Activation maps; Bone scintigraphy; Deep learning; Metastatic bone disease; Radiology, Nuclear Medicine and imaging; Oncology; General Medicine; Radiological and Ultrasound Technology
Abstract :
[en] ("[en] PURPOSE: Metastatic bone disease (MBD) is the most common form of metastases, most frequently deriving from prostate cancer. MBD is screened with bone scintigraphy (BS), which have high sensitivity but low specificity for the diagnosis of MBD, often requiring further investigations. Deep learning (DL) - a machine learning technique designed to mimic human neuronal interactions- has shown promise in the field of medical imaging analysis for different purposes, including segmentation and classification of lesions. In this study, we aim to develop a DL algorithm that can classify areas of increased uptake on bone scintigraphy scans.
METHODS: We collected 2365 BS from three European medical centres. The model was trained and validated on 1203 and 164 BS scans respectively. Furthermore we evaluated its performance on an external testing set composed of 998 BS scans. We further aimed to enhance the explainability of our developed algorithm, using activation maps. We compared the performance of our algorithm to that of 6 nuclear medicine physicians.
RESULTS: The developed DL based algorithm is able to detect MBD on BSs, with high specificity and sensitivity (0.80 and 0.82 respectively on the external test set), in a shorter time compared to the nuclear medicine physicians (2.5 min for AI and 30 min for nuclear medicine physicians to classify 134 BSs). Further prospective validation is required before the algorithm can be used in the clinic.","[en] ","")
Disciplines :
Radiology, nuclear medicine & imaging
Author, co-author :
Ibrahim, Abdalla Khalil ; Université de Liège - ULiège > GIGA ; The D-Lab, Department of Precision Medicine, GROW- School for Oncology and Developmental Biology, Maastricht University, Maastricht, The Netherlands ; Department of Radiology and Nuclear Medicine, Columbia University Irving Medical Center, New York, United States ; Department of Nuclear Medicine and Comprehensive diagnostic centre Aachen (CDCA), University Hospital RWTH Aachen University, Aachen, Germany
Vaidyanathan, Akshayaa ; Université de Liège - ULiège > Département de pharmacie ; The D-Lab, Department of Precision Medicine, GROW- School for Oncology and Developmental Biology, Maastricht University, Maastricht, The Netherlands. akshayaa.vaidyanathan@radiomics.bio ; Radiomics (Oncoradiomics SA), Liege, Belgium. akshayaa.vaidyanathan@radiomics.bio
Primakov, Sergey; The D-Lab, Department of Precision Medicine, GROW- School for Oncology and Developmental Biology, Maastricht University, Maastricht, The Netherlands ; Department of Radiology and Nuclear Medicine, Columbia University Irving Medical Center, New York, United States
Refaee, Turkey; The D-Lab, Department of Precision Medicine, GROW- School for Oncology and Developmental Biology, Maastricht University, Maastricht, The Netherlands ; Department of Diagnostic Radiology, Faculty of Applied Medical Sciences, Jazan University, Jazan, Saudi Arabia
Lovinfosse, Pierre ; Centre Hospitalier Universitaire de Liège - CHU > > Service médical de médecine nucléaire et imagerie onco
Jadoul, Alexandre ; Centre Hospitalier Universitaire de Liège - CHU > > Service médical de médecine nucléaire et imagerie onco
Derwael, Céline ; Centre Hospitalier Universitaire de Liège - CHU > > Service médical de médecine nucléaire et imagerie onco
Hertel, Fabian; Department of Nuclear Medicine and Comprehensive diagnostic centre Aachen (CDCA), University Hospital RWTH Aachen University, Aachen, Germany
Woodruff, Henry C; The D-Lab, Department of Precision Medicine, GROW- School for Oncology and Developmental Biology, Maastricht University, Maastricht, The Netherlands ; Department of Radiology and Nuclear Medicine, Columbia University Irving Medical Center, New York, United States
Zacho, Helle D; Department of Nuclear Medicine, Clinical Cancer Research Centre, Aalborg University Hospital, Aalborg, Denmark ; Department of Clinical Medicine, Aalborg University, Aalborg, Denmark
Hanin, François-Xavier; Department of Nuclear Medicine, Universite´CatholiqueUniversite´Catholique de Louvain, CHU-UCL-Namur, Ottignies-Louvain-la-Neuve, Belgium
Lambin, Philippe ; Université de Liège - ULiège > Département des sciences cliniques ; The D-Lab, Department of Precision Medicine, GROW- School for Oncology and Developmental Biology, Maastricht University, Maastricht, The Netherlands ; Department of Radiology and Nuclear Medicine, Columbia University Irving Medical Center, New York, United States
Mottaghy, Felix M; Department of Radiology and Nuclear Medicine, Columbia University Irving Medical Center, New York, United States ; Department of Nuclear Medicine and Comprehensive diagnostic centre Aachen (CDCA), University Hospital RWTH Aachen University, Aachen, Germany
Hustinx, Roland ; Centre Hospitalier Universitaire de Liège - CHU > > Service médical de médecine nucléaire et imagerie onco
ERC - European Research Council [BE] EU - European Union [BE] Interreg EMR - Interreg Euregio Meuse-Rhine [NL]
Funding text :
Authors acknowledge financial support from ERC advanced grant (ERCADG-2015 n° 694812 - Hypoximmuno), ERC-2020-PoC: 957565-AUTO.DISTINCT, Authors also acknowledge financial support from the European Union’s Horizon 2020 research and innovation programme under grant agreement: ImmunoSABR n° 733008, MSCA-ITN-PREDICT n° 766276, CHAIMELEON n° 952172, EuCanImage n° 952103,Interreg V-A Euregio Meuse-Rhine (EURADIOMICS n° EMR4) and Maastricht-Liege Imaging Valley grant, project no. “DEEP-NUCLE”.
Coleman RE. Clinical features of metastatic bone disease and risk of skeletal morbidity. Clin cancer Res an Off J Am Assoc Cancer Res United States. 2006;12:6243s–9. DOI: 10.1158/1078-0432.CCR-06-0931
Migliorini F, Maffulli N, Trivellas A, Eschweiler J, Tingart M, Driessen A. Bone metastases: a comprehensive review of the literature. Mol Biol Rep [Internet]. Department of Orthopaedics, University Clinic Aachen, RWTH Aachen University Clinic, Pauwelsstraße 30, 52074, Aachen, Germany. migliorini.md@gmail.com.; 2020;47:6337–45. Available from: http://europepmc.org/abstract/MED/32749632
Huang J-F, Shen J, Li X, Rengan R, Silvestris N, Wang M et al. Incidence of patients with bone metastases at diagnosis of solid tumors in adults: a large population-based study. Ann Transl Med [Internet]. AME Publishing Company; 2020;8:482. Available from: https://pubmed.ncbi.nlm.nih.gov/32395526
Coleman RE. Metastatic bone disease: clinical features, pathophysiology and treatment strategies. Cancer Treat Rev Netherlands. 2001;27:165–76. DOI: 10.1053/ctrv.2000.0210
Macedo F, Ladeira K, Pinho F, Saraiva N, Bonito N, Pinto L, et al. Bone metastases: an overview. Oncol Rev. 2017;11:321.
Ryan PJ, Fogelman I. Bone scintigraphy in metabolic bone disease. Semin Nucl Med United States. 1997;27:291–305. DOI: 10.1016/S0001-2998(97)80030-X
Ziessman HA, O’Malley JP, Thrall JHBT-NM, Fourth E, editors., editors. Chapter 7 - Skeletal Scintigraphy. Philadelphia: W.B. Saunders; 2014. p. 98–130. Available from: https://www.sciencedirect.com/science/article/pii/B9780323082990000079
Van den Wyngaert T, Strobel K, Kampen WU, Kuwert T, van der Bruggen W, Mohan HK, et al. The EANM practice guidelines for bone scintigraphy. Eur J Nucl Med Mol Imaging. 2016;43:1723–38. DOI: 10.1007/s00259-016-3415-4
LeCun Y, Bengio Y, Hinton G. Deep learning. Nature [Internet]. 2015;521:436–44. Available from: https://doi.org/10.1038/nature14539
McCulloch WS, Pitts W. A logical calculus of the ideas immanent in nervous activity. Bull Math Biophys [Internet]. 1943;5:115–33. Available from: https://doi.org/10.1007/BF02478259
Deng L. A tutorial survey of architectures, algorithms, and applications for deep learning. APSIPA Trans Signal Inf Process [Internet]. 2014/01/22. Cambridge University Press; 2014;3:e2. Available from: https://www.cambridge.org/core/article/tutorial-survey-of-architectures-algorithms-and-applications-for-deep-learning/023B6ADF962FA37F8EC684B209E3DFAE
Aslam YNS. A Review of Deep Learning Approaches for Image Analysis. Int Conf Smart Syst Inven Technol. 2019;2019:709–14.
Janocha K, Czarnecki WM. On loss functions for deep neural networks in classification. Schedae Informaticae. 2016;25:49–59.
Cheng D.-C, Hsieh T.-C, Yen K.-Y, Kao C.-H. Lesion-Based Bone Metastasis Detection in Chest Bone Scintigraphy Images of Prostate Cancer Patients Using Pre-Train, Negative Mining, and Deep Learning. Diagnostics. 2021;11:518. 10.3390/diagnostics11030518.
Papandrianos, N.; Papageorgiou, E.; Anagnostis, A.; Papageorgiou, K. Efficient Bone Metastasis Diagnosis in Bone Scintigraphy Using a Fast Convolutional Neural Network Architecture. Diagnostics. 2020;10:532. 10.3390/diagnostics10080532.
Aoki Y, Nakayama M, Nomura K, Tomita Y, Nakajima K, Yamashina M, et al. The utility of a deep learning-based algorithm for bone scintigraphy in patient with prostate cancer. Ann Nucl Med Japan. 2020;34:926–31. DOI: 10.1007/s12149-020-01524-0
Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization. Int J Comput Vis [Internet]. 2020;128:336–59. Available from: 10.1007/s11263-019-01228-7
Dubost F, Adams H, Yilmaz P, Bortsova G, van Tulder G, Ikram MA et al. Weakly supervised object detection with 2D and 3D regression neural networks. Med Image Anal [Internet]. 2020;65:101767. Available from: https://www.sciencedirect.com/science/article/pii/S1361841520301316
Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-CAM: visual explanations from deep networks via gradient-based localization. Int J Comput Vis Springer. 2016;128:336–59. DOI: 10.1007/s11263-019-01228-7
World Medical Association. Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA United States. 2013;310:2191–4.
Simonyan, K. and Zisserman, A. (2015) Very Deep Convolutional Networks for Large-Scale Image Recognition. The 3rd International Conference on Learning Representations (ICLR2015). https://arxiv.org/abs/1409.1556.
Shorten C, Khoshgoftaar TM. A survey on Image Data Augmentation for Deep Learning. J Big Data [Internet]. 2019;6:60. Available from: 10.1186/s40537-019-0197-0
Calin O, Activation Functions BT. - Deep Learning Architectures: A Mathematical Approach. In: Calin O, editor. Cham: Springer International Publishing; 2020. p. 21–39. Available from: 10.1007/978-3-030-36721-3_2
Kingma, D. and Ba, J. (2015) Adam: A Method for Stochastic Optimization. Proceedings of the 3rd International Conference on Learning Representations (ICLR 2015). https://arxiv.org/abs/1412.6980.
Mongan J, Moy L, Kahn CE. Checklist for Artificial Intelligence in Medical Imaging (CLAIM): A Guide for Authors and Reviewers. Radiol Artif Intell [Internet]. Radiological Society of North America; 2020;2:e200029. Available from: 10.1148/ryai.2020200029
Cohen JF, Korevaar DA, Altman DG, Bruns DE, Gatsonis CA, Hooft L et al. STARD 2015 guidelines for reporting diagnostic accuracy studies: explanation and elaboration. BMJ Open [Internet]. 2016;6:e012799. Available from: http://bmjopen.bmj.com/content/6/11/e012799.abstract
Liu S, Feng M, Qiao T, Cai H, Xu K, Yu X, et al. Deep learning for the Automatic diagnosis and analysis of bone metastasis on bone scintigrams. Cancer Manag Res. 2022;14:51–65. DOI: 10.2147/CMAR.S340114
Han S, Oh J.S, Lee J.J. Diagnostic performance of deep learning models for detecting bone metastasis on whole-body bone scan in prostate cancer. Eur J Nucl Med Mol Imaging. 2022;49:585–595. 10.1007/s00259-021-05481-2.
Anand A, Morris MJ, Kaboteh R, Båth L, Sadik M, Gjertsson P, et al. Analytic Validation of the automated bone scan index as an imaging biomarker to standardize quantitative changes in bone scans of patients with metastatic prostate Cancer. J Nucl Med. 2016;57:41–5. DOI: 10.2967/jnumed.115.160085
Narasinga Rao MR, Venkatesh Prasad D, Sai Teja V, Zindavali P, Phanindra Reddy M. A Survey on Prevention of Overfitting in Convolution neural networks using machine learning techniques. Int J Eng Technol. 2018;7:177. DOI: 10.14419/ijet.v7i2.32.15399