Abstract
Machine learning can be used to create rapid and accurate segmentations of tumors on images to aid in a patient’s overall clinical assessment.
A 68-year-old woman with a history of hepatocellular carcinoma underwent conventional transarterial chemoembolization. Manual tumor segmentation on images, which can be used to assess disease progression, is time consuming and may suffer from interobserver reliability issues. The authors present a how-to guide to develop machine learning algorithms for fully automatic segmentation of hepatocellular carcinoma and other tumors for lesion tracking over time.
© RSNA, 2022
references
- one. . Guide for diagnosis and treatment of hepatocellular carcinoma . World J Hepatol 2015 ; 7 (12): 1632–1651. Crossref, MedlineGoogle Scholar
- two. . EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma . J Hepatol 2018 ; 69 (1): 182–23. [Published correction appears in J Hepatol 2019;70(4):817.]. Crossref, MedlineGoogle Scholar
- 3. . Modified RECIST (mRECIST) assessment for hepatocellular carcinoma . Semin Liver Dis 2010 ; 30 (1): 52–60. Crossref, MedlineGoogle Scholar
- Four. . Lack of response after initial chemoembolization for hepatocellular carcinoma: does it predict failure of subsequent treatment? Radiology 2012 ; 265(1):115–123. linkGoogle Scholar
- 5. . Use of a semi-automated cardiac segmentation tool improves reproducibility and speed of segmentation of contaminated right heart magnetic resonance angiography . Int J Cardiovasc Imaging 2016 ; 32 (8): 1273–1279. Crossref, MedlineGoogle Scholar
- 6. . Accuracy and reproducibility of manual and semiautomated quantification of MS lesions by MRI . J Magn Reson Imaging 2003 ; 17 (3): 300–308. Crossref, MedlineGoogle Scholar
- 7. . Interobserver agreement of semi-automated and manual measurements of functional MRI metrics of treatment response in hepatocellular carcinoma . Eur J Radiol 2014 ; 83 (3): 487–496. Crossref, MedlineGoogle Scholar
- 8. . Automated segmentation of MR images of brain tumors . Radiology 2001 ; 218 (2): 586–591. linkGoogle Scholar
- 9. . Comparison between manual and semi-automatic segmentation of nasal cavity and paranasal sinuses from CT images. . Annu Int Conf IEEE Eng Med Biol Soc 2007 ; 2007 (5505): 5508. Google Scholar
- 10. . The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) . IEEE Trans Med Imaging 2015 ; 34 (10): 1993–2024. Crossref, MedlineGoogle Scholar
- eleven. . Three-Plane-assembled Deep Learning Segmentation of Gliomas . Radiol Artif Intell 2020 ; 2 (2): e190011. linkGoogle Scholar
- 12. . CT-based True- and False-Lumen Segmentation in Type B Aortic Dissection Using Machine Learning . Radiol Cardiothoracic Imaging 2020 ; 2(3):e190179. linkGoogle Scholar
- 13. . Deep Learning for Automated Segmentation of Liver Lesions at CT in Patients with Colorectal Cancer Liver Metastases . Radiol Artif Intell 2019 ; 1(2):180014. linkGoogle Scholar
- 14. . A New Approach for Automated Image Segmentation of Organs at Risk in Cervical Cancer . Radiol Imaging Cancer 2020 ; 2 (2): e204010. linkGoogle Scholar
- fifteen. . 3D slicer .
In: 2004 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro (IEEE Cat No. 04EX821) ,Arlington, VA ,April 18, 2004 . Piscataway, NJ: IEEE, 2004 ; 632–635. Google Scholar - 16. . OsiriX: an open-source software for navigating in multidimensional DICOM images . J Digit Imaging 2004 ; 17 (3): 205–216. Crossref, MedlineGoogle Scholar
- 17. . PROSTATEx Challenges for computerized classification of prostate lesions from multiparametric magnetic resonance images . J Med Imaging (Bellingham) 2018 ; 5(4):044501. MedlineGoogle Scholar
- 18. . An Effective Interactive Medical Image Segmentation Method using Fast GrowCut .
In: International Conference of Medical Imaging Computing and Computer Assisted Intervention , 2014 . Google Scholar - 19. . 3D Slicer Editor . https://www.slicer.org/wiki/Documentation/4.10/Modules/Editor. Published 2019. Accessed July 26, 2021. Google Scholar
- twenty. . The State of the Octoverse: machine learning. Github Blog Website . https://github.blog/2019-01-24-the-state-of-the-octoverse-machine-learning/. Accessed July 27, 2021. Google Scholar
- twenty-one. . Anaconda Inc. Anaconda Documentation Web site . https://docs.anaconda.com/. Published 2020. Accessed July 27, 2021. Google Scholar
- 22. . Jupyter notebook-based tools for building structured datasets from the Sequence Read Archive . F1000 Res 2020 ; 9:376. Crossref, MedlineGoogle Scholar
- 23. . Why Jupyter is data scientists’ computational notebook of choice . Nature 2018 ; 563 (7729): 145–146. Crossref, MedlineGoogle Scholar
- 24. . Scientific Notebook Software: Applications for Academic Radiology . Curr Trouble Diagnostic Radiol 2018 ; 47 (6): 368–377. Crossref, MedlineGoogle Scholar
- 25. . Teaching Radiology Physics Interactively with Scientific Notebook Software . Acad Radiol 2018 ; 25 (6): 801–810. Crossref, MedlineGoogle Scholar
- 26. . Machine Learning Classification of Cerebral Aneurysm Rupture Status with Morphologic Variables and Hemodynamic Parameters . Radiol Artif Intell 2020 ; 2 (1): e190077. linkGoogle Scholar
- 27. . Measures of the Amount of Ecologic Association Between Species . ecology [1945 ; 26 (3): 297–302. crossrefGoogle Scholar
- 28. . The Regression Analysis of Binary Sequences . JR Stat Soc Ser B Methodol 1958 ; 20 (2): 215–242. Google Scholar
- 29. . Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: concepts, tools, and techniques to build intelligent systems . 2nd ed. Sebastopol, Calif: O’Reilly Media, 2019 . Google Scholar
- 30. . Performance of a Deep-Learning Neural Network Model in Assessing Skeletal Maturity on Pediatric Hand Radiographs . Radiology 2018 ; 287 (1): 313–322. linkGoogle Scholar
- 31. . Radiomic versus Convolutional Neural Networks Analysis for Classification of Contrast-enhancing Lesions at Multiparametric Breast MRI . Radiology 2019 ; 290 (2): 290–297. linkGoogle Scholar
- 32. . Detecting Large Vessel Occlusion at Multiphase CT Angiography by Using a Deep Convolutional Neural Network . Radiology 2020 ; 297 (3): 640–649. linkGoogle Scholar
- 33. . Deep Convolutional Neural Network-based Software Improves Radiologist Detection of Malignant Lung Nodules on Chest Radiographs . Radiology 2020 ; 294 (1): 199–209. linkGoogle Scholar
- 3. 4. . The Liver Tumor Segmentation Benchmark (LiTS) . ArXiv 1901.04056 [preprint]. https://arxiv.org/abs/1901.04056. Posted January 13, 2019. Accessed July 27, 2021. Google Scholar
- 35. . Deep learning and artificial intelligence in radiology: Current applications and future directions . PLoS Med 2018 ; 15(11):e1002707. Crossref, MedlineGoogle Scholar
- 36. . Federated learning improves site performance in multicenter deep learning without data sharing . J Am Med Inform Assoc 2021 ; 28 (6): 1259–1264. Crossref, MedlineGoogle Scholar
- 37. . Artificial Intelligence Algorithm Improves Radiologist Performance in Skeletal Age Assessment: A Prospective Multicenter Randomized Controlled Trial . Radiology 2021 ; 301 (3): 692–699. linkGoogle Scholar
Article History
Received: Sep 24 2021
Revision requested: Nov 3 2021
Revision received: Feb 3 2022
Accepted: Feb 8 2022
Published online: May 10 2022