Assisting the Diagnosis of Cirrhosis in Chronic Hepatitis C Patients Based on Machine Learning Algorithms: A Novel
Emre Dirican1, Tayibe Bal2, Yusuf Onlen2
1Department of Biostatistics, Faculty of Medicine, Hatay Mustafa Kemal University, Hatay, Turkey.
Insights
Machine learning accurately predicts cirrhosis in chronic hepatitis C patients using routine lab tests. Key indicators include platelet count, alpha-fetoprotein, age, GGT, and prothrombin time.
Area of Science:
- Hepatology
- Machine Learning in Medicine
- Biostatistics
Background:
- Cirrhosis detection in chronic hepatitis C (CHC) is crucial for patient management.
- Routine laboratory tests offer a potential avenue for non-invasive cirrhosis assessment.
- Machine learning (ML) algorithms can analyze complex datasets to identify predictive patterns.
Purpose of the Study:
- To identify key laboratory features and their cut-off values for predicting cirrhosis in CHC patients.
- To evaluate the performance of ML algorithms in classifying cirrhosis based on routine tests.
- To establish an accurate ML-based model for cirrhosis detection.
Main Methods:
- Retrospective analysis of 1164 biopsy-proven CHC patients from a multicenter registry.
- Application of three ML algorithms to classify cirrhosis presence/absence.
- Feature importance analysis to determine key predictive laboratory tests.
Main Results:
- The Random Forest (RF) algorithm achieved the highest performance (Accuracy=0.89, AUC=0.87).
- The most significant features for cirrhosis prediction were platelet count, alpha-fetoprotein (AFP), age, gamma-glutamyl transferase (GGT), and prothrombin time (PT).
- Specific cut-off values were determined for these features, demonstrating significant risk coefficients (e.g., platelet < 182.000/mm³).
Conclusions:
- The RF-based ML model effectively classifies cirrhosis in CHC patients with high accuracy.
- Identified key laboratory features and cut-off values provide valuable tools for physicians in cirrhosis detection.
- Alpha-fetoprotein (AFP) emerged as a significant predictor, despite not being part of current non-invasive indices.
Aim:
This study aimed to determine the important features and cut-off values after demonstrating the detectability of cirrhosis using routine laboratory test results of chronic hepatitis C (CHC) patients in machine learning (ML) algorithms.
Methods:
This retrospective multicenter (37 referral centers) study included the data obtained from the Hepatitis C Turkey registry of 1164 patients with biopsy-proven CHC. Three different ML algorithms were used to classify the presence/absence of cirrhosis with the determined features.
Results:
The highest performance in the prediction of cirrhosis (Accuracy = 0.89, AUC = 0.87) was obtained from the Random Forest (RF) method. The five most important features that contributed to the classification were platelet, αlpha-feto protein (AFP), age, gamma-glutamyl transferase (GGT), and prothrombin time (PT). The cut-off values of these features were obtained as platelet < 182.000/mm3, AFP > 5.49 ng/mL, age > 52 years, GGT > 39.9 U/L, and PT > 12.35 s. Using cut-off values, the risk coefficients were AOR = 4.82 for platelet, AOR = 3.49 for AFP, AOR = 4.32 for age, AOR = 3.04 for GGT, and AOR = 2.20 for PT.
Conclusion:
These findings indicated that the RF-based ML algorithm could classify cirrhosis with high accuracy. Thus, crucial features and cut-off values for physicians in the detection of cirrhosis were determined. In addition, although AFP is not included in non-invasive indexes, it had a remarkable contribution in predicting cirrhosis.
Trial Registration:
Clinicaltrials.gov identifier: NCT03145844.
