P-L02 Machine learning to evaluate liver reserve function based on venous blood biochemistry

Background The accurate and comprehensive evaluation of liver reserve function is crucial for daily follow-up and medical treatment of patients with liver disease. However, existing techniques are too complex and costly for universal implementation.To develop a convenient, reliable method to evaluate liver reserve function based on eight biochemical indicators measured from venous blood. Methods Blood test results (albumin (Alb), total bilirubin (TBIL), prothrombin time (PT), international normalized ratio (INR), total cholesterol (TC), cholinesterase (ChE), aspartate amino transferase (AST), and alanine transaminase (ALT)) were collected retrospectively from 660 patients treated at the first hospital of Ianzhou University from 2016 to 2018. As the reference standard for liver reserve function, indocyanine green (ICG) clearance test results were also collected from the same patients at the same times. The patient data were processed and analyzed to construct a machine learning model, eXtreme Gradient Boosting (XGBoost), and a generalized linear model (GLM) to predict liver reserve function based on the eight biochemical indicators. Results Results showed that the predicted XGBoost values were closely correlated with the actual ICG 15-minute retention rates (R = 0.969, R2 = 0.939), while the GLM values had a relatively low correlation (R = 0.566, R2 = 0.320). These findings indicate that the developed model can be used to evaluate liver reserve function with comparable performance to the ICG clearance test. Furthermore, the XGBoost model exhibited superior prediction compared with the GLM. Hence, the XGBoost model developed using machine learning can be utilized to evaluate liver reserve function from eight biochemical indicators that are closely related to liver function, commonly used clinically, and easier to obtain than ICG clearance measures. Conclusions The results predicted by the XGBoost model were highly accurate when compared with the results of the actual ICG test, demonstrating the strong practical clinical value of the model.

Comparison of preoperative two-dimensional shear wave elastography, indocyanine green clearance test and biomarkers for post hepatectomy liver failure prediction in patients with hepatocellular carcinoma

Background The preoperative prediction of post hepatectomy liver failure (PHLF) is essential, but there is no gold standard for the prediction at present, and the efficacy of different methods for the prediction has not been compared systematically. In this study, we aimed to compare the efficacy of preoperative two-dimensional shear wave elastography (2D-SWE), indocyanine green (ICG) clearance test and biomarkers for PHLF prediction in patients with hepatocellular carcinoma (HCC). Methods We retrospectively studied 215 patients with HCC, who had undergone major liver resection in our hospital. Preoperative data of each patient, including liver stiffness value (LSV) of underlying hepatic parenchyma measured by 2D-SWE, ICG retention rate at 15 min (ICG-R15) measured by ICG clearance test, albumin-bilirubin (ALBI) scores, aspartate aminotransferase–platelet ratio index (APRI), and Fibrosis-4 (FIB-4) were collected for analysis. Post hepatectomy outcomes of study patients were also recorded for assessment of PHLF. The study patients were divided into development cohort (133 patients without PHLF, and 17 patients with PHLF) and validation cohort (59 patients without PHLF, and 6 patients with PHLF) randomly. Results In the development cohort, LSV, ICG-R15 and ALBI scores were significantly different between patients with and without PHLF, while no significant difference of APRI and FIB-4 scores was found. LSV had higher AUC (the area under the receiver operating characteristic curve) (AUC = 0.795) for PHLF prediction than ICG-R15 (AUC = 0.619) and ALBI scores (AUC = 0.686) (p < 0.05 for all comparisons). In the validation cohort, the cutoff value of LSV obtained from the development cohort, 10.35 kPa, revealed higher specificity (76.3%) for PHLF prediction than ICG-R15 (specificity: 66.1%) and ALBI scores (specificity: 69.5%) (p < 0.0001). Conclusions Compared with ICG-R15, ALBI scores, APRI and FIB-4, LSV measured by 2D-SWE may demonstrate better efficacy for preoperative PHLF prediction in patients with HCC.

The technique of 3D reconstruction combining with biochemistry to build an equivalent formula of indocyanine green (ICG) clearance test to assess the liver reserve function

Background The indocyanine green (ICG) clearance test is the main method of evaluating the liver reserve function before hepatectomy. However, some patients may be allergic to ICG or the equipment of ICG clearance test was lack, leading to be difficult to evaluate liver reserve function. We aim to find an alternative tool to assist the clinicians to evaluate the liver reserve function for those who were allergic to the ICG or lack of equipment before hepatectomy. Methods We retrospected 300 patients to investigate the risk factors affecting the liver reserve function and to build an equivalent formula to predict ICG 15 min retention rate (ICG-R15) value. Results We found that the independent risk factors affecting ICG clearance test were total bilirubin, albumin, and spleen-to-non-neoplastic liver volume ratio (SNLR). The equivalent formula of the serological index combining with SNLR was: ICG-R15 = 0.36 × TB (umol/L) − 0.78 × ALB(g/L) + 7.783 × SNLR + 0.794 × PT (s) − 0.016 × PLT(/109) − 0.039 × ALT (IU/L) + 0.043 × AST (IU/L) + 23.846. The equivalent formula of serum index was: ICG-R152 = 24.665 + 0.382 × TB (umol/L) − 0.799 × ALB(g/L) − 0.025 × PLT(/109) + 0.048 × AST(IU/L) − 0.045 × ALT(IU/L). And the area under the ROC curve (AUC) of predicting ICG-R15 ≥ 10% was 0.861 and 0.857, respectively. Conclusion We found that SNLR was an independent risk factor affecting liver reserve function. Combining with SNLR to evaluate the liver reserve function was better than just basing on serology.

The technique of 3D reconstruction combining with biochemistry to build an equivalent formula of indocyanine green (ICG) clearance test to assess the liver reserve function

Background: The indocyanine green(ICG) clearance test is the main method of evaluating the liver reserve function before hepatectomy. However, some patients may be allergic to ICG or the equipment of ICG clearance test was lack, leading to be difficult to evaluate liver reserve function. We aim to find an alternative tool to assist the clinicians to evaluate the liver reserve function for those who were allergic to the ICG or lack of equipment before hepatectomy.Methods: We retrospected 300 patients to investigate the risk factors affecting the liver reserve function and to build an equivalent formula to predict ICG 15 minutes retention rate (ICG-R15) value. Results：We found that the independent risk factors affecting ICG clearance test were total bilirubin, albumin, and spleen-to-non-neoplastic liver volume ratio (SNLR). The equivalent formula of the serological index combining with SNLR was: ICG-R15=0.36×TB (umol/L) - 0.78 × ALB(g/L) + 7.783×SNLR + 0.794×PT (s) - 0.016×PLT(/109) - 0.039× ALT (IU/L) + 0.043 × AST (IU/L) + 23.846. The equivalent formula of serum index was: ICG-R152=24.665 + 0.382×TB (umol/L) - 0.799 × ALB(g/L) - 0.025 × PLT(/109) + 0.048 × AST(IU/L) - 0.045 × ALT(IU/L). And the area under the ROC curve (AUC) of predicting ICG-R15≥10% was 0.861 and 0.857, respectively. Conclusion: We found that SNLR was an independent risk factor affecting liver reserve function. Combining with SNLR to evaluate the liver reserve function was better than just basing on serology.

The technique of 3D reconstruction combining with biochemistry to build an equivalent formula of indocyanine green (ICG) clearance test to assess the liver reserve function

Background: The indocyanine green(ICG) clearance test is the main method of evaluating the liver reserve function before hepatectomy. However, some patients may be allergic to ICG or the equipment of ICG clearance test was lack, leading to be difficult to evaluate liver reserve function. We aim to find an alternative tool to assist the clinicians to evaluate the liver reserve function for those who were allergic to the ICG or lack of equipment before hepatectomy.Methods: We retrospected 300 patients to investigate the risk factors affecting the liver reserve function and to build an equivalent formula to predict ICG 15 minutes retention rate (ICG-R15) value. Results: We found that the independent risk factors affecting ICG clearance test were total bilirubin, albumin, and spleen-to-non-neoplastic liver volume ratio (SNLR). The equivalent formula of the serological index combining with SNLR was: ICG-R15=0.36×TB (umol/L)- 0.78 × ALB(g/L) + 7.783×SNLR + 0.794×PT (s) - 0.016×PLT(/109) - 0.039× ALT (IU/L) + 0.043 × AST (IU/L) + 23.846. The equivalent formula of serum index was: ICG-R152=24.6650.382×TB (umol/L) - 0.799 × ALB(g/L) - 0.025 × PLT(/109) + 0.048 × AST(IU/L) - 0.045 × ALT(IU/L). And the area under the ROC curve (AUC) of predicting ICG-R15≥10% was 0.861 and 0.857, respectively. Conclusion: We found that SNLR was an independent risk factor affecting liver reserve function. Combining with SNLR to evaluate the liver reserve function was better than just basing on serology.

The technique of 3D reconstruction combining with biochemistry to build an equivalent formula of indocyanine green (ICG) clearance test to assess the liver reserve function

Background: The indocyanine green(ICG) clearance test is the main method of evaluating the liver reserve function before hepatectomy. However, some patients may be allergic to ICG or the equipment of ICG clearance test was lack, leading to be difficult to evaluate liver reserve function. We aim to find an alternative tool to assist the clinicians to evaluate the liver reserve function for those who were allergic to the ICG or lack of equipment before hepatectomy.Methods: We retrospected 300 patients to investigate the risk factors affecting the liver reserve function and to build an equivalent formula to predict ICG 15 minutes retention rate (ICG-R15) value.Results: We found that the independent risk factors affecting ICG clearance test were total bilirubin, albumin, and spleen-to-non-neoplastic liver volume ratio (SNLR). The equivalent formula of the serological index combining with SNLR was: ICG-R15 = 0.36 × TB (umol/L)- 0.78 × ALB(g/L) + 7.783 × SNLR + 0.794 × PT (s) − 0.016 × PLT(/109) − 0.039 × ALT (IU/L) + 0.043 × AST (IU/L) + 23.846. The equivalent formula of serum index was: ICG-R152 = 24.6650.382 × TB (umol/L) − 0.799 × ALB(g/L) − 0.025 × PLT(/109) + 0.048 × AST(IU/L) − 0.045 × ALT(IU/L). And the area under the ROC curve (AUC) of predicting ICG-R15 ≥ 10% was 0.861 and 0.857, respectively.Conclusion: We found that SNLR was an independent risk factor affecting liver reserve function. Combining with SNLR to evaluate the liver reserve function was better than just basing on serology.

Child-Pugh Parameters and Platelet Count as an Alternative to ICG Test for Assessing Liver Function for Major Hepatectomy

Objective. To study the correlations and discrepancies between Child-Pugh system and indocyanine green (ICG) clearance test in assessing liver function reserve and explore the possibility of combining two systems to gain an overall liver function assessment. Design. Retrospective analysis of 2832 hepatocellular carcinoma (HCC) patients graded as Child-Pugh A and Child-Pugh B with ICG clearance test being performed was conducted. Results. ICG retention rate at 15 minutes (ICG15) correlates with Child-Pugh score, however, with a large variance. Platelet count improves the correlation between Child-Pugh score and ICG15. ICG15 can be estimated using the following regression formula: estimated ICG15 (eICG15) = 45.1 + 0.435 × bilirubin − 0.917 × albumin + 0.491 × prothrombin time − 0.0283 × platelet (R2=0.455). Patients with eICG15 >20.0% who underwent major hepatectomy had a tendency towards more posthepatectomy liver failure (4.1% versus 8.0%, p=0.09) and higher in-hospital mortality (3.7% versus 8.0%, p=0.052). They also had shorter median overall survival (5.10±0.553 versus 3.01±0.878 years, p=0.015) and disease-free survival (1.37±0.215 versus 0.707±0.183 years, p=0.018). Conclusion. eICG15 can be predicted from Child-Pugh parameters and platelet count. eICG15 correlates with in-hospital mortality after major hepatectomy and predicts long-term survival.