BackgroundSince most published articles comparing the performance of artificial neural network (ANN) models and logistic regression (LR) models for predicting hepatocellular carcinoma (HCC) outcomes used only a single dataset, the essential issue of internal validity (reproducibility) of the models has not been addressed. The study purposes to validate the use of ANN model for predicting in-hospital mortality in HCC surgery patients in Taiwan and to compare the predictive accuracy of ANN with that of LR model.Methodology/Principal FindingsPatients who underwent a HCC surgery during the period from 1998 to 2009 were included in the study. This study retrospectively compared 1,000 pairs of LR and ANN models based on initial clinical data for 22,926 HCC surgery patients. For each pair of ANN and LR models, the area under the receiver operating characteristic (AUROC) curves, Hosmer-Lemeshow (H-L) statistics and accuracy rate were calculated and compared using paired T-tests. A global sensitivity analysis was also performed to assess the relative significance of input parameters in the system model and the relative importance of variables. Compared to the LR models, the ANN models had a better accuracy rate in 97.28% of cases, a better H-L statistic in 41.18% of cases, and a better AUROC curve in 84.67% of cases. Surgeon volume was the most influential (sensitive) parameter affecting in-hospital mortality followed by age and lengths of stay.Conclusions/SignificanceIn comparison with the conventional LR model, the ANN model in the study was more accurate in predicting in-hospital mortality and had higher overall performance indices. Further studies of this model may consider the effect of a more detailed database that includes complications and clinical examination findings as well as more detailed outcome data.

Neoadjuvant concurrent chemoradiotherapy has been widely used for rectal cancer to improve local tumor control. The varied response of individual tumors encouraged us to search for useful biomarkers to predict the therapeutic response. The study was aimed to evaluate the prognostic impact of lipid biosynthesis-associated biomarkers in rectal cancer patients treated with preoperative chemoradiotherapy. Through analysis of the previously published gene expression profiling database focusing on genes associated with lipid biosynthesis, we found that HSD17B2 and HMGCS2 were the top two significantly upregulated genes in the non-responders. We further evaluated their expression by immunohistochemistry in the pre-treatment tumor specimens from 172 patients with rectal cancer and statistically analyzed the associations between their expression and various clinicopathological factors, as well as survival. High expression of HMGCS2 or HSD17B2 was significantly associated with advanced pre- and post-treatment tumor or nodal status (P < 0.001) and lower tumor regression grade (P < 0.001). More importantly, high expression of either HMGCS2 or HSD17B2 was of prognostic significance, with HMGCS2 overexpression indicating poor prognosis for disease-free survival (P = 0.0003), local recurrence-free survival (P = 0.0115), and metastasis-free survival (P = 0.0119), while HSD17B2 overexpression was associated with poor prognosis for disease-free survival (P <0.0001), local recurrence-free survival (P = 0.0009), and metastasis-free survival (P < 0.0001). In multivariate analysis, only HSD17B2 overexpression remained as an independent prognosticator for shorter disease-free survival (P < 0.001) and metastasis-free survival (P = 0.008). In conclusion, high expression of either HSD17B2 or HMGCS2 predicted poor susceptibility of rectal cancer to preoperative chemoradiotherapy. Both acted as promising prognostic factors, particularly HSD17B2.

The data in this study indicate that clinicians and health researchers should weight disease-specific measures more heavily than generic measures when evaluating treatment outcomes.

In comparison with the conventional LR model, the ANN model in this study was more accurate in predicting 5-year mortality. Further studies of this model may consider the effect of a more detailed database that includes complications and clinical examination findings as well as more detailed outcome data.

The data revealed dramatically improved post-cholecystectomy QOL. However, QOL change was simultaneously associated with preoperative functional status and demographic characteristics.