Dirican Emre, Bal Tayibe, Onlen Yusuf, Sarigul Figen, User Ulku, Sari Nagehan Didem, Kurtaran Behice, Senates Ebubekir, Gunduz Alper, Zerdali Esra, Karsen Hasan, Batirel Ayse, Karaali Ridvan, Guner Hatice Rahmet, Yamazhan Tansu, Kose Sukran, Erben Nurettin, Ince Nevin Koc, Koksal Iftihar, Oztoprak Nefise, Yoruk Gulsen, Komur Suheyla, Kaya Sibel Yildiz, Bozkurt Ilkay, Gunal Ozgur, Yildiz Ilknur Esen, Inan Dilara, Barut Sener, Namiduru Mustafa, Tosun Selma, Turker Kamuran, Sener Alper, Hizel Kenan, Baykam Nurcan, Duygu Fazilet, Bodur Hurrem, Can Guray, Gul Hanefi Cem, Tartar Ayse Sagmak, Celebi Guven, Sunnetcioglu Mahmut, Karabay Oguz, Kumbasar Karaosmanoglu Hayat, Sirmatel Fatma, Tabak Omer Fehmi
Department of Biostatistics, Faculty of Medicine, Hatay Mustafa Kemal University, Hatay, Turkey.
Department of Infectious Diseases and Clinical Microbiology, Faculty of Medicine, Bolu Abant Izzet Baysal University, Bolu, Turkey.
J Clin Lab Anal. 2025 Jun;39(12):e70054. doi: 10.1002/jcla.70054. Epub 2025 May 19.
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.
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.
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/mm, 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.
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.
Clinicaltrials.gov identifier: NCT03145844.
本研究旨在利用慢性丙型肝炎（CHC）患者的常规实验室检测结果，在机器学习（ML）算法中证明肝硬化的可检测性之后，确定其重要特征和临界值。
这项回顾性多中心（37个转诊中心）研究纳入了从土耳其丙型肝炎登记处获得的1164例经活检证实为CHC患者的数据。使用三种不同的ML算法，根据确定的特征对肝硬化的存在与否进行分类。
随机森林（RF）方法在肝硬化预测方面表现最佳（准确率=0.89，曲线下面积=0.87）。对分类有贡献的五个最重要特征是血小板、甲胎蛋白（AFP）、年龄、γ-谷氨酰转移酶（GGT）和凝血酶原时间（PT）。这些特征的临界值分别为血小板<182,000/mm、AFP>5.49 ng/mL、年龄>52岁、GGT>39.9 U/L和PT>12.35秒。使用临界值，血小板的风险系数比值比（AOR）=4.82，AFP的AOR=3.49，年龄的AOR=4.32，GGT的AOR=3.04，PT的AOR=2.20。
这些发现表明，基于RF的ML算法能够高精度地对肝硬化进行分类。因此，确定了医生在检测肝硬化时的关键特征和临界值。此外，尽管AFP未包含在非侵入性指标中，但它在预测肝硬化方面有显著贡献。
Clinicaltrials.gov标识符：NCT03145844。
