID-19 pneumonia on chest X-ray images by integrating CNN with sparse autoencoder and feed forward neural network. Comput Biol Med. https://doi.org/10.1016/j.compbiomed.2021.105134CrossRef

Gedik N (2016) A new feature extraction method based on multi-resolution representations of mammograms. Appl Soft Comput J 44:128–133. https://doi.org/10.1016/j.asoc.2016.04.004CrossRef

Grattarola D, Alippi C (2021) Graph neural networks in Tensorflow and Keras with Spektral [application notes]. IEEE Comput Intell Mag 16(1):99–106. https://doi.org/10.1109/MCI.2020.3039072CrossRef

He Q, Yang Q, Xie M (2023) HCTNet: a hybrid CNN-transformer network for breast ultrasound image segmentation. Comput Biol Med. https://doi.org/10.1016/j.compbiomed.2023.106629CrossRefPubMed

Huang Q, Huang Y, Luo Y, Yuan F, Li X (2020) Segmentation of breast ultrasound image with semantic classification of superpixels. Med Image Analysis. https://doi.org/10.1016/j.media.2020.101657CrossRef

Jabeen K, Khan MA, Alhaisoni M, Usman Tariq Y, Zhang D, Hamza A, Mickus A, Damaševičius R (2022) Breast cancer classification from ultrasound images using probability-based optimal deep learning feature fusion. Sensors. https://doi.org/10.3390/s22030807CrossRefPubMedPubMedCentral

Khan IU, Azam S, Montaha S, Mahmud AAl, Rafid AKMRH, Hasan MZ, Jonkman M (2022) An effective approach to address processing time and computational complexity employing modified CCT for lung disease classification. Intell Syst Appl. https://doi.org/10.1016/j.iswa.2022.200147

Kriti JV, Agarwal R (2020) Deep feature extraction and classification of breast ultrasound images. Multimedia Tools Appl 79(37–38):27257–27292. https://doi.org/10.1007/s11042-020-09337-zCrossRef

Lim H, Kim DW (2021) Pairwise dependence-based unsupervised feature selection. Pattern Recogn. https://doi.org/10.1016/j.patcog.2020.107663CrossRef

Liu Bo, Cheng HD, Huang J, Tian J, Tang X, Liu J (2010) Fully automatic and segmentation-robust classification of breast tumors based on local texture analysis of ultrasound images. Pattern Recogn 43(1):280–298. https://doi.org/10.1016/j.patcog.2009.06.002CrossRef

Liu Y, Ren L, Cao X, Tong Y (2020) Breast tumors recognition based on edge feature extraction using support vector machine. Biomed Signal Processing Control. https://doi.org/10.1016/j.bspc.2019.101825CrossRef

Liu W, Guo M, Liu P, Yongzhao D (2022) MfdcModel: a novel classification model for classification of benign and malignant breast tumors in ultrasound images. Electronics (switz). https://doi.org/10.3390/electronics11162583CrossRef

Mendelson EB, Böhm-Vélez M, Berg, Whitman GJ, Madjar H, Rizzatto G, Baker JA et al. n.d. “ACR BI-RADS® Ultrasound 2013 As of 12/05/2013”

Meng H, Liu X, Niu J, Wang Y, Liao J, Li Q, Chen C (2023) DGANet: a dual global attention neural network for breast lesion detection in ultrasound images. Ultrasound Med Biol 49(1):31–44. https://doi.org/10.1016/j.ultrasmedbio.2022.07.006CrossRefPubMed

Mishra AK, Roy P, Bandyopadhyay S, Das SK (2021) Breast ultrasound tumour classification: a machine learning—radiomics based approach. Expert Syst. https://doi.org/10.1111/exsy.12713CrossRefPubMedPubMedCentral

Mohammed MA, Al-Khateeb B, Rashid AN, Ibrahim DA, Ghani MKA, Mostafa SA (2018) Neural network and multi-fractal dimension features for breast cancer classification from ultrasound images. Comput Electr Eng 70:871–882. https://doi.org/10.1016/j.compeleceng.2018.01.033CrossRef

Montaha S, Azam S, Rafid AKMRH, Pronab Ghosh M, Hasan Z, Jonkman M, De Boer F (2021) BreastNet18: a high accuracy fine-tuned VGG16 model evaluated using ablation study for diagnosing breast cancer from enhanced mammography images. Biology. https://doi.org/10.3390/biology10121347CrossRefPubMedPubMedCentral

Moon WK, Shen YW, Huang CS, Chiang LR, Chang RF (2011) Computer-aided diagnosis for the classification of breast masses in automated whole breast ultrasound images. Ultrasound Med Biol 37(4):539–548. https://doi.org/10.1016/j.ultrasmedbio.2011.01.006CrossRefPubMed

Moon WK, Lee YW, Ke HH, Lee SH, Huang CS, Chang RF (2020) Computer-aided diagnosis of breast ultrasound images using ensemble learning from convolutional neural networks. Comput Methods Programs Biomed. https://doi.org/10.1016/j.cmpb.2020.105361CrossRefPubMed

Özcan H (2023) BUS−CAD: a computer-aided diagnosis system for breast tumor classification in ultrasound images using grid-search-optimized machine learning algorithms with extended and Boruta-selected features. Int J Imaging Syst Technol. https://doi.org/10.1002/ima.22873CrossRef

Rafid AKM, Haque R, Azam S, Montaha S, Karim A, Fahim KU, Hasan MZ (2022) An effective ensemble machine learning approach to classify breast cancer based on feature selection and lesion segmentation using preprocessed mammograms. Biology. https://doi.org/10.3390/biology11111654CrossRefPubMedPubMedCentral

Ranjbarzadeh R, Sarshar NT, Ghoushchi SJ, Esfahani MS, Parhizkar M, Pourasad Y, Anari S, Bendechache M (2022) MRFE-CNN: multi-route feature extraction model for breast tumor segmentation in mammograms using a convolutional neural network. Annals Operat Res. https://doi.org/10.1007/s10479-022-04755-8CrossRef

Sahu A, Das PK, Meher S (2023) High accuracy hybrid CNN classifiers for breast cancer detection using mammogram and ultrasound datasets. Biomed Signal Process Control. https://doi.org/10.1016/j.bspc.2022.104292CrossRef

Sajid U, Khan RA, Shah SM, Arif S (2023) Breast cancer classification using deep learned features boosted with handcrafted features. Biomed Signal Process Control. https://doi.org/10.1016/j.bspc.2023.105353CrossRef

Sayed AM, Zaghloul E, Nassef TM (2016) Automatic classification of breast tumors using features extracted from magnetic resonance images. Proc Comput Sci 95:392–398. https://doi.org/10.1016/j.procs.2016.09.350CrossRef

Sellami L, Ben Sassi O, Chtourou K, Ben Hamida A (2015) Breast cancer ultrasound images’ sequence exploration using BI-RADS features’ extraction: towards an advanced clinical aided tool for precise lesion characterization. IEEE Trans Nanobiosci 14(7):740–745. https://doi.org/10.1109/TNB.2015.2486621CrossRef

Shia WC, Chen DR (2021) Classification of malignant tumors in breast ultrasound using a pretrained deep residual network model and support vector machine. Comput Med Imaging Graphics. https://doi.org/10.1016/j.compmedimag.2020.101829CrossRef

Shia WC, Lin LS, Chen DR (2021) Classification of malignant tumours in breast ultrasound using unsupervised machine learning approaches. Sci Rep. https://doi.org/10.1038/s41598-021-81008-xCrossRefPubMedPubMedCentral

Singh J, Thakur D, Ali F, Gera T, Kwak KS (2020) Deep feature extraction and classification of android malware images. Sensors (switz) 20(24):1–29. https://doi.org/10.3390/s20247013CrossRef

Sirjani N, Oghli MG, Tarzamni MK, Gity M, Shabanzadeh A, Ghaderi P, Shiri I, Akhavan A, Faraji M, Taghipour M (2023) A novel deep learning model for breast lesion classification using ultrasound images: a multicenter data evaluation. Phys Med. https://doi.org/10.1016/j.ejmp.2023.102560CrossRefPubMed

Telagarapu P, Poonguzhali S (2018) Assessment of texture feature extraction to classify the benign and malignant lesions from breast ultrasound images. Adv Intell Syst Comput 668:725–732. https://doi.org/10.1007/978-981-10-7868-2_69CrossRef

“The Graph Neural Network Model.” n.d.

Triyani Y, Khabzli W, Styorini W (2023) Margin, orientation, and posterior. J Electron Electromed Eng Med Inform 5(2):75–81. https://doi.org/10.35882/jeemi.v5i2.286CrossRef

Truong TT, Dinh-Cong D, Lee J, Nguyen-Thoi T (2020) An effective deep feedforward neural networks (DFNN) method for damage identification of truss structures using noisy incomplete modal data. J Build Eng. https://doi.org/10.1016/j.jobe.2020.101244CrossRef

Umer MJ, Sharif M, Wang SH (2022) Breast cancer classification and segmentation framework using multiscale CNN and U-shaped dual decoded attention network. Expert Syst. https://doi.org/10.1111/exsy.13192CrossRef

Wu Z, Pan S, Chen F, Long G, Zhang C, Philip SYu (2021) A comprehensive survey on graph neural networks. IEEE Transact Neural Netw Learn Syst 32(1):4–24. https://doi.org/10.1109/TNNLS.2020.2978386CrossRef

Yu H, Sun H, Li J, Shi L, Bao N, Li H, Qian W, Zhou S (2021) Effective diagnostic model construction based on discriminative breast ultrasound image regions using deep feature extraction. Med Phys 48(6):2920–2928. https://doi.org/10.1002/mp.14832CrossRefPubMed

Zhang X, Cui H, Nana Hu, Han P, Fan W, Wang P, Zuo X et al (2023) Correlation of androgen receptor with ultrasound, clinicopathological features and clinical outcomes in breast cancer. Insights Imaging 14(1):46. https://doi.org/10.1186/s13244-023-01387-9CrossRefPubMedPubMedCentral

Zhou S, Shi J, Zhu J, Cai Y, Wang R (2013) Shearlet-based texture feature extraction for classification of breast tumor in ultrasound image. Biomed Signal Process Control 8(6):688–696. https://doi.org/10.1016/j.bspc.2013.06.011CrossRef

Zhou Q, Wang Q, Bao Y, Kong L, Jin X, Weihua O (2022) LAEDNet: a lightweight attention encoder–decoder network for ultrasound medical image segmentation. Comput Electrical Eng. https://doi.org/10.1016/j.compeleceng.2022.107777CrossRef

Zhuang Z, Yang Z, Raj ANJ, Wei C, Jin P, Zhuang S (2021) Breast ultrasound tumor image classification using image decomposition and fusion based on adaptive multi-model spatial feature fusion. Comput Methods Programs Biomed. https://doi.org/10.1016/j.cmpb.2021.106221CrossRefPubMed

Title: Graph neural network-based breast cancer diagnosis using ultrasound images with optimized graph construction integrating the medically significant features
Authors: Sadia Sultana Chowa
Sami Azam
Sidratul Montaha
Israt Jahan Payel
Md Rahad Islam Bhuiyan
Md. Zahid Hasan
Mirjam Jonkman
Publication date: 20-11-2023
Publisher: Springer Berlin Heidelberg
Keywords: Breast Cancer
Breast Cancer
Breast Ultrasound
Published in: Journal of Cancer Research and Clinical Oncology / Issue 20/2023
Print ISSN: 0171-5216
Electronic ISSN: 1432-1335
DOI: https://doi.org/10.1007/s00432-023-05464-w

2024 ESC Congress

Noise pollution dangerous for heart health

02-09-2024
Myocardial Infarction
News

Two studies underscore the adverse impact of noise pollution on the risk for and prognosis after myocardial infarction.

Antihypertensives: more effective at morning or night?

02-09-2024
Hypertension
News

Low Lp(a) levels unlikely to increase the risk of diabetes

02-09-2024
Type 2 Diabetes
News

» View more highlights on the 2024 ESC Congress hub page

A quick guide to ECGs

Electrocardiography
Training Course

Improve your ECG interpretation skills with this comprehensive, rapid, interactive course. Expert advice provides detailed feedback as you work through 50 ECGs covering the most common cardiac presentations to ensure your practice stays up to date.

PD Dr. Carsten W. Israel

Developed by: Springer Medizin

Start the cases

At a glance: The STEP trials

Obesity
Clinical Trial Summary

A round-up of the STEP phase 3 clinical trials evaluating semaglutide for weight loss in people with overweight or obesity.

Developed by: Springer Medicine

EASD 2024

Springer Medicine

Graph neural network-based breast cancer diagnosis using ultrasound images with optimized graph construction integrating the medically significant features

Abstract

Purpose

Method

Results

Conclusion

2024 ESC Congress

Noise pollution dangerous for heart health

Post-TAVI follow up: deploying virtual assistants

Antihypertensives: more effective at morning or night?

Low Lp(a) levels unlikely to increase the risk of diabetes

» View more highlights on the 2024 ESC Congress hub page

A quick guide to ECGs

At a glance: The STEP trials

EASD 2024

Springer Medicine

Abstract

Purpose

Method

Results

Conclusion

Please log in to get access to this content

Other articles of this Issue 20/2023

Epidemiological and clinical characteristics of children with peripheral neuroblastic tumors: a study on a Moroccan population

EGR3 and estrone are involved in the tamoxifen resistance and progression of breast cancer

High density of CXCL12-positive immune cell infiltration predicts chemosensitivity and recurrence-free survival in ovarian carcinoma

Proposal of a diagnostic algorithm for radiation-induced dropped head syndrome in long-term childhood cancer survivors based on a prospective study in a specialized clinical setting and a review of the literature

Effects of the CALM intervention on resilience in Chinese patients with early breast cancer: a randomized trial

Mortality due to non-AIDS-defining cancers among people living with HIV in Spain over 18 years of follow-up

2024 ESC Congress

Noise pollution dangerous for heart health

Post-TAVI follow up: deploying virtual assistants

Antihypertensives: more effective at morning or night?

Low Lp(a) levels unlikely to increase the risk of diabetes

» View more highlights on the 2024 ESC Congress hub page

A quick guide to ECGs

At a glance: The STEP trials