Meme Kanserinin Histopatolojik Sonuçlarının Öngörülmesinde Stir Sekanslardaki Radyomik Özelliklerin Performansı

Abstract

ÖZ Amaç: Meme kanserinin histopatolojik özellikleri tedavi ve prognozun belirlenmesinde kritik öneme sahiptir. Bu çalışmanın amacı, short tau inversion recovery (STIR) sekanslarındaki radyomik özelliklerin, invazif meme kanserinin histopatolojik sonuçlarının tahmin edilmesindeki performansını araştırmaktır. Gereç ve Yöntemler: İnvazif meme kanseri tanısı alan kadınların tedavi öncesi Manyetik rezonans görüntüleme (MRG) incelemeleri retrospektif olarak değerlendirildi. Histolojik grade, östrojen reseptörü (ER), progesteron reseptörü (PR), Her-2, Ki-67 ekspresyonları ve moleküler alt tipler not edildi. Lezyonlar, 3D Slicer programında STIR sekanslarından manuel olarak segmente edildi ve ilgilenilen hacim elde edildi. Makine öğrenimi (MÖ) analizi Python 3.11, Pycaret kütüphane programı kullanılarak yapıldı. MÖ algoritmalarının başarısı eğri altındaki alan (AUC), doğruluk, geri hatırlama, kesinlik ve F1 puanı ile değerlendirildi. Bulgular: Çalışmaya yaş ortalaması 50,72±46 (dağılım 28-82) yıl olan toplam 197 hasta dahil edildi. Ortalama lezyon boyutu 23,71±14,86 (5-120) mm idi. En iyi performans ER+/- ayrımında ve luminal A+B tahmininde elde edildi. CatBoost sınıflandırıcısı en başarılı MÖ algoritması olup, AUC, doğruluk, geri hatırlama, kesinlik ve F1 değerleri ER+/- için sırasıyla 0,75, %80, 0,96, 0,81, 0,88 olarak bulunmuştur. Luminal A+B tümörlerini öngörmede AUC, doğruluk, hatırlama, kesinlik ve F1 değerleri sırasıyla 0,72, %79, 0,79, 0,64, 0,70 olarak belirlendi. Sonuç: STIR sekanslardan elde edilen radyomik özellikler, ER reseptör durumunu ve luminal tümörleri tahmin etme potansiyeline sahiptir.

Keywords

• Meme neoplazmı
• Manyetik rezonans görüntüleme
• Yapay Zeka
• Histoloji
• Makine öğrenimi

Performance of Radiomic Features in Stir Sequences in Predicting Histopathological Outcomes of Breast Cancer

Abstract

ABSTRACT Objective: Histopathological features of breast cancer are critical in determining treatment and prognosis. The aim of this study is to investigate the performance of radiomic features in short tau inversion recovery (STIR) sequences in predicting histopathological outcomes of invasive breast cancer. Methods: Pre-treatment (Magnetic resonance imaging) MRI examinations of women diagnosed with invasive breast cancer were evaluated retrospectively. Histologic grade, estrogen receptor (ER), progesterone receptor (PR), Her-2, Ki-67 expressions, and molecular subtypes were noted. Lesions were manually segmented from STIR sequences using the 3D Slicer software, and volumes of interest was obtained. Machine learning (ML) analysis was performed using Python 3.11, Pycaret library. The performance of ML algorithms was evaluated with area under curve (AUC), accuracy, recall, precision and F1 score. Results: A total of 197 patients with a mean age of 50.72±11 (range 28-82) years were included in the study. The mean lesion size was 23.71±14.86 (5-120) mm. The best performance was obtained in ER+/- discrimination and luminal A+B prediction. The CatBoost classifier was the most successful ML algorithm with AUC, accuracy, recall, precision and F1 values were 0.75, 80%, 0.96, 0.81, 0.88 respectively for ER+/-.In predicting luminal A+B tumors, AUC, accuracy, recall, precision and F1 values were 0.72, 79%, 0.79, 0.64, 0.70, respectively. Conclusion: Radiomic features obtained from STIR sequences have potential utility in predicting ER receptor status and luminal tumors.

Keywords

• Breast neoplasms
• Magnetic resonance imaging
• Artificial Intelligence
• Histology
• Machine learning

References

1. Wilkinson L, Gathani T. Understanding breast cancer as a global health concern. Br J Radiol. 2022;95(1130):20211033.
2. Sarhangi N, Hajjari S, Heydari SF, Ganjizadeh M, Rouhollah F, Hasanzad M. Breast cancer in the era of precision medicine. Mol Biol Rep. 2022;49(10):10023-37.
3. Tarighati E, Keivan H, Mahani H. A review of prognostic and predictive biomarkers in breast cancer. Clin Exp Med. 2023;23(1):1-16.
4. Yin L, Duan JJ, Bian XW, Yu SC. Triple-negative breast cancer molecular subtyping and treatment progress. Breast Cancer Res. 2020;22(1):61.
5. Roy M, Fowler AM, Ulaner GA, Mahajan A. Molecular Classification of Breast Cancer. PET Clin. 2023;18(4):441-58.
6. Turner KM, Yeo SK, Holm TM, Shaughnessy E, Guan JL. Heterogeneity within molecular subtypes of breast cancer. Am J Physiol Cell Physiol. 2021;321(2):C343-54.
7. Wolf DM, Yau C, Wulfkuhle J, Brown-Swigart L, Gallagher RI, Lee PRE, Zhu Z, Magbanua MJ, Sayaman R, O'Grady N, Basu A, Delson A, Coppé JP, Lu R, Braun J; I-SPY2 Investigators; Asare SM, Sit L, Matthews JB, Perlmutter J, Hylton N, Liu MC, Pohlmann P, Symmans WF, Rugo HS, Isaacs C, DeMichele AM, Yee D, Berry DA, Pusztai L, Petricoin EF, Hirst GL, Esserman LJ, van 't Veer LJ. Redefining breast cancer subtypes to guide treatment prioritization and maximize response: Predictive biomarkers across 10 cancer therapies. Cancer Cell. 2022 ;40(6):609-623.e6.
8. Tsang JYS, Tse GM. Molecular Classification of Breast Cancer. Adv Anat Pathol. 2020;27(1):27-35.

9. van Timmeren JE, Cester D, Tanadini-Lang S, Alkadhi H, Baessler B. Radiomics in medical imaging-"how-to" guide and critical reflection. Insights Imaging. 2020;11(1):91.
10. Conti A, Duggento A, Indovina I, Guerrisi M, Toschi N. Radiomics in breast cancer classification and prediction. Semin Cancer Biol. 2021;72:238-50.
11. Lin JY, Ye JY, Chen JG, Lin ST, Lin S, Cai SQ. Prediction of Receptor Status in Radiomics: Recent Advances in Breast Cancer Research. Acad Radiol. 2024;31(7):3004-14.
12. Saha A, Harowicz MR, Grimm LJ, Kim CE, Ghate SV, Walsh R, Mazurowski MA. A machine learning approach to radiogenomics of breast cancer: a study of 922 subjects and 529 DCE-MRI features. Br J Cancer. 2018;119(4):508-16.
13. Chang RF, Chen HH, Chang YC, Huang CS, Chen JH, Lo CM. Quantification of breast tumor heterogeneity for ER status, HER2 status, and TN molecular subtype evaluation on DCE-MRI. Magn Reson Imaging. 2016;34(6):809-19.
14. Demircioglu A, Grueneisen J, Ingenwerth M, Hoffmann O, Pinker-Domenig K, Morris E, Haubold J, Forsting M, Nensa F, Umutlu L. A rapid volume of interest-based approach of radiomics analysis of breast MRI for tumor decoding and phenotyping of breast cancer. PLoS One. 2020;15(6):e0234871.
15. Li C, Song L, Yin J. Intratumoral and Peritumoral Radiomics Based on Functional Parametric Maps from Breast DCE-MRI for Prediction of HER-2 and Ki-67 Status. J Magn Reson Imaging. 202;54(3):703-14.
16. Bian X, Du S, Yue Z, Gao S, Zhao R, Huang G, Guo L, Peng C, Zhang L. Potential Antihuman Epidermal Growth Factor Receptor 2 Target Therapy Beneficiaries: The Role of MRI-Based Radiomics in Distinguishing Human Epidermal Growth Factor Receptor 2-Low Status of Breast Cancer. J Magn Reson Imaging. 2023;58(5):1603-14.
17. Szep M, Pintican R, Boca B, Perja A, Duma M, Feier D, Epure F, Fetica B, Eniu D, Roman A, Dudea SM, Chiorean A. Whole-Tumor ADC Texture Analysis Is Able to Predict Breast Cancer Receptor Status. Diagnostics (Basel). 2023;13(8):1414.
18. Ma W, Ji Y, Qi L, Guo X, Jian X, Liu P. Breast cancer Ki67 expression prediction by DCE-MRI radiomics features. Clin Radiol. 2018;73(10):909.e1-909.e5.
19. Sheng W, Xia S, Wang Y, Yan L, Ke S, Mellisa E, Gong F, Zheng Y, Tang T. Invasive ductal breast cancer molecular subtype prediction by MRI radiomic and clinical features based on machine learning. Front Oncol. 2022;12:964605.