From: Image-based AI diagnostic performance for fatty liver: a systematic review and meta-analysis
Author | Year | study design | country | Single/Multi center | Inclusion criteria | Exclusion criteria | age | Source of data | Data range(time) | imaging technique | Reference | Algorithm |
---|---|---|---|---|---|---|---|---|---|---|---|---|
G.Li et al. | 2008 | Retrospective | China | Single | NR | NR | NR | WuHan municipal hospital | NR | Conventional ultrasound | US | SVM |
M.Hájek et al. | 2011 | Retrospective | Czech Republic | Single | With transplanted livers for 1–11 years who underwent regular clinical examination, including protocol liver biopsies. | NR | NR | NR | March 2009 to November 2009 and March 2010 to July 2010 | MR spectrum | pathology | Linear regression |
U.R.Acharya et al. | 2012 | Retrospective | Singapore | Single | NR | NR | NR | NR | NR | Conventional ultrasound | US | Decision Tree |
F.U.A.A.Minhas et al. | 2012 | Retrospective | USA | Single | NR | NR | NR | Multan Institute of Nuclear Medicine and Radiotherapy (MINAR) | NR | Conventional ultrasound | US | SVM |
R.Ribeiro et al. | 2012 | Retrospective | Portugal | Single | had known diagnosis based on liver biopsy results | NR | NR | Hospital de Santa Maria, Medical School of Lisbon, Portugal | NR | Conventional ultrasound | pathology | Bayes |
R.T.Ribeiro et al. | 2014 | Retrospective | Portugal | Single | All patients were outpatients with known diagnosis, based on the clinical and US criteria. | NR | NR | The Gastroenterology Department of the Santa Maria Hospital, Lisbon | NR | Conventional ultrasound | US | Bayes |
M.Owjimehr et al. | 2015 | Retrospective | Iran | NR | NR | NR | NR | NR | NR | Conventional ultrasound | US | SVM |
L.Saba et al. | 2016 | Retrospective | Italy | Single | NR | NR | NR | NR | NR | Conventional ultrasound | US | BPNN |
V.Kuppili et al. | 2017 | Retrospective | India | Single | The patient with normal body mass index. | NR | NR | Instituto Superior Tecnico (IST), University of Lisbon, Portugal | NR | Conventional ultrasound | pathology | ELM |
M.Biswas et al. | 2018 | Retrospective | India | Single | NR | NR | NR | The Santa Maria Hospital (ethics approval granted), in Lisbon, Portugal | NR | Conventional ultrasound | pathology | CNN |
V.Sharma et al. | 2018 | Retrospective | India | Single | age group of 25–60 years old | NR | NR | Delta Diagnostic Centre Patiala, India | NR | Conventional ultrasound | US | SVM |
A.Han et al. | 2020 | Prospective | Poland | Single | Age of at least 18 years, willingness and ability to participate in. | Clinical, laboratory, or histologic evidence of a liver disease other than NAFLD; excessive alcohol consumption (0.30 g per day within the past 10 years or 0.10 g per day in the previous year); and steatogenic or hepatoxic medication use. | NAFLD 52 ± 14; control 46 ± 21 | The University of California, San Diego, NAFLD Research Center | Between February 2012 and March 2014 | Conventional ultrasound | MRI | CNN |
E.C.Constantinescu et al. | 2021 | Retrospective | Romania | Single | Age ranging from 18 to 92 from the outpatient clinic of a private healthcare network. | Excessive alcohol consumption or > 20 g ethanol per day, history or clinical and/or laboratory evidence of liver disease, hepatotoxic medication use. | NR | The University of Medicine and Pharmacy, Research Center of Gastroenterology and Hepatology, Craiova from the outpatient clinic of a private healthcare network. | NR | B-mode ultrasound and ARFI elastography | US | Inception v3 |
M.Byra et al. | 2022 | Prospective | USA | Single | Age ≥ 18 years old, known or suspected NAFLD, and willingness and ability to participate. | Clinical, laboratory, or histology evidence of a liver disease other than NAFLD, excessive alcohol consumption (≥ 14 (men) or ≥ 7(women)drinks/week), and steatogenic or hepatoxic medication use. | NR | The University of California, San Diego (UCSD) | Between 2016 and 2018 | Conventional ultrasound | MRI | ResNet50 CNN;LR |
F.Destrempes et al. | 2022 | Retrospective | Canada | Single | A liver biopsy was scheduled as part of their clinical standard of care. | (a) images were not acquired with the probe assigned to the study protocol; (b) the underlying pathology did not meet eligibility criteria due to ethanol consumption, drug-induced hepatitis, sarcoidosis, cholestasis, or no biopsy was performed. |  | Centre hospitalier de l’Universite´ de Montreál and McGill University Health Centre | Between October 2014 and September 2018 | Quantitative ultrasound(QUS) and Point shear wave elastography (pSWE) | pathology | Random forest |