卷积神经网络从b超图像中对肝脂肪变性进行分类的诊断准确性：在印度泰伦加纳社区环境中进行的荟萃分析和新验证的系统回顾

IF 6.2 Q1 HEALTH CARE SCIENCES & SERVICES

The Lancet regional health. Southeast Asia Pub Date : 2025-07-31 DOI:10.1016/j.lansea.2025.100644

Akshay Jagadeesh , Chanchanok Aramrat , Santosh Rai , Fathima Hana Maqsood , Adarsh Kibballi Madhukeshwar , Santhi Bhogadi , Judith Lieber , Hemant Mahajan , Santosh Kumar Banjara , Alexandra Lewin , Sanjay Kinra , Poppy Mallinson

{"title":"卷积神经网络从b超图像中对肝脂肪变性进行分类的诊断准确性：在印度泰伦加纳社区环境中进行的荟萃分析和新验证的系统回顾","authors":"Akshay Jagadeesh , Chanchanok Aramrat , Santosh Rai , Fathima Hana Maqsood , Adarsh Kibballi Madhukeshwar , Santhi Bhogadi , Judith Lieber , Hemant Mahajan , Santosh Kumar Banjara , Alexandra Lewin , Sanjay Kinra , Poppy Mallinson","doi":"10.1016/j.lansea.2025.100644","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Ultrasound is a widely available, inexpensive, and non-invasive modality for evaluating hepatic steatosis (HS). However, the scarcity of radiological expertise limits its utility. Convolutional Neural Networks (CNNs) have potential for automated classification of HS using B-mode ultrasound images. We aimed to assess their diagnostic accuracy and generalisability across diverse study settings and populations.</div></div><div><h3>Methods</h3><div>We systematically reviewed two biomedical databases up to Dec 12, 2023, to identify studies that applied CNNs in the classification of HS using B-mode ultrasound images as input (PROSPERO: CRD42024501483). We supplemented this review with a novel analysis of the community-based Andhra Pradesh Children and Parents’ Study (APCAPS) in India to address the overrepresentation of hospital samples and lack of data on South Asian populations who exhibit a distinct central adiposity phenotype that could influence CNN performance. We quantitatively synthesised diagnostic accuracy metrics for eligible studies using random-effects meta-analyses.</div></div><div><h3>Findings</h3><div>Our search returned 289 studies, of which 17 were eligible. All but one of the 17 studies were based in hospital or clinical outpatient settings with curated cases and controls. Studies were conducted exclusively in East Asian, European, or North American populations. Studies employed varying gold standards: seven studies (41.18%) used liver biopsy, three (17.64%) used MRI proton density fat fraction, and seven (41.18%) used clinician-evaluated ultrasound-based HS grades. The APCAPS sample included 219 participants with radiologist-assigned HS grades. Across the range of study settings and populations, CNNs demonstrated good diagnostic accuracy. Meta-analysis of studies with low risk of bias reporting on five unique datasets showed a pooled area under the receiver operating characteristic curve of 0.93 (95% CI 0.73–0.98) for detecting any severity and 0.86 (95% CI 0.77–0.92) for detecting moderate-to-severe HS severity grades, respectively.</div></div><div><h3>Interpretation</h3><div>CNNs have good diagnostic accuracy and generalisability for HS classification, suggesting potential for real-world application.</div></div><div><h3>Funding</h3><div><span>Medical Research Council</span>, UK (<span><span>MR/T038292/1</span></span>, <span><span>MR/V001221/1</span></span>).</div></div>","PeriodicalId":75136,"journal":{"name":"The Lancet regional health. Southeast Asia","volume":"40 ","pages":"Article 100644"},"PeriodicalIF":6.2000,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Diagnostic accuracy of convolutional neural networks in classifying hepatic steatosis from B-mode ultrasound images: a systematic review with meta-analysis and novel validation in a community setting in Telangana, India\",\"authors\":\"Akshay Jagadeesh , Chanchanok Aramrat , Santosh Rai , Fathima Hana Maqsood , Adarsh Kibballi Madhukeshwar , Santhi Bhogadi , Judith Lieber , Hemant Mahajan , Santosh Kumar Banjara , Alexandra Lewin , Sanjay Kinra , Poppy Mallinson\",\"doi\":\"10.1016/j.lansea.2025.100644\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Background</h3><div>Ultrasound is a widely available, inexpensive, and non-invasive modality for evaluating hepatic steatosis (HS). However, the scarcity of radiological expertise limits its utility. Convolutional Neural Networks (CNNs) have potential for automated classification of HS using B-mode ultrasound images. We aimed to assess their diagnostic accuracy and generalisability across diverse study settings and populations.</div></div><div><h3>Methods</h3><div>We systematically reviewed two biomedical databases up to Dec 12, 2023, to identify studies that applied CNNs in the classification of HS using B-mode ultrasound images as input (PROSPERO: CRD42024501483). We supplemented this review with a novel analysis of the community-based Andhra Pradesh Children and Parents’ Study (APCAPS) in India to address the overrepresentation of hospital samples and lack of data on South Asian populations who exhibit a distinct central adiposity phenotype that could influence CNN performance. We quantitatively synthesised diagnostic accuracy metrics for eligible studies using random-effects meta-analyses.</div></div><div><h3>Findings</h3><div>Our search returned 289 studies, of which 17 were eligible. All but one of the 17 studies were based in hospital or clinical outpatient settings with curated cases and controls. Studies were conducted exclusively in East Asian, European, or North American populations. Studies employed varying gold standards: seven studies (41.18%) used liver biopsy, three (17.64%) used MRI proton density fat fraction, and seven (41.18%) used clinician-evaluated ultrasound-based HS grades. The APCAPS sample included 219 participants with radiologist-assigned HS grades. Across the range of study settings and populations, CNNs demonstrated good diagnostic accuracy. Meta-analysis of studies with low risk of bias reporting on five unique datasets showed a pooled area under the receiver operating characteristic curve of 0.93 (95% CI 0.73–0.98) for detecting any severity and 0.86 (95% CI 0.77–0.92) for detecting moderate-to-severe HS severity grades, respectively.</div></div><div><h3>Interpretation</h3><div>CNNs have good diagnostic accuracy and generalisability for HS classification, suggesting potential for real-world application.</div></div><div><h3>Funding</h3><div><span>Medical Research Council</span>, UK (<span><span>MR/T038292/1</span></span>, <span><span>MR/V001221/1</span></span>).</div></div>\",\"PeriodicalId\":75136,\"journal\":{\"name\":\"The Lancet regional health. Southeast Asia\",\"volume\":\"40 \",\"pages\":\"Article 100644\"},\"PeriodicalIF\":6.2000,\"publicationDate\":\"2025-07-31\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"The Lancet regional health. Southeast Asia\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2772368225001155\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"HEALTH CARE SCIENCES & SERVICES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"The Lancet regional health. Southeast Asia","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2772368225001155","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"HEALTH CARE SCIENCES & SERVICES","Score":null,"Total":0}

引用次数: 0

摘要

超声是一种广泛可用、廉价且无创的肝脂肪变性（HS）评估方法。然而，放射学专业知识的匮乏限制了它的实用性。卷积神经网络（cnn）具有利用b超图像自动分类HS的潜力。我们的目的是评估它们在不同研究环境和人群中的诊断准确性和普遍性。方法我们系统地回顾了截至2023年12月12日的两个生物医学数据库，以识别将cnn用于HS分类的研究，这些研究使用b超图像作为输入（PROSPERO: CRD42024501483）。我们对以社区为基础的印度安得拉邦儿童和父母研究（APCAPS）进行了一项新的分析，以补充这一综述，以解决医院样本的过度代表性和南亚人群数据的缺乏，这些人群表现出明显的中心肥胖表型，可能影响CNN的表现。我们使用随机效应荟萃分析定量地综合了符合条件的研究的诊断准确性指标。我们检索到289项研究，其中17项符合条件。17项研究中，除了一项研究外，其余研究都是在医院或临床门诊环境中进行的，有精心策划的病例和对照。研究仅在东亚、欧洲或北美人群中进行。研究采用不同的金标准：7项研究（41.18%）使用肝活检，3项研究（17.64%）使用MRI质子密度脂肪分数，7项研究（41.18%）使用临床评估的基于超声的HS分级。APCAPS样本包括219名具有放射科医生指定HS等级的参与者。在研究设置和人群的范围内，cnn表现出良好的诊断准确性。对五个独特数据集报告的低偏倚风险研究的荟萃分析显示，检测任何严重程度的受试者工作特征曲线下的合并面积为0.93 (95% CI 0.73-0.98)，检测中度至重度HS严重等级的受试者工作特征曲线下的合并面积为0.86 （95% CI 0.77-0.92）。cnn对HS分类具有良好的诊断准确性和通用性，具有实际应用的潜力。资助医学研究理事会，英国（MR/T038292/1, MR/V001221/1）。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Diagnostic accuracy of convolutional neural networks in classifying hepatic steatosis from B-mode ultrasound images: a systematic review with meta-analysis and novel validation in a community setting in Telangana, India

Background

Ultrasound is a widely available, inexpensive, and non-invasive modality for evaluating hepatic steatosis (HS). However, the scarcity of radiological expertise limits its utility. Convolutional Neural Networks (CNNs) have potential for automated classification of HS using B-mode ultrasound images. We aimed to assess their diagnostic accuracy and generalisability across diverse study settings and populations.

Methods

We systematically reviewed two biomedical databases up to Dec 12, 2023, to identify studies that applied CNNs in the classification of HS using B-mode ultrasound images as input (PROSPERO: CRD42024501483). We supplemented this review with a novel analysis of the community-based Andhra Pradesh Children and Parents’ Study (APCAPS) in India to address the overrepresentation of hospital samples and lack of data on South Asian populations who exhibit a distinct central adiposity phenotype that could influence CNN performance. We quantitatively synthesised diagnostic accuracy metrics for eligible studies using random-effects meta-analyses.

Findings

Our search returned 289 studies, of which 17 were eligible. All but one of the 17 studies were based in hospital or clinical outpatient settings with curated cases and controls. Studies were conducted exclusively in East Asian, European, or North American populations. Studies employed varying gold standards: seven studies (41.18%) used liver biopsy, three (17.64%) used MRI proton density fat fraction, and seven (41.18%) used clinician-evaluated ultrasound-based HS grades. The APCAPS sample included 219 participants with radiologist-assigned HS grades. Across the range of study settings and populations, CNNs demonstrated good diagnostic accuracy. Meta-analysis of studies with low risk of bias reporting on five unique datasets showed a pooled area under the receiver operating characteristic curve of 0.93 (95% CI 0.73–0.98) for detecting any severity and 0.86 (95% CI 0.77–0.92) for detecting moderate-to-severe HS severity grades, respectively.