Abstract
A computer-aided detection (CADe) system identifies features on a medical image and brings them to the radiologist’s attention. However, the spread of CADe systems in clinical sites is limited. Our aim is to establish a generalized framework for developing various CADe systems. The framework would provide opportunities for many clinicians without the CADe system expertise to develop and use such systems that meet their specific needs. We proposed a pilot version of the framework, including four pretrained algorithms: preprocessing, extraction of a candidate area, candidate detection, and candidate classification. We experimentally confirmed that two different types of CADe system were developed successfully through the generalized framework using two different datasets. We also proposed two feature generation methods to improve the generalized framework. One is the use of multiple deep convolutional autoencoders (DCAEs) trained with a normal dataset. The other is by transfer using a deep convolutional neural network (DCNN) pretrained with an anatomical landmark dataset from whole-body CT. An evaluation of these methods using head MR angiography datasets shows that the DCAEs could extract useful features for lesion classification.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
van Ginneken B, Schaefer-Prokop CM, Prokop M. Computer-aided diagnosis: how to move from the laboratory to the clinic. Radiology. 2011;261:719–32.
Chan HP, Lo SCB, Sahiner B, Lam KL, Helvie MA. Computer-aided detection of mammographic microcalcifications: pattern recognition with an artificial neural network. Med Phys. 1995;22:1555–67.
Gokturk SB, Tomasi C, Acar B, Beaulieu CF, Paik DS, Jeffrey RB Jr, Yee J, Napel S. A statistical 3-D pattern processing method for computer-aided detection of polyps in CT colonography. IEEE Trans Med Imaging. 2001;20:1251–60.
Nemoto M, Hayashi N, Hanaoka S, Nomura Y, Miki S, Yoshikawa T. Feasibility study of a generalized framework for developing computer-aided detection systems—a new paradigm. J Digit Imaging. 2017;30:629–39.
Nemoto M, Ushifusa K, Kimura Y, Hayashi N. Pilot study to generate image features by deep autoencoder for computer-aided detection systems. Int Forum on Medical Imaging in Asia (IFMIA), Proceedings SPIE. 2019;11050:10501J.
Ushifusa K, Nemoto M, Kimura Y, Nagaoka T, Yamada T, Tanaka A, Hayashi N. A generalized image feature generation based on unsupervised deep learning with small scale normal dataset. Int J Comput Assist Radiol Surg. 2020;15(Suppl 1):S210–2.
Nemoto M. A pilot study for transferring deep convolutional neural network pre-trained by local anatomical structures to computer-aided detection. Int J Comput Assist Radiol Surg. 2020;15(Suppl 1):S110–2.
Rohr K. Landmark-based image analysis: using geometric and intensity models. Amsterdam: Springer Netherlands; 2001.
Hanaoka S, Shimizu A, Nemoto M, et al. Automatic detection of over 100 anatomical landmarks in medical CT images – a framework with independent detectors and combinatorial optimization. Med Image Anal. 2016;35:192–214.
Viola P, Jones M. Rapid object detection using a boosted cascade of simple features. Proc IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2001). 2001;1:I511–I518.
Iba W, Langley P. Induction of one-level decision trees. Proc International Conference on Machine Learning (ICML 1992). 1992:233–240.
Sun Y, Kamel MS, Wong AK, Wang Y. Cost-sensitive boosting for classification of imbalanced data. Pattern Recogn. 2007;40:3358–78.
Tu Z, Zhou XS, Barbu A, Bogoni L, Comaniciu D. Probabilistic 3D polyp detection in CT images – The role of sample alignment. Proc IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR). 2006;
Boureau Y, Ponce J, LeCun Y. A theoretical analysis of feature pooling in visual recognition. Proc International Conference on Machine Learning (ICML). 2010:111–118.
van Ginneken B, Armato SG, de Hoop B, et al. Comparing and combining algorithms for computer-aided detection of pulmonary nodules in computed tomography scans – The ANODE09 study. Med Image Anal. 2010;14:707–22.
Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60–88.
Masci J, Meier U, Cireşan D, Schmidhuber J. Stacked convolutional auto-encoders for hierarchical feature extraction. Artificial Neural Networks and Machine Learning – ICANN. 2011;2011:52–9.
Roth HR, Lu L, Seff A, Cherry KM. A new 2.5D representation for lymph node detection using random sets of deep convolutional neural network observations. Medical Image Computing and Computer-Assisted Intervention – MICCAI. 2014;2014:520–7.
Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. The International Conference on Learning Representations – ICLR, 2015.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Nemoto, M., Hayashi, N. (2022). AI: A Machine-Learning-Based Framework for Developing Various Computer-Aided Detection Systems with Generated Image Features. In: Hashizume, M. (eds) Multidisciplinary Computational Anatomy. Springer, Singapore. https://doi.org/10.1007/978-981-16-4325-5_49
Download citation
DOI: https://doi.org/10.1007/978-981-16-4325-5_49
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-4324-8
Online ISBN: 978-981-16-4325-5
eBook Packages: MedicineMedicine (R0)