Karunanayake Nalan, Lu Lin, Yang Hao, Geng Pengfei, Akin Oguz, Furberg Helena, Schwartz Lawrence H, Zhao Binsheng
Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, NY 10017, USA.
Tomography. 2025 Jan 3;11(1):3. doi: 10.3390/tomography11010003.
Accurate kidney and tumor segmentation of computed tomography (CT) scans is vital for diagnosis and treatment, but manual methods are time-consuming and inconsistent, highlighting the value of AI automation. This study develops a fully automated AI model using vision transformers (ViTs) and convolutional neural networks (CNNs) to detect and segment kidneys and kidney tumors in Contrast-Enhanced (CECT) scans, with a focus on improving sensitivity for small, indistinct tumors.
The segmentation framework employs a ViT-based model for the kidney organ, followed by a 3D UNet model with enhanced connections and attention mechanisms for tumor detection and segmentation. Two CECT datasets were used: a public dataset (KiTS23: 489 scans) and a private institutional dataset (Private: 592 scans). The AI model was trained on 389 public scans, with validation performed on the remaining 100 scans and external validation performed on all 592 private scans. Tumors were categorized by TNM staging as small (≤4 cm) (KiTS23: 54%, Private: 41%), medium (>4 cm to ≤7 cm) (KiTS23: 24%, Private: 35%), and large (>7 cm) (KiTS23: 22%, Private: 24%) for detailed evaluation.
Kidney and kidney tumor segmentations were evaluated against manual annotations as the reference standard. The model achieved a Dice score of 0.97 ± 0.02 for kidney organ segmentation. For tumor detection and segmentation on the KiTS23 dataset, the sensitivities and average false-positive rates per patient were as follows: 0.90 and 0.23 for small tumors, 1.0 and 0.08 for medium tumors, and 0.96 and 0.04 for large tumors. The corresponding Dice scores were 0.84 ± 0.11, 0.89 ± 0.07, and 0.91 ± 0.06, respectively. External validation on the private data confirmed the model's effectiveness, achieving the following sensitivities and average false-positive rates per patient: 0.89 and 0.15 for small tumors, 0.99 and 0.03 for medium tumors, and 1.0 and 0.01 for large tumors. The corresponding Dice scores were 0.84 ± 0.08, 0.89 ± 0.08, and 0.92 ± 0.06.
The proposed model demonstrates consistent and robust performance in segmenting kidneys and kidney tumors of various sizes, with effective generalization to unseen data. This underscores the model's significant potential for clinical integration, offering enhanced diagnostic precision and reliability in radiological assessments.
计算机断层扫描（CT）扫描中准确的肾脏和肿瘤分割对于诊断和治疗至关重要，但手动方法既耗时又不一致，凸显了人工智能自动化的价值。本研究开发了一种使用视觉变换器（ViT）和卷积神经网络（CNN）的全自动人工智能模型，用于在对比增强（CECT）扫描中检测和分割肾脏及肾脏肿瘤，重点在于提高对小的、边界不清的肿瘤的敏感性。
分割框架采用基于ViT的模型进行肾脏器官分割，随后是具有增强连接和注意力机制的3D UNet模型用于肿瘤检测和分割。使用了两个CECT数据集：一个公共数据集（KiTS23：489次扫描）和一个私人机构数据集（私人：592次扫描）。人工智能模型在389次公共扫描上进行训练，在其余100次扫描上进行验证，并在所有592次私人扫描上进行外部验证。肿瘤根据TNM分期分为小（≤4 cm）（KiTS23：54%，私人：41%）、中（>4 cm至≤7 cm）（KiTS23：24%，私人：35%）和大（>7 cm）（KiTS23：22%，私人：24%）以进行详细评估。
以手动标注作为参考标准对肾脏和肾脏肿瘤分割进行评估。该模型在肾脏器官分割方面的Dice评分为0.97±0.02。对于KiTS23数据集上的肿瘤检测和分割，每个患者的敏感性和平均假阳性率如下：小肿瘤为0.90和0.23，中肿瘤为1.0和0.08，大肿瘤为0.96和0.04。相应的Dice评分分别为0.84±0.11、0.89±0.07和0.91±0.06。对私人数据的外部验证证实了该模型的有效性，每个患者的敏感性和平均假阳性率如下：小肿瘤为0.89和0.15，中肿瘤为0.99和0.03，大肿瘤为1.0和0.01。相应的Dice评分分别为0.84±0.08、0.89±0.08和0.92±0.06。
所提出的模型在分割各种大小的肾脏和肾脏肿瘤方面表现出一致且稳健的性能，并能有效地推广到未见过的数据。这突出了该模型在临床整合方面的巨大潜力，在放射学评估中提供了更高的诊断精度和可靠性。
