The paper considers the problem of handling short sets of medical data. Effectively solving this problem will provide the ability to solve numerous classification and regression tasks in case of limited data in health decision support systems. Many similar tasks arise in various fields of medicine. The authors improved the regression method of data analysis based on artificial neural networks by introducing additional elements into the formula for calculating the output signal of the existing RBF-based input-doubling method. This improvement provides averaging of the result, which is typical for ensemble methods, and allows compensating for the errors of different signs of the predicted values. These two advantages make it possible to significantly increase the accuracy of the methods of this class. It should be noted that the duration of the training algorithm of the advanced method remains the same as for existing method. Experimental modeling was performed using a real short medical data. The regression task in rheumatology was solved based on only 77 observations. The optimal parameters of the method, which provide the highest prediction accuracy based on MAE and RMSE, were selected experimentally. A comparison of its efficiency with other methods of this class has been performed. The highest accuracy of the proposed RBF-based additive input-doubling method among the considered ones is established. The method can be modified by using other nonlinear artificial intelligence tools to implement its training and application algorithms and such methods can be applied in various fields of medicine.
Citation: Ivan Izonin, Roman Tkachenko, Ivanna Dronyuk, Pavlo Tkachenko, Michal Gregus, Mariia Rashkevych. Predictive modeling based on small data in clinical medicine: RBF-based additive input-doubling method[J]. Mathematical Biosciences and Engineering, 2021, 18(3): 2599-2613. doi: 10.3934/mbe.2021132
The paper considers the problem of handling short sets of medical data. Effectively solving this problem will provide the ability to solve numerous classification and regression tasks in case of limited data in health decision support systems. Many similar tasks arise in various fields of medicine. The authors improved the regression method of data analysis based on artificial neural networks by introducing additional elements into the formula for calculating the output signal of the existing RBF-based input-doubling method. This improvement provides averaging of the result, which is typical for ensemble methods, and allows compensating for the errors of different signs of the predicted values. These two advantages make it possible to significantly increase the accuracy of the methods of this class. It should be noted that the duration of the training algorithm of the advanced method remains the same as for existing method. Experimental modeling was performed using a real short medical data. The regression task in rheumatology was solved based on only 77 observations. The optimal parameters of the method, which provide the highest prediction accuracy based on MAE and RMSE, were selected experimentally. A comparison of its efficiency with other methods of this class has been performed. The highest accuracy of the proposed RBF-based additive input-doubling method among the considered ones is established. The method can be modified by using other nonlinear artificial intelligence tools to implement its training and application algorithms and such methods can be applied in various fields of medicine.
| [1] | D. I. Christine, M. Thinyane, Small Data approaches provide nuance and context to health datasets, Available from: http://theconversation.com/small-data-approaches-provide-nuance-and-context-to-health-datasets-121972. |
| [2] |
N. Melnykova, N. Shakhovska, M. G. ml, V. Melnykov, Using big data for formalization the patient's personalized data, Proc. Comp. Scien., 155 (2019), 624–629. doi: 10.1016/j.procs.2019.08.088
|
| [3] |
E. K. Wang, F. Wang, R. P. Sun, X. Liu, A new privacy attack network for remote sensing images classification with small training samples, Math. Biosci. Eng. 16 (2019), 4456–4476. doi: 10.3934/mbe.2019222
|
| [4] |
T. Mao, L. Yu, Y. Zhang, L. Zhou, Modified mahalanobis-taguchi system based on proper orthogonal decomposition for high-dimensional-small-sample-size data classification, Math. Biosci. Eng. 18 (2021), 426–444. doi: 10.3934/mbe.2021023
|
| [5] | L. Mochurad, M. Yatskiv, Simulation of a human operator's response to stressors under production conditions, CEUR-WS, 2753 (2020), 156–169. |
| [6] | V. Kotsovsky, F. Geche, A. Batyuk, On the computational complexity of learning bithreshold neural units and networks, Adv. Intel. Syst. Comp., 1020 (2019), 189–202. |
| [7] | V. Kotsovsky, F. Geche, A. Batyuk, Finite generalization of the offline spectral learning, in 2018 IEEE 2nd Intern. Conf. on Data Stream Mining Processing (DSMP), 2018,356–360. |
| [8] |
S. Fedushko, M. G. ml, T. Ustyianovych, Medical card data imputation and patient psychological and behavioral profile construction, Proc. Comp. Scien., 160 (2019), 354–361. doi: 10.1016/j.procs.2019.11.080
|
| [9] | S. Huang, H. Deng, Data Analytics A Small Data Approach, 1st edition, Routledge & CRC Press, 2021. |
| [10] | T. Hovorushchenko, A Herts, Y. Hnatchuk, Concept of intelligent decision support system in the legal regulation of the surrogate motherhood, CEUR-WS, 2488 (2019), 57–68. |
| [11] |
E.B. Hekler, P. Klasnja, G. Chevance G, N. M. Golaszewski, D. Lewis, I. Sim, Why we need a small data paradigm, BMC Med., 17 (2019), 133. doi: 10.1186/s12916-019-1366-x
|
| [12] |
N. Shahid, T. Rappon, W. Berta, Applications of artificial neural networks in health care organizational decision-making: a scoping review, PLOS ONE, 14 (2019), e0212356. doi: 10.1371/journal.pone.0212356
|
| [13] | S. Kaczor, N. Kryvinska, It is all about services-fundamentals, drivers, and business models, J. Serv. Sci. Res., 5 (2013), 125–154. |
| [14] | J. Wang, Data Warehousing and Mining: Concepts, Methodologies, Tools, and Applications, IGI Global, 2008. |
| [15] |
D. Wu, K. Warwick, Z. Ma, M. N. Gasson, J. G. Burgess, S. Pan, et. al., Prediction of parkinson's disease tremor onset using a radial basis function neural network based on particle swarm optimization, Int. J. Neural. Syst., 20 (2010), 109–116. doi: 10.1142/S0129065710002292
|
| [16] |
M. E. Karar, Robust RBF neural network‑based backstepping controller for implantable cardiac pacemakers, I. J. Adap. Cont. Sign. Proc., 32 (2018), 1040–1051. doi: 10.1002/acs.2884
|
| [17] | W. Aftab, M. Moinuddin, M. S. Shaikh, A novel kernel for RBF based neural networks, Abs. Appl. Anal., 2 (2014), 176253 |
| [18] | M. Moinuddin, I. Naseem, W. Aftab, S. A. Bencherif, A. Memic, A weighted cosine RBF neural networks, J. Mol. Biol. Biotech., 2 (2017), 2. |
| [19] |
T. Shaikhina, N. A. Khovanova, Handling limited datasets with neural networks in medical applications: a small-data approach, Artifl. Intel. Med., 75 (2017), 51–63. doi: 10.1016/j.artmed.2016.12.003
|
| [20] | T. Shaikhina, D. Lowe, S. Daga, D. Briggs, R. Higgins, N. Khovanova, Machine learning for predictive modelling based on small data in biomedical engineering, IFAC-PapersOnLine, 48 (2015), 469–474. |
| [21] | I. Izonin, R. Tkachenko, S. Fedushko, D. Koziy, K. Zub, O, Vovk, RBF-based input doubling method for small medical data processing, Adv. Intell. Syst. Comput., 2021. |
| [22] | Y. Bodyanskiy, I. Pliss, A. Deineko, Multilayer radial-basis function network and its learning, IEEE Int. Conf. Comp. Sci. Inf. Tech., (2020), 92–95. |
| [23] | S. Babichev, J. Škvor, Technique of gene expression profiles extraction based on the complex use of clustering and classification methods, Diagnostics, 10 (2020), 584. |
| [24] |
T. Kohonen, Essentials of the self-organizing map, Neur. Netw., 37 (2013), 52–65. doi: 10.1016/j.neunet.2012.09.018
|
| [25] |
S. Subbotin, Radial-basis function neural network synthesis on the basis of decision tree, Opt. Mem. Neur. Networks, 29 (2020), 7–18. doi: 10.3103/S1060992X20010051
|
| [26] | F. Samuelson, D. G. Brown, Application of cover's theorem to the evaluation of the performance of CI observers, Int. Joint Conf. Neur. Netw., (2011), 1020–1026. |
| [27] |
Ye. V. Bodyanskiy, A. O. Deineko, Ya. V. Kutsenko, On-line kernel clustering based on the general regression neural network and kohonen's self-organizing map, Aut. Control Comp. Sci., 51 (2017), 55–62. doi: 10.3103/S0146411617010023
|
| [28] | M. Deshp, Using neural networks for regression: radial basis function networks, Available from: https://pythonmachinelearning.pro/using-neural-networks-for-regression-radial-basis-function-networks/. |
| [29] | R. Tkachenko, H. Kutucu, I. Izonin, A. Doroshenko, Y. Tsymbal, Non-iterative neural-like predictor for solar energy in libya, CEUR-WS, 2105 (2018), 35–45. |
Figures(4) / Tables(2)
Ivan Izonin, Roman Tkachenko, Ivanna Dronyuk, Pavlo Tkachenko, Michal Gregus, Mariia Rashkevych. Predictive modeling based on small data in clinical medicine: RBF-based additive input-doubling method[J]. Mathematical Biosciences and Engineering, 2021, 18(3): 2599-2613. doi: 10.3934/mbe.2021132
DownLoad: