DEEPNOID Inc., Seoul, Korea.
Department of Urology, Chung-Ang University Gwangmyeong Hospital, Chung-Ang University College of Medicine, Gyeonggi-do, Korea.
J Endourol. 2024 Oct;38(10):1036-1043. doi: 10.1089/end.2024.0250. Epub 2024 Jul 1.
An investigation of various convolutional neural network (CNN)-based deep learning algorithms was conducted to select the appropriate artificial intelligence (AI) model for calculating the diagnostic performance of bladder tumor classification on cystoscopy images, with the performance of the selected model to be compared against that of medical students and urologists. A total of 3,731 cystoscopic images that contained 2,191 tumor images were obtained from 543 bladder tumor cases and 219 normal cases were evaluated. A total of 17 CNN models were trained for tumor classification with various hyperparameters. The diagnostic performance of the selected AI model was compared with the results obtained from urologists and medical students by using the receiver operating characteristic (ROC) curve graph and metrics. EfficientNetB0 was selected as the appropriate AI model. In the test results, EfficientNetB0 achieved a balanced accuracy of 81%, sensitivity of 88%, specificity of 74%, and an area under the curve (AUC) of 92%. In contrast, human-derived diagnostic statistics for the test data showed an average balanced accuracy of 75%, sensitivity of 94%, and specificity of 55%. Specifically, urologists had an average balanced accuracy of 91%, sensitivity of 95%, and specificity of 88%, while medical students had an average balanced accuracy of 69%, sensitivity of 94%, and specificity of 44%. Among the various AI models, we suggest that EfficientNetB0 is an appropriate AI classification model for determining the presence of bladder tumors in cystoscopic images. EfficientNetB0 showed the highest performance among several models and showed high accuracy and specificity compared to medical students. This AI technology will be helpful for less experienced urologists or nonurologists in making diagnoses. Image-based deep learning classifies bladder cancer using cystoscopy images and shows promise for generalized applications in biomedical image analysis and clinical decision making.
对各种基于卷积神经网络（CNN）的深度学习算法进行了调查，以选择合适的人工智能（AI）模型来计算膀胱肿瘤分类在膀胱镜图像上的诊断性能，所选择的模型的性能将与医学生和泌尿科医生的性能进行比较。从 543 例膀胱肿瘤病例和 219 例正常病例中获得了包含 2191 个肿瘤图像的 3731 张膀胱镜图像。使用各种超参数对 17 个 CNN 模型进行了肿瘤分类训练。使用接收器操作特征（ROC）曲线和指标比较了所选 AI 模型的诊断性能与泌尿科医生和医学生的结果。选择了 EfficientNetB0 作为合适的 AI 模型。在测试结果中，EfficientNetB0 达到了 81%的平衡准确率、88%的敏感度、74%的特异性和 92%的曲线下面积（AUC）。相比之下，对于测试数据，人类诊断统计数据的平均平衡准确率为 75%、敏感度为 94%和特异性为 55%。具体而言，泌尿科医生的平均平衡准确率为 91%、敏感度为 95%和特异性为 88%，而医学生的平均平衡准确率为 69%、敏感度为 94%和特异性为 44%。在各种 AI 模型中，我们建议 EfficientNetB0 是一种用于确定膀胱镜图像中膀胱肿瘤存在的合适的 AI 分类模型。EfficientNetB0 在几个模型中表现最好，与医学生相比，其准确性和特异性都很高。这项 AI 技术将有助于经验较少的泌尿科医生或非泌尿科医生进行诊断。基于图像的深度学习使用膀胱镜图像对膀胱癌进行分类，有望在生物医学图像分析和临床决策制定的广泛应用中得到推广。
