Assisting the Diagnosis of Cirrhosis in Chronic Hepatitis C Patients Based on Machine Learning Algorithms: A Novel Non-Invasive Approach

Emre Dirican; Tayibe Bal; Yusuf Onlen; Figen Sarigul; Ulku User; Nagehan Didem Sari; Behice Kurtaran; Ebubekir Senates; Alper Gunduz; Esra Zerdali; Hasan Karsen; Ayse Batirel; Ridvan Karaali; Hatice Rahmet Guner; Tansu Yamazhan; Sukran Kose; Nurettin Erben; Nevin Koc Ince; Iftihar Koksal; Nefise Oztoprak; Gulsen Yoruk; Suheyla Komur; Sibel Yildiz Kaya; Ilkay Bozkurt; Ozgur Gunal; Ilknur Esen Yildiz; Dilara Inan; Sener Barut; Mustafa Namiduru; Selma Tosun; Kamuran Turker; Alper Sener; Kenan Hizel; Nurcan Baykam; Fazilet Duygu; Hurrem Bodur; Guray Can; Hanefi Cem Gul; Ayse Sagmak Tartar; Guven Celebi; Mahmut Sunnetcioglu; Oguz Karabay; Hayat Kumbasar Karaosmanoglu; Fatma Sirmatel; Omer Fehmi Tabak

doi:10.1002/jcla.70054

Assisting the Diagnosis of Cirrhosis in Chronic Hepatitis C Patients Based on Machine Learning Algorithms: A Novel Non-Invasive Approach

Emre Dirican
, Tayibe Bal
, Yusuf Onlen
, Figen Sarigul
, Ulku User
, Nagehan Didem Sari
, Behice Kurtaran
, Ebubekir Senates
, Alper Gunduz
, Esra Zerdali
, Hasan Karsen
, Ayse Batirel
, Ridvan Karaali
, Hatice Rahmet Guner
, Tansu Yamazhan
, Sukran Kose
, Nurettin Erben
, Nevin Koc Ince
, Iftihar Koksal
, Nefise Oztoprak

Gulsen Yoruk, Suheyla Komur, Sibel Yildiz Kaya, Ilkay Bozkurt, Ozgur Gunal, Ilknur Esen Yildiz, Dilara Inan, Sener Barut, Mustafa Namiduru, Selma Tosun, Kamuran Turker, Alper Sener, Kenan Hizel, Nurcan Baykam, Fazilet Duygu, Hurrem Bodur, Guray Can, Hanefi Cem Gul, Ayse Sagmak Tartar, Guven Celebi, Mahmut Sunnetcioglu, Oguz Karabay, Hayat Kumbasar Karaosmanoglu, Fatma Sirmatel, Omer Fehmi Tabak

Research output: Contribution to journal › Article › peer-review

Abstract

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.

Original language	English
Article number	e70054
Journal	Journal of Clinical Laboratory Analysis
Volume	39
Issue number	12
DOIs	https://doi.org/10.1002/jcla.70054
Publication status	Published - 19 May 2025
Externally published	Yes

Keywords

alfa-feto protein
chronic hepatitis C
classification
diagnosis of cirrhosis
machine learning

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1002/jcla.70054Licence: CC BY

Clinical Laboratory Analysis - 2025 - Dirican - Assisting the Diagnosis of Cirrhosis in Chronic Hepatitis C Patients BasedFinal published version, 970 KBLicence: CC BY

Cite this

Dirican, E., Bal, T., Onlen, Y., Sarigul, F., User, U., Sari, N. D., Kurtaran, B., Senates, E., Gunduz, A., Zerdali, E., Karsen, H., Batirel, A., Karaali, R., Guner, H. R., Yamazhan, T., Kose, S., Erben, N., Ince, N. K., Koksal, I., ... Tabak, O. F. (2025). Assisting the Diagnosis of Cirrhosis in Chronic Hepatitis C Patients Based on Machine Learning Algorithms: A Novel Non-Invasive Approach. Journal of Clinical Laboratory Analysis, 39(12), Article e70054. https://doi.org/10.1002/jcla.70054

@article{4c0bdbd004a3496eb81bdca98ab0dd32,

title = "Assisting the Diagnosis of Cirrhosis in Chronic Hepatitis C Patients Based on Machine Learning Algorithms: A Novel Non-Invasive Approach",

abstract = "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.",

keywords = "alfa-feto protein, chronic hepatitis C, classification, diagnosis of cirrhosis, machine learning",

author = "Emre Dirican and Tayibe Bal and Yusuf Onlen and Figen Sarigul and Ulku User and Sari, \{Nagehan Didem\} and Behice Kurtaran and Ebubekir Senates and Alper Gunduz and Esra Zerdali and Hasan Karsen and Ayse Batirel and Ridvan Karaali and Guner, \{Hatice Rahmet\} and Tansu Yamazhan and Sukran Kose and Nurettin Erben and Ince, \{Nevin Koc\} and Iftihar Koksal and Nefise Oztoprak and Gulsen Yoruk and Suheyla Komur and Kaya, \{Sibel Yildiz\} and Ilkay Bozkurt and Ozgur Gunal and Yildiz, \{Ilknur Esen\} and Dilara Inan and Sener Barut and Mustafa Namiduru and Selma Tosun and Kamuran Turker and Alper Sener and Kenan Hizel and Nurcan Baykam and Fazilet Duygu and Hurrem Bodur and Guray Can and Gul, \{Hanefi Cem\} and Tartar, \{Ayse Sagmak\} and Guven Celebi and Mahmut Sunnetcioglu and Oguz Karabay and \{Kumbasar Karaosmanoglu\}, Hayat and Fatma Sirmatel and Tabak, \{Omer Fehmi\}",

note = "Publisher Copyright: {\textcopyright} 2025 The Author(s). Journal of Clinical Laboratory Analysis published by Wiley Periodicals LLC.",

year = "2025",

month = may,

day = "19",

doi = "10.1002/jcla.70054",

language = "English",

volume = "39",

journal = "Journal of Clinical Laboratory Analysis",

issn = "0887-8013",

publisher = "Wiley-Liss Inc.",

number = "12",

}

Dirican, E, Bal, T, Onlen, Y, Sarigul, F, User, U, Sari, ND, Kurtaran, B, Senates, E, Gunduz, A, Zerdali, E, Karsen, H, Batirel, A, Karaali, R, Guner, HR, Yamazhan, T, Kose, S, Erben, N, Ince, NK, Koksal, I, Oztoprak, N, Yoruk, G, Komur, S, Kaya, SY, Bozkurt, I, Gunal, O, Yildiz, IE, Inan, D, Barut, S, Namiduru, M, Tosun, S, Turker, K, Sener, A, Hizel, K, Baykam, N, Duygu, F, Bodur, H, Can, G, Gul, HC, Tartar, AS, Celebi, G, Sunnetcioglu, M, Karabay, O, Kumbasar Karaosmanoglu, H, Sirmatel, F & Tabak, OF 2025, 'Assisting the Diagnosis of Cirrhosis in Chronic Hepatitis C Patients Based on Machine Learning Algorithms: A Novel Non-Invasive Approach', Journal of Clinical Laboratory Analysis, vol. 39, no. 12, e70054. https://doi.org/10.1002/jcla.70054

TY - JOUR

T1 - Assisting the Diagnosis of Cirrhosis in Chronic Hepatitis C Patients Based on Machine Learning Algorithms: A Novel Non-Invasive Approach

AU - Dirican, Emre

AU - Bal, Tayibe

AU - Onlen, Yusuf

AU - Sarigul, Figen

AU - User, Ulku

AU - Sari, Nagehan Didem

AU - Kurtaran, Behice

AU - Senates, Ebubekir

AU - Gunduz, Alper

AU - Zerdali, Esra

AU - Karsen, Hasan

AU - Batirel, Ayse

AU - Karaali, Ridvan

AU - Guner, Hatice Rahmet

AU - Yamazhan, Tansu

AU - Kose, Sukran

AU - Erben, Nurettin

AU - Ince, Nevin Koc

AU - Koksal, Iftihar

AU - Oztoprak, Nefise

AU - Yoruk, Gulsen

AU - Komur, Suheyla

AU - Kaya, Sibel Yildiz

AU - Bozkurt, Ilkay

AU - Gunal, Ozgur

AU - Yildiz, Ilknur Esen

AU - Inan, Dilara

AU - Barut, Sener

AU - Namiduru, Mustafa

AU - Tosun, Selma

AU - Turker, Kamuran

AU - Sener, Alper

AU - Hizel, Kenan

AU - Baykam, Nurcan

AU - Duygu, Fazilet

AU - Bodur, Hurrem

AU - Can, Guray

AU - Gul, Hanefi Cem

AU - Tartar, Ayse Sagmak

AU - Celebi, Guven

AU - Sunnetcioglu, Mahmut

AU - Karabay, Oguz

AU - Kumbasar Karaosmanoglu, Hayat

AU - Sirmatel, Fatma

AU - Tabak, Omer Fehmi

PY - 2025/5/19

Y1 - 2025/5/19

N2 - 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.

AB - 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.

KW - alfa-feto protein

KW - chronic hepatitis C

KW - classification

KW - diagnosis of cirrhosis

KW - machine learning

U2 - 10.1002/jcla.70054

DO - 10.1002/jcla.70054

M3 - Article

C2 - 40384539

AN - SCOPUS:105005551496

SN - 0887-8013

VL - 39

JO - Journal of Clinical Laboratory Analysis

JF - Journal of Clinical Laboratory Analysis

IS - 12

M1 - e70054

ER -

Assisting the Diagnosis of Cirrhosis in Chronic Hepatitis C Patients Based on Machine Learning Algorithms: A Novel Non-Invasive Approach

Abstract

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this