Assessing the performance of ChatGPT in answer- ing questions regarding cirrhosis and hepatocellu- lar carcinoma

被引：382

作者：

Yeo, Yee Hui ^{[1
]}

Samaan, Jamil S. ^{[1
]}

Ng, Wee Han ^{[2
]}

Ting, Peng-Sheng ^{[3
]}

Trivedi, Hirsh ^{[1
,4
]}

Vipani, Aarshi ^{[1
]}

Ayoub, Walid ^{[1
,4
]}

Yang, Ju Dong ^{[1
,4
,5
]}

Liran, Omer ^{[6
,7
]}

Spiegel, Brennan ^{[1
,7
]}

Kuo, Alexander ^{[1
,4
]}

机构：

[1] Cedars Sinai Med Ctr, Karsh Div Gastroenterol & Hepatol, Dept Med, Los Angeles, CA USA

[2] Univ Bristol, Bristol Med Sch, Bristol, England

[3] Tulane Univ, Sch Med, New Orleans, LA USA

[4] Cedars Sinai Med Ctr, Comprehens Transplant Ctr, Los Angeles, CA USA

[5] Cedars Sinai Med Ctr, Samuel Oschin Comprehens Canc Inst, Los Angeles, CA USA

[6] Cedars Sinai, Dept Psychiat & Behav Sci, Los Angeles, CA USA

[7] Cedars Sinai, Div Hlth Serv Res, Dept Med, Los Angeles, CA USA

来源：

CLINICAL AND MOLECULAR HEPATOLOGY | 2023年 / 29卷 / 03期

关键词：

Artificial intelligence; Patient education as topic; Health communication; Telemedicine; Chronic disease management; PATIENT; COMPLICATIONS; SURVEILLANCE;

D O I：

10.3350/cmh.2023.0089

中图分类号：

R57 [消化系及腹部疾病];

学科分类号：

摘要：

Background/Aims: Patients with cirrhosis and hepatocellular carcinoma (HCC) require extensive and personalized care to improve outcomes. ChatGPT (Generative Pre-trained Transformer), a large language model, holds the potential to provide professional yet patient-friendly support. We aimed to examine the accuracy and reproducibility of ChatGPT in answering questions regarding knowledge, management, and emotional support for cirrhosis and HCC. Methods: ChatGPT's responses to 164 questions were independently graded by two transplant hepatologists and resolved by a third reviewer. The performance of ChatGPT was also assessed using two published questionnaires and 26 questions formulated from the quality measures of cirrhosis management. Finally, its emotional support capacity was tested. Results: We showed that ChatGPT regurgitated extensive knowledge of cirrhosis (79.1% correct) and HCC (74.0% cor-rect), but only small proportions (47.3% in cirrhosis, 41.1% in HCC) were labeled as comprehensive. The performance was better in basic knowledge, lifestyle, and treatment than in the domains of diagnosis and preventive medicine. For the quality measures, the model answered 76.9% of questions correctly but failed to specify decision-making cut-off s and treatment durations. ChatGPT lacked knowledge of regional guidelines variations, such as HCC screening criteria. How-ever, it provided practical and multifaceted advice to patients and caregivers regarding the next steps and adjusting to a new diagnosis. Conclusions: We analyzed the areas of robustness and limitations of ChatGPT's responses on the management of cirrhosis and HCC and relevant emotional support. ChatGPT may have a role as an adjunct informational tool for patients and physicians to improve outcomes. (Clin Mol Hepatol 2023;29:721-732)

引用

页码：721 / 732

页数：13

共 30 条

[1] Guide for diagnosis and treatment of hepatocellular carcinoma [J].

Attwa, Magdy Hamed ;

El-Etreby, Shahira Aly .

WORLD JOURNAL OF HEPATOLOGY, 2015, 7 (12) :1632-1651

[2]

Bogost Ian., 2015, The Atlantic

[3]

Brown AF, 2019, AM J PUBLIC HEALTH, V109, pS72, DOI [10.2105/AJPH.2018.304844, 10.2105/ajph.2018.304844]

[4]

Christiano PF, 2017, 31 C NEURAL INFORM P

[5] Upper digestive bleeding in cirrhosis. Post-therapeutic outcome and prognostic indicators [J].

D'Amico, G ;

De Franchis, R .

HEPATOLOGY, 2003, 38 (03) :599-612

[6] Increasing Economic Burden in Hospitalized Patients With Cirrhosis: Analysis of a National Database [J].

Desai, Archita P. ;

Mohan, Prashanthinie ;

Nokes, Brandon ;

Sheth, Deekksha ;

Knapp, Shannon ;

Boustani, Malaz ;

Chalasani, Naga ;

Fallon, Michael B. ;

Calhoun, Elizabeth A. .

CLINICAL AND TRANSLATIONAL GASTROENTEROLOGY, 2019, 10

[7] Patient-Reported Barriers Are Associated With Lower Hepatocellular Carcinoma Surveillance Rates in Patients With Cirrhosis [J].

Farvardin, Sherean ;

Patel, Jaimin ;

Khambaty, Maleka ;

Yerokun, Olutola A. ;

Mok, Huram ;

Tiro, Jasmin A. ;

Yopp, Adam C. ;

Parikh, Neehar D. ;

Marrero, Jorge A. ;

Singal, Amit G. .

HEPATOLOGY, 2017, 65 (03) :875-884

[8] The global, regional, and national burden of gastro-oesophageal reflux disease in 195 countries and territories, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017 [J].

Dirac M.A. ;

Safiri S. ;

Tsoi D. ;

Adedoyin R.A. ;

Afshin A. ;

Akhlaghi N. ;

Alahdab F. ;

Almulhim A.M. ;

Amini S. ;

Ausloos F. ;

Bacha U. ;

Banach M. ;

Bhagavathula A.S. ;

Bijani A. ;

Biondi A. ;

Borzì A.M. ;

Colombara D. ;

Dagnew B. ;

Daryani A. ;

Davitoiu D.V. ;

Demeke F.M. ;

Demoz G.T. ;

Do H.P. ;

Etemadi A. ;

Farzadfar F. ;

Fischer F. ;

Gebre A.K. ;

Gebremariam H. ;

Gebremichael B. ;

Ghashghaee A. ;

Ghoshal U.C. ;

Hamidi S. ;

Hasankhani M. ;

Hassan S. ;

Hay S.I. ;

Hoang C.L. ;

Hole M.K. ;

Ikuta K.S. ;

Ilesanmi O.S. ;

Irvani S.S.N. ;

James S.L. ;

Joukar F. ;

Kabir A. ;

Kassaye H.G. ;

Kavetskyy T. ;

Kengne A.P. ;

Khalilov R. ;

Khan M.U. ;

Khan E.A. ;

Khan M. .

LANCET GASTROENTEROLOGY & HEPATOLOGY, 2020, 5 (06) :561-581

[9] How Does ChatGPT Perform on the United States Medical Licensing Examination (USMLE)? The Implications of Large Language Models for Medical Education and Knowledge Assessment [J].

Gilson, Aidan ;

Safranek, Conrad W. ;

Huang, Thomas ;

Socrates, Vimig ;

Chi, Ling ;

Taylor, Richard Andrew ;

Chartash, David .

JMIR MEDICAL EDUCATION, 2023, 9

[10] Liver cirrhosis [J].

Gines, Pere ;

Krag, Aleksander ;

Abraldes, Juan G. ;

Sola, Elsa ;

Fabrellas, Nuria ;

Kamath, Patrick S. .

LANCET, 2021, 398 (10308) :1359-1376

← 1 2 3 →