End-to-end Deep Learning Radiomics: Development and Validation of a Novel Attention-based Aggregate Convolutional Neural Network to Distinguish Breast Diffuse Large B-cell Lymphoma from Breast Invasive Ductal Carcinoma

Overview

Journal Quant Imaging Med Surg

Specialty Radiology

Date 2023 Oct 23

PMID 37869296

Authors

Wen Chen

Fei Liu

Rui Wang

Ming Qi

Jianping Zhang

Xiaosheng Liu

Shaoli Song

Affiliations

Soon will be listed here.

Abstract

Background: Apart from invasive pathological examination, there is no effective method to differentiate breast diffuse large B-cell lymphoma (DLBCL) from breast invasive ductal carcinoma (IDC). In this study, we aimed to develop and validate an effective deep learning radiomics model to discriminate between DLBCL and IDC.

Methods: A total of 324 breast nodules from 236 patients with baseline F-fluorodeoxyglucose (F-FDG) positron emission tomography/computed tomography (PET/CT) were retrospectively analyzed. After grouping breast DLBCL and breast IDC patients, external and internal datasets were divided according to the data collected by different centers. Preprocessing was then used to process the original PET/CT images and an attention-based aggregate convolutional neural network (AACNN) model was designed. The AACNN model was trained using patches of CT or PET tumor images and optimized with an improved loss function. The final ensemble predictive model was built using distance weight voting. Finally, the model performance was evaluated and statistically verified.

Results: A total of 249 breast nodules from Fudan University Shanghai Cancer Center (FUSCC) and 75 breast nodules from Shanghai Proton and Heavy Ion Center (SPHIC) were selected as internal and external datasets, respectively. On the internal testing, our method yielded an area under the curve (AUC), accuracy (ACC), sensitivity (SEN), specificity (SPE), positive predictive value (PPV), negative predictive value (NPV), and harmonic mean of precision and sensitivity (F1) of 0.886, 83.0%, 80.9%, 85.0%, 84.8%, 81.2%, and 0.828, respectively. Meanwhile on the external testing, the results were 0.788, 71.6%, 61.4%, 84.7%, 84.0%, 62.6%, and 0.709, respectively.

Conclusions: Our study outlines a deep learning radiomics method which can automatically, noninvasively, and accurately differentiate breast DLBCL from breast IDC, which will be more in line with the needs and strategies of precision medicine, individualized diagnosis, and treatment.

Citing Articles

Ultrasound-based radiomics and clinical factors-based nomogram for early intracranial hypertension detection in patients with decompressive craniotomy.

Fu Z, Peng L, Guo L, Qin C, Yu Y, Zhang J Front Med Technol. 2025; 7:1485244.

PMID: 39974430 PMC: 11835818. DOI: 10.3389/fmedt.2025.1485244.

A Systematic Review of the Applications of Deep Learning for the Interpretation of Positron Emission Tomography Images of Patients with Lymphoma.

Kanavos T, Birbas E, Zanos T Cancers (Basel). 2025; 17(1.

PMID: 39796698 PMC: 11719749. DOI: 10.3390/cancers17010069.

Enhancing Lymphoma Diagnosis, Treatment, and Follow-Up Using F-FDG PET/CT Imaging: Contribution of Artificial Intelligence and Radiomics Analysis.

Hasanabadi S, Aghamiri S, Abin A, Abdollahi H, Arabi H, Zaidi H Cancers (Basel). 2024; 16(20).

PMID: 39456604 PMC: 11505665. DOI: 10.3390/cancers16203511.

References

Cheson B . Staging and response assessment in lymphomas: the new Lugano classification. Chin Clin Oncol. 2015; 4(1):5. DOI: 10.3978/j.issn.2304-3865.2014.11.03. View

Mohanty S, Hughes D, Salathe M . Using Deep Learning for Image-Based Plant Disease Detection. Front Plant Sci. 2016; 7:1419. PMC: 5032846. DOI: 10.3389/fpls.2016.01419. View

Picasso R, Tagliafico A, Calabrese M, Martinoli C, Pistoia F, Rossi A . Primary and Secondary Breast Lymphoma: Focus on Epidemiology and Imaging Features. Pathol Oncol Res. 2019; 26(3):1483-1488. DOI: 10.1007/s12253-019-00730-0. View

Thanarajasingam G, Bennani-Baiti N, Thompson C . PET-CT in Staging, Response Evaluation, and Surveillance of Lymphoma. Curr Treat Options Oncol. 2016; 17(5):24. DOI: 10.1007/s11864-016-0399-z. View

Nicholson B, Bhatti R, Glassman L . Extranodal Lymphoma of the Breast. Radiol Clin North Am. 2016; 54(4):711-26. DOI: 10.1016/j.rcl.2016.03.005. View

Lee J, Lee S . Radiomics in Oncological PET/CT: Clinical Applications. Nucl Med Mol Imaging. 2018; 52(3):170-189. PMC: 5995782. DOI: 10.1007/s13139-017-0500-y. View

Gong J, Liu J, Li H, Zhu H, Wang T, Hu T . Deep Learning-Based Stage-Wise Risk Stratification for Early Lung Adenocarcinoma in CT Images: A Multi-Center Study. Cancers (Basel). 2021; 13(13). PMC: 8269183. DOI: 10.3390/cancers13133300. View

Eertink J, van de Brug T, Wiegers S, Zwezerijnen G, Pfaehler E, Lugtenburg P . F-FDG PET baseline radiomics features improve the prediction of treatment outcome in diffuse large B-cell lymphoma. Eur J Nucl Med Mol Imaging. 2021; 49(3):932-942. PMC: 8803694. DOI: 10.1007/s00259-021-05480-3. View

McBee M, Awan O, Colucci A, Ghobadi C, Kadom N, Kansagra A . Deep Learning in Radiology. Acad Radiol. 2018; 25(11):1472-1480. DOI: 10.1016/j.acra.2018.02.018. View

10.

Zheng R, Zhang S, Zeng H, Wang S, Sun K, Chen R . Cancer incidence and mortality in China, 2016. J Natl Cancer Cent. 2024; 2(1):1-9. PMC: 11256658. DOI: 10.1016/j.jncc.2022.02.002. View

11.

Lam N, Sun H, Song L, Yang D, Zhi S, Ren G . Development and validation of bone-suppressed deep learning classification of COVID-19 presentation in chest radiographs. Quant Imaging Med Surg. 2022; 12(7):3917-3931. PMC: 9246721. DOI: 10.21037/qims-21-791. View

12.

Surov A, Holzhausen H, Wienke A, Schmidt J, Thomssen C, Arnold D . Primary and secondary breast lymphoma: prevalence, clinical signs and radiological features. Br J Radiol. 2012; 85(1014):e195-205. PMC: 3474114. DOI: 10.1259/bjr/78413721. View

13.

Matasar M, Zelenetz A . Overview of lymphoma diagnosis and management. Radiol Clin North Am. 2008; 46(2):175-98, vii. DOI: 10.1016/j.rcl.2008.03.005. View

14.

Sung H, Ferlay J, Siegel R, Laversanne M, Soerjomataram I, Jemal A . Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021; 71(3):209-249. DOI: 10.3322/caac.21660. View

15.

Eertink J, Burggraaff C, Heymans M, Duhrsen U, Huttmann A, Schmitz C . Optimal timing and criteria of interim PET in DLBCL: a comparative study of 1692 patients. Blood Adv. 2021; 5(9):2375-2384. PMC: 8114547. DOI: 10.1182/bloodadvances.2021004467. View

16.

Liu Q, Sun D, Li N, Kim J, Feng D, Huang G . Predicting EGFR mutation subtypes in lung adenocarcinoma using F-FDG PET/CT radiomic features. Transl Lung Cancer Res. 2020; 9(3):549-562. PMC: 7354146. DOI: 10.21037/tlcr.2020.04.17. View

17.

Le N, Kha Q, Nguyen V, Chen Y, Cheng S, Chen C . Machine Learning-Based Radiomics Signatures for EGFR and KRAS Mutations Prediction in Non-Small-Cell Lung Cancer. Int J Mol Sci. 2021; 22(17). PMC: 8431041. DOI: 10.3390/ijms22179254. View

18.

Gurney K, Cartwright R . Increasing incidence and descriptive epidemiology of extranodal non-Hodgkin lymphoma in parts of England and Wales. Hematol J. 2002; 3(2):95-104. DOI: 10.1038/sj.thj.6200154. View

19.

Ou X, Wang J, Zhou R, Zhu S, Pang F, Zhou Y . Ability of F-FDG PET/CT Radiomic Features to Distinguish Breast Carcinoma from Breast Lymphoma. Contrast Media Mol Imaging. 2019; 2019:4507694. PMC: 6410462. DOI: 10.1155/2019/4507694. View

20.

Sousaris N, Barr R . Sonoelastography of Breast Lymphoma. Ultrasound Q. 2016; 32(3):208-11. DOI: 10.1097/RUQ.0000000000000213. View