Ibrahim Miriam Naim, Blázquez-García Raul, Lightstone Adi, Meng Fankun, Bhat Mamatha, El Kaffas Ahmed, Ukwatta Eranga
University of Guelph, Faculty of Engineering, Biomedical Engineering, Guelph, Ontario, Canada.
Oncoustics, Toronto, Ontario, Canada.
J Med Imaging (Bellingham). 2023 May;10(3):034505. doi: 10.1117/1.JMI.10.3.034505. Epub 2023 Jun 5.
Non-alcoholic fatty liver disease (NAFLD) is an increasing global health concern, with a prevalence of 25% worldwide. The rising incidence of NAFLD, an asymptomatic condition, reinforces the need for systematic screening strategies in primary care. We present the use of non-expert acquired point-of-care ultrasound (POCUS) B-mode images for the development of an automated steatosis classification algorithm.
We obtained a Health Insurance Portability and Accountability Act compliant dataset consisting of 478 patients [body mass index , age ], imaged with POCUS by non-expert health care personnel. A U-Net deep learning (DL) model was used for liver segmentation in the POCUS B-mode images, followed by patch extraction of liver parenchyma. Several DL models including VGG-16, ResNet-50, Inception V3, and DenseNet-121 were trained for binary classification of steatosis. All layers of each tested model were unfrozen, and the final layer was replaced with a custom classifier. Majority voting was applied for patient-level results.
On a hold-out test set of 81 patients, the final DenseNet-121 model yielded an area under the receiver operator characteristic curve of 90.1%, sensitivity of 95.0%, and specificity of 85.2% for the detection of liver steatosis. Average cross-validation performance in models using patches of liver parenchyma as input outperformed methods using complete B-mode frames.
Despite minimal POCUS acquisition training, and low-quality B-mode images, it is possible to detect steatosis using DL algorithms. Implementation of this algorithm in POCUS software may offer an accessible, low-cost steatosis screening technology, for use by non-expert health care personnel.
非酒精性脂肪性肝病（NAFLD）是一个日益引起全球健康关注的问题，在全球的患病率为25%。NAFLD作为一种无症状疾病，其发病率不断上升，这凸显了在初级保健中采用系统筛查策略的必要性。我们展示了如何利用非专业人员获取的即时超声（POCUS）B模式图像来开发一种自动脂肪变性分类算法。
我们获得了一个符合《健康保险流通与责任法案》的数据集，该数据集由478名患者[体重指数，年龄]组成，由非专业医护人员使用POCUS进行成像。使用U-Net深度学习（DL）模型对POCUS B模式图像中的肝脏进行分割，随后对肝实质进行图像块提取。包括VGG-16、ResNet-50、Inception V3和DenseNet-121在内的几种DL模型被训练用于脂肪变性的二元分类。对每个测试模型的所有层进行解冻，并将最后一层替换为自定义分类器。对患者水平的结果应用多数投票法。
在一个由81名患者组成的保留测试集上，最终的DenseNet-121模型在检测肝脏脂肪变性时，受试者操作特征曲线下面积为90.1%，灵敏度为95.0%，特异性为85.2%。使用肝实质图像块作为输入的模型的平均交叉验证性能优于使用完整B模式帧的方法。
尽管POCUS采集培训最少且B模式图像质量较低，但使用DL算法仍有可能检测出脂肪变性。在POCUS软件中实施该算法可能会提供一种易于使用、低成本的脂肪变性筛查技术，供非专业医护人员使用。
