Comparative analysis of radiomics and deep-learning algorithms for survival prediction in hepatocellular carcinoma.


Journal

Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288

Informations de publication

Date de publication:
05 Jan 2024
Historique:
received: 16 07 2023
accepted: 20 12 2023
medline: 6 1 2024
pubmed: 6 1 2024
entrez: 5 1 2024
Statut: epublish

Résumé

To examine the comparative robustness of computed tomography (CT)-based conventional radiomics and deep-learning convolutional neural networks (CNN) to predict overall survival (OS) in HCC patients. Retrospectively, 114 HCC patients with pretherapeutic CT of the liver were randomized into a development (n = 85) and a validation (n = 29) cohort, including patients of all tumor stages and several applied therapies. In addition to clinical parameters, image annotations of the liver parenchyma and of tumor findings on CT were available. Cox-regression based on radiomics features and CNN models were established and combined with clinical parameters to predict OS. Model performance was assessed using the concordance index (C-index). Log-rank tests were used to test model-based patient stratification into high/low-risk groups. The clinical Cox-regression model achieved the best validation performance for OS (C-index [95% confidence interval (CI)] 0.74 [0.57-0.86]) with a significant difference between the risk groups (p = 0.03). In image analysis, the CNN models (lowest C-index [CI] 0.63 [0.39-0.83]; highest C-index [CI] 0.71 [0.49-0.88]) were superior to the corresponding radiomics models (lowest C-index [CI] 0.51 [0.30-0.73]; highest C-index [CI] 0.66 [0.48-0.79]). A significant risk stratification was not possible (p > 0.05). Under clinical conditions, CNN-algorithms demonstrate superior prognostic potential to predict OS in HCC patients compared to conventional radiomics approaches and could therefore provide important information in the clinical setting, especially when clinical data is limited.

Identifiants

pubmed: 38182664
doi: 10.1038/s41598-023-50451-3
pii: 10.1038/s41598-023-50451-3
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Pagination

590

Informations de copyright

© 2024. The Author(s).

Références

Altekruse, S. F., Devesa, S. S., Dickie, L. A., McGlynn, K. A. & Kleiner, D. E. Histological classification of liver and intrahepatic bile duct cancers in SEER registries. J. Regist. Manag. 38, 201–205 (2011).
Jemal, A. et al. Annual report to the nation on the status of cancer, 1975–2014 featuring survival. J. Natl. Cancer Inst. 109, djx030 (2017).
pubmed: 28376154 pmcid: 5409140 doi: 10.1093/jnci/djx030
Reig, M. et al. BCLC strategy for prognosis prediction and treatment recommendation: The 2022 update. J. Hepatol. 76, 681–693 (2022).
pubmed: 34801630 doi: 10.1016/j.jhep.2021.11.018
Vitale, A. et al. Survival benefit of liver resection for patients with hepatocellular carcinoma across different Barcelona clinic liver cancer stages: A multicentre study. J. Hepatol. 62, 617–624 (2015).
pubmed: 25450706 doi: 10.1016/j.jhep.2014.10.037
Huitzil-Melendez, F.-D. et al. Advanced hepatocellular carcinoma: Which staging systems best predict prognosis?. J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol. 28, 2889–2895 (2010).
doi: 10.1200/JCO.2009.25.9895
Wakabayashi, T. et al. Radiomics in hepatocellular carcinoma: A quantitative review. Hepatol. Int. 13, 546–559 (2019).
pubmed: 31473947 doi: 10.1007/s12072-019-09973-0
Ahn, J. C., Qureshi, T. A., Singal, A. G., Li, D. & Yang, J.-D. Deep learning in hepatocellular carcinoma: Current status and future perspectives. World J. Hepatol. 13, 2039–2051 (2021).
pubmed: 35070007 pmcid: 8727204 doi: 10.4254/wjh.v13.i12.2039
Denzler, S. et al. Impact of CT convolution kernel on robustness of radiomic features for different lung diseases and tissue types. Br. J. Radiol. 94, 20200947 (2021).
pubmed: 33544646 pmcid: 8010534 doi: 10.1259/bjr.20200947
Leitlinienprogramm Onkologie (Deutsche Krebsgesellschaft, Deutsche Krebshilfe, AWMF): Diagnostik und Therapie des Hepatozellulären Karzinoms und biliärer Karzinome, Langversion 4.0, 2023, AWMF-Registernummer: 032–053OL. https://www.leitlinienprogramm-onkologie.de/leitlinien/hcc-und-biliaere-karzinome .
Wiesner, R. H. et al. MELD and PELD: Application of survival models to liver allocation. Liver Transplant. Off. Publ. Am. Assoc. Study Liver Dis. Int. Liver Transplant. Soc. 7, 567–580 (2001).
Malinchoc, M. et al. A model to predict poor survival in patients undergoing transjugular intrahepatic portosystemic shunts. Hepatology 31, 864–871 (2000).
pubmed: 10733541 doi: 10.1053/he.2000.5852
Johnson, P. J. et al. Assessment of liver function in patients with hepatocellular carcinoma: A new evidence-based approach—The ALBI grade. J. Clin. Oncol. 33, 550–558 (2015).
pubmed: 25512453 doi: 10.1200/JCO.2014.57.9151
Pugh, R. N., Murray-Lyon, I. M., Dawson, J. L., Pietroni, M. C. & Williams, R. Transection of the oesophagus for bleeding oesophageal varices. Br. J. Surg. 60, 646–649 (1973).
pubmed: 4541913 doi: 10.1002/bjs.1800600817
Nebelung, H. et al. Radioembolization versus portal vein embolization for contralateral liver lobe hypertrophy: Effect of cirrhosis. Abdom. Radiol. 46, 4046–4055 (2021).
doi: 10.1007/s00261-021-03048-1
Fedorov, A. et al. 3D slicer as an image computing platform for the quantitative imaging network. Magn. Reson. Imaging 30, 1323–1341 (2012).
pubmed: 22770690 pmcid: 3466397 doi: 10.1016/j.mri.2012.05.001
Han, J. K. et al. Factors influencing vascular and hepatic enhancement at CT: Experimental study on injection protocol using a canine model. J. Comput. Assist. Tomogr. 24, 400 (2000).
pubmed: 10864075 doi: 10.1097/00004728-200005000-00008
Fiz, F. et al. Contrast administration impacts CT-based radiomics of colorectal liver metastases and non-tumoral liver parenchyma revealing the “radiological” tumour microenvironment. Diagnostics 11, 1162 (2021).
pubmed: 34202253 pmcid: 8305553 doi: 10.3390/diagnostics11071162
Zwanenburg, A., Leger, S., Vallières, M. & Löck, S. Image biomarker standardisation initiative. Radiology 295, 328–338 (2020).
pubmed: 32154773 doi: 10.1148/radiol.2020191145
Zwanenburg, A. et al. Assessing robustness of radiomic features by image perturbation. Sci. Rep. 9, 614 (2019).
pubmed: 30679599 pmcid: 6345842 doi: 10.1038/s41598-018-36938-4
Zwanenburg, A. & Löck, S. familiar: End-to-end automated machine learning and model evaluation (2021).
Paszke, A. et al. PyTorch: An imperative style, high-performance deep learning library. In Advances in Neural Information Processing Systems Vol. 32 (eds Paszke, A. et al.) (Curran Associates Inc, 2019).
Hosny, A. et al. Deep learning for lung cancer prognostication: A retrospective multi-cohort radiomics study. PLoS Med. 15, e1002711 (2018).
pubmed: 30500819 pmcid: 6269088 doi: 10.1371/journal.pmed.1002711
Starke, S. et al. 2D and 3D convolutional neural networks for outcome modelling of locally advanced head and neck squamous cell carcinoma. Sci. Rep. 10, 15625 (2020).
pubmed: 32973220 pmcid: 7518264 doi: 10.1038/s41598-020-70542-9
Nie, D. et al. Multi-channel 3D deep feature learning for survival time prediction of brain tumor patients using multi-modal neuroimages. Sci. Rep. 9, 1103 (2019).
pubmed: 30705340 pmcid: 6355868 doi: 10.1038/s41598-018-37387-9
Katzman, J. L. et al. DeepSurv: Personalized treatment recommender system using a Cox proportional hazards deep neural network. BMC Med. Res. Methodol. 18, 24 (2018).
pubmed: 29482517 pmcid: 5828433 doi: 10.1186/s12874-018-0482-1
Efron, B. 10 Nonparametric Confidence Intervals. In The Jackknife, the Bootstrap and Other Resampling Plans (ed. Efron, B.) 75–90 (Society for Industrial and Applied Mathematics, 1982). https://doi.org/10.1137/1.9781611970319.ch10 .
doi: 10.1137/1.9781611970319.ch10
Wei, L. et al. A deep survival interpretable radiomics model of hepatocellular carcinoma patients. Phys. Med. 82, 295–305 (2021).
pubmed: 33714190 pmcid: 8035300 doi: 10.1016/j.ejmp.2021.02.013
Liu, Q.-P., Xu, X., Zhu, F.-P., Zhang, Y.-D. & Liu, X.-S. Prediction of prognostic risk factors in hepatocellular carcinoma with transarterial chemoembolization using multi-modal multi-task deep learning. eClinicalMedicine 23, 100379 (2020).
pubmed: 32548574 pmcid: 7284069 doi: 10.1016/j.eclinm.2020.100379
Wang, H. et al. Development and validation of a deep learning model for survival prognosis of transcatheter arterial chemoembolization in patients with intermediate-stage hepatocellular carcinoma. Eur. J. Radiol. 156, 110527 (2022).
pubmed: 36152524 doi: 10.1016/j.ejrad.2022.110527
Zhang, L. et al. Deep learning predicts overall survival of patients with unresectable hepatocellular carcinoma treated by transarterial chemoembolization plus sorafenib. Front. Oncol. https://doi.org/10.3389/fonc.2020.593292 (2020).
doi: 10.3389/fonc.2020.593292 pubmed: 33928019 pmcid: 7787160
Liu, Y. et al. CT radiomics combined with clinical variables for predicting the overall survival of hepatocellular carcinoma patients after hepatectomy. Transl. Oncol. 26, 101536 (2022).
pubmed: 36115077 pmcid: 9483805 doi: 10.1016/j.tranon.2022.101536
Liu, Q. et al. A radiomics nomogram for the prediction of overall survival in patients with hepatocellular carcinoma after hepatectomy. Cancer Imaging 20, 82 (2020).
pubmed: 33198809 pmcid: 7667801 doi: 10.1186/s40644-020-00360-9
Zheng, B.-H. et al. Radiomics score: A potential prognostic imaging feature for postoperative survival of solitary HCC patients. BMC Cancer 18, 1148 (2018).
pubmed: 30463529 pmcid: 6249916 doi: 10.1186/s12885-018-5024-z
Meng, X.-P. et al. Radiomics analysis on multiphase contrast-enhanced CT: A survival prediction tool in patients with hepatocellular carcinoma undergoing transarterial chemoembolization. Front. Oncol. https://doi.org/10.3389/fonc.2020.01196 (2020).
doi: 10.3389/fonc.2020.01196 pubmed: 33680905 pmcid: 7786369
Bernatz, S. et al. CT-radiomics and clinical risk scores for response and overall survival prognostication in TACE HCC patients. Sci. Rep. 13, 533 (2023).
pubmed: 36631548 pmcid: 9834236 doi: 10.1038/s41598-023-27714-0
Mackin, D. et al. Measuring computed tomography scanner variability of radiomics features. Investig. Radiol. 50, 757–765 (2015).
doi: 10.1097/RLI.0000000000000180
Berenguer, R. et al. Radiomics of CT features may be nonreproducible and redundant: Influence of CT acquisition parameters. Radiology 288, 407–415 (2018).
pubmed: 29688159 doi: 10.1148/radiol.2018172361
Lu, L., Ehmke, R. C., Schwartz, L. H. & Zhao, B. Assessing agreement between radiomic features computed for multiple CT imaging settings. PLoS One 11, e0166550 (2016).
pubmed: 28033372 pmcid: 5199063 doi: 10.1371/journal.pone.0166550
Jin, H. & Kim, J. H. Evaluation of feature robustness against technical parameters in CT radiomics: Verification of phantom study with patient dataset. J. Signal Process. Syst. 92, 277–287 (2020).
doi: 10.1007/s11265-019-01496-z
Ziegelmayer, S. et al. Feature robustness and diagnostic capabilities of convolutional neural networks against radiomics features in computed tomography imaging. Investig. Radiol. 57, 171–177 (2022).
doi: 10.1097/RLI.0000000000000827
Huynh, B. N. et al. Head and neck cancer treatment outcome prediction: A comparison between machine learning with conventional radiomics features and deep learning radiomics. Front. Med. https://doi.org/10.3389/fmed.2023.1217037 (2023).
doi: 10.3389/fmed.2023.1217037
Whitfield, J. B. Gamma glutamyl transferase. Crit. Rev. Clin. Lab. Sci. 38, 263–355 (2001).
pubmed: 11563810 doi: 10.1080/20014091084227
Xu, X.-S. et al. Model based on γ-glutamyltransferase and alkaline phosphatase for hepatocellular carcinoma prognosis. World J. Gastroenterol. WJG 20, 10944–10952 (2014).
pubmed: 25152598 doi: 10.3748/wjg.v20.i31.10944
Yang, Z. et al. Elevation of serum GGT and LDH levels, together with higher BCLC staging are associated with poor overall survival from hepatocellular carcinoma: A retrospective analysis. Discov. Med. 19, 409–418 (2015).
pubmed: 26175398
Ma, W., Wang, H. & Teng, L. Correlation analysis of preoperative serum alpha-fetoprotein (AFP) level and prognosis of hepatocellular carcinoma (HCC) after hepatectomy. World J. Surg. Oncol. 11, 212 (2013).
pubmed: 23981851 pmcid: 3844510 doi: 10.1186/1477-7819-11-212
Tandon, P. & Garcia-Tsao, G. Prognostic indicators in hepatocellular carcinoma: A systematic review of 72 studies. Liver Int. Off. J. Int. Assoc. Study Liver 29, 502–510 (2009).
Lu, X.-Y. et al. Pathobiological features of small hepatocellular carcinoma: Correlation between tumor size and biological behavior. J. Cancer Res. Clin. Oncol. 137, 567–575 (2011).
pubmed: 20508947 doi: 10.1007/s00432-010-0909-5
Wu, G. et al. Importance of tumor size at diagnosis as a prognostic factor for hepatocellular carcinoma survival: A population-based study. Cancer Manag. Res. 10, 4401–4410 (2018).
pubmed: 30349373 pmcid: 6188157 doi: 10.2147/CMAR.S177663
Lau, T. & Ahmad, J. Clinical applications of the model for end-stage liver disease (MELD) in hepatic medicine. Hepatic Med. Evid. Res. 5, 1–10 (2013).
Hagan, M. T. et al. Liver volume in the cirrhotic patient: Does size matter?. Dig. Dis. Sci. 59, 886–891 (2014).
pubmed: 24504591 pmcid: 4565509 doi: 10.1007/s10620-014-3038-1
Kim, S. et al. Radiomics on gadoxetic acid-enhanced magnetic resonance imaging for prediction of postoperative early and late recurrence of single hepatocellular carcinoma. Clin. Cancer Res. Off. J. Am. Assoc. Cancer Res. 25, 3847–3855 (2019).
doi: 10.1158/1078-0432.CCR-18-2861
Teixeira, L. O. et al. Impact of lung segmentation on the diagnosis and explanation of COVID-19 in chest X-ray images. Sensors 21, 7116 (2021).
pubmed: 34770423 pmcid: 8587284 doi: 10.3390/s21217116

Auteurs

Felix Schön (F)

Institute and Polyclinic for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Carl Gustav Carus, TU Dresden, Dresden, Germany. felix.schoen@uniklinikum-dresden.de.

Aaron Kieslich (A)

OncoRay‑National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, TU Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany. aaron.kieslich@oncoray.de.

Heiner Nebelung (H)

Institute and Polyclinic for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Carl Gustav Carus, TU Dresden, Dresden, Germany.

Carina Riediger (C)

Department of Visceral, Thoracic and Vascular Surgery, Faculty of Medicine and University Hospital Carl Gustav Carus, TU Dresden, Dresden, Germany.

Ralf-Thorsten Hoffmann (RT)

Institute and Polyclinic for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Carl Gustav Carus, TU Dresden, Dresden, Germany.

Alex Zwanenburg (A)

OncoRay‑National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, TU Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany.
National Center for Tumor Diseases (NCT/UCC) Dresden, Dresden, Germany.
German Cancer Research Center (DKFZ), Heidelberg, Germany.

Steffen Löck (S)

OncoRay‑National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, TU Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany.

Jens-Peter Kühn (JP)

Institute and Polyclinic for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Carl Gustav Carus, TU Dresden, Dresden, Germany.

Classifications MeSH