Comparison of Neural Network and Logistic Regression Methods to Predict Xerostomia after Radiotherapy

To evaluate the ability to predict xerostomia after
radiotherapy, we constructed and compared neural network and
logistic regression models. In this study, 61 patients who completed a
questionnaire about their quality of life (QoL) before and after a full
course of radiation therapy were included. Based on this questionnaire,
some statistical data about the condition of the patients’ salivary
glands were obtained, and these subjects were included as the inputs of
the neural network and logistic regression models in order to predict
the probability of xerostomia. Seven variables were then selected from
the statistical data according to Cramer’s V and point-biserial
correlation values and were trained by each model to obtain the
respective outputs which were 0.88 and 0.89 for AUC, 9.20 and 7.65
for SSE, and 13.7% and 19.0% for MAPE, respectively. These
parameters demonstrate that both neural network and logistic
regression methods are effective for predicting conditions of parotid
glands.





References:
<p>[1] T. F. Lee, P. J. Chao, H. M. Ting, S. H. Lo, Y. W. Wang, C. C. Tuan, F.
M. Fang, and T. J. Su, &quot;Comparative analysis of SmartArc-based dual arc
volumetric-modulated arc radiotherapy (VMAT) versus
intensity-modulated radiotherapy (IMRT) for nasopharyngeal
carcinoma,&quot; J Appl Clin Med Phys, vol. 12, p. 3587, 2011.
[2] C. Y. Hsiung, H. M. Ting, H. Y. Huang, C. H. Lee, E. Y. Huang, and H. C.
Hsu, &quot;Parotid-sparing intensity-modulated radiotherapy (IMRT) for
nasopharyngeal carcinoma: preserved parotid function after IMRT on
quantitative salivary scintigraphy, and comparison with historical data
after conventional radiotherapy,&quot; Int J Radiat Oncol Biol Phys, vol. 66,
pp. 454-61, 2006.
[3] M. Agulnik and J. B. Epstein, &quot;Nasopharyngeal carcinoma: current
management, future directions and dental implications,&quot; Oral Oncol, vol.
44, pp. 617-27, 2008.
[4] Y. P. Talmi, Z. Horowitz, L. Bedrin, M. Wolf, G. Chaushu, J.
Kronenberg, and M. R. Pfeffer, &quot;Quality of life of nasopharyngeal
carcinoma patients,&quot; Cancer, vol. 94, pp. 1012-7, 2002.
[5] V. W. Wu, M. T. Ying, and D. L. Kwong, &quot;Evaluation of
radiation-induced changes to parotid glands following conventional
radiotherapy in patients with nasopharygneal carcinoma,&quot; Br J Radiol,
vol. 84, pp. 843-9, 2011.
[6] J.-C. Shyu and H.-Y. Liou, &quot; The financial distress prediction model
under consideration of business cycle and industry factors - the
application of logistic regression model and DEA-DA model &quot; Journal of
Risk Management, vol. 12, pp. 157-183, 2010.
[7] J. O. Deasy, J. R. Alaly, and K. Zakaryan, &quot;Obstacles and advances in
intensity-modulated radiation therapy treatment planning,&quot; Front Radiat
Ther Oncol, vol. 40, pp. 42-58, 2007.
[8] M. Isaksson, J. Jalden, and M. J. Murphy, &quot;On using an adaptive neural
network to predict lung tumor motion during respiration for radiotherapy
applications,&quot; Med Phys, vol. 32, pp. 3801-9, 2005.
[9] L. Zhang, L. Jia, and W. Zhu, &quot;Overview of traffic flow hybrid ANN
forecasting algorithm study,&quot; in 2010 International Conference on
Computer Application and System Modeling (ICCASM), , 2010, pp.
V1-615-V1-619.
[10] M. H. Zweig and G. Campbell, &quot;Receiver-operating characteristic (ROC)
plots: a fundamental evaluation tool in clinical medicine,&quot; Clin Chem, vol.
39, pp. 561-77, 1993.
[11] H.-L. Chen and D.-W. Tsai, &quot;A study of predictive abilities for different
models,&quot; Journal of Soil and Water Conservation, vol. 37, pp. 127-138,
2005.
[12] M. Z. Liu, L. L. Tang, J. F. Zong, Y. Huang, Y. Sun, Y. P. Mao, L. Z. Liu,
A. H. Lin, and J. Ma, &quot;Evaluation of sixth edition of AJCC staging system
for nasopharyngeal carcinoma and proposed improvement,&quot; Int J Radiat
Oncol Biol Phys, vol. 70, pp. 1115-23, 2008.
[13] A. Lin, H. M. Kim, J. E. Terrell, L. A. Dawson, J. A. Ship, and A.
Eisbruch, &quot;Quality of life after parotid-sparing IMRT for head-and-neck
cancer: a prospective longitudinal study,&quot; Int J Radiat Oncol Biol Phys,
vol. 57, pp. 61-70, 2003.
[14] F. M. Fang, W. L. Tsai, T. F. Lee, K. C. Liao, H. C. Chen, and H. C. Hsu,
&quot;Multivariate analysis of quality of life outcome for nasopharyngeal
carcinoma patients after treatment,&quot; Radiother Oncol, vol. 97, pp. 263-9,
2010.
[15] I. Beetz, C. Schilstra, A. van der Schaaf, E. R. van den Heuvel, P.
Doornaert, P. van Luijk, A. Vissink, B. F. van der Laan, C. R. Leemans,
H. P. Bijl, M. E. Christianen, R. J. Steenbakkers, and J. A. Langendijk,
&quot;NTCP models for patient-rated xerostomia and sticky saliva after
treatment with intensity modulated radiotherapy for head and neck
cancer: the role of dosimetric and clinical factors,&quot; Radiother Oncol, vol.
105, pp. 101-6, 2012.
[16] A. Borque, G. Sanz, C. Allepuz, L. Plaza, P. Gil, and L. A. Rioja, &quot;The
use of neural networks and logistic regression analysis for predicting
pathological stage in men undergoing radical prostatectomy: a population
based study,&quot; J Urol, vol. 166, pp. 1672-8, 2001.
[17] J. H. Song, S. S. Venkatesh, E. A. Conant, P. H. Arger, and C. M. Sehgal,
&quot;Comparative analysis of logistic regression and artificial neural network
for computer-aided diagnosis of breast masses,&quot; Acad Radiol, vol. 12, pp.
487-95, 2005.
[18] B. Eftekhar, K. Mohammad, H. E. Ardebili, M. Ghodsi, and E. Ketabchi,
&quot;Comparison of artificial neural network and logistic regression models
for prediction of mortality in head trauma based on initial clinical data,&quot;
BMC Medical Informatics and Decision Making, vol. 5, p. 3, 2005.</p>