Two-stage machine learning models for bowel lesions characterisation using self-propelled capsule dynamics
-
Published:2023-09-08
Issue:20
Volume:111
Page:19387-19410
-
ISSN:0924-090X
-
Container-title:Nonlinear Dynamics
-
language:en
-
Short-container-title:Nonlinear Dyn
Author:
Afebu Kenneth Omokhagbo, Tian Jiyuan, Papatheou Evangelos, Liu YangORCID, Prasad Shyam
Abstract
AbstractTo foster early bowel cancer diagnosis, a non-invasive biomechanical characterisation of bowel lesions is proposed. This method uses the dynamics of a self-propelled capsule and a two-stage machine learning procedure. As the capsule travels and encounters lesions in the bowel, its exhibited dynamics are envisaged to be of biomechanical significance being a highly sensitive nonlinear dynamical system. For this study, measurable capsule dynamics including acceleration and displacement have been analysed for features that may be indicative of biomechanical differences, Young’s modulus in this case. The first stage of the machine learning involves the development of supervised regression networks including multi-layer perceptron (MLP) and support vector regression (SVR), that are capable of predicting Young’s moduli from dynamic signals features. The second stage involves an unsupervised categorisation of the predicted Young’s moduli into clusters of high intra-cluster similarity but low inter-cluster similarity using K-means clustering. Based on the performance metrics including coefficient of determination and normalised mean absolute error, the MLP models showed better performances on the test data compared to the SVR. For situations where both displacement and acceleration were measurable, the displacement-based models outperformed the acceleration-based models. These results thus make capsule displacement and MLP network the first-line choices for the proposed bowel lesion characterisation and early bowel cancer diagnosis.
Funder
Engineering and Physical Sciences Research Council
Publisher
Springer Science and Business Media LLC
Subject
Electrical and Electronic Engineering,Applied Mathematics,Mechanical Engineering,Ocean Engineering,Aerospace Engineering,Control and Systems Engineering
Reference77 articles.
1. Afebu, K.O., Tian, J., Liu, Y., Papatheou, E., Prasad, S.: AI-assisted dynamic tissue evaluation for early bowel cancer diagnosis using a vibrational capsule. IEEE Robot. Autom. Lett. 6, 66 (2023) 2. Akhtar, R., Sherratt, M.J., Cruickshank, J.K., Derby, B.: Characterizing the elastic properties of tissues. Mater. Today 14(3), 96–105 (2011) 3. Ansari, H.R., Gholami, A.: An improved support vector regression model for estimation of saturation pressure of crude oils. Fluid Phase Equilib. 402, 124–132 (2015) 4. AnX Robotica: Navicam small bowel capsule-system. https://www.anxrobotics.com/products/navicam-sb-capsule-system/ (2022). Accessed 11.05.2022 5. Banerjee, A., Pasupuleti, S., Mondal, K., Nezhad, M.M.: Application of data driven machine learning approach for modelling of non-linear filtration through granular porous media. Int. J. Heat Mass Transf. 179, 121650 (2021)
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
|
|