深度学习架构的比较分析,用于全景牙科放射图中的自动牙细分:平衡精度和计算效率
Alperen Yalım1, Emre Aytugar1, Fahrettin Kalabalık2
1Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Izmir Katip Celebi University, Izmir 35640, Turkey.
Diagnostics (Basel, Switzerland)
|January 28, 2026
概括
这项研究对U-Net深度学习模型进行了基准测试,用于牙科放射图中的牙细分. EfficientNet-B0为此任务提供了高精度和低计算成本的最佳平衡.
科学领域:
- 医疗成像中的人工智能
- 用于医学图像分析的深度学习
- 牙科放射学和诊断 牙科
背景情况:
- 在全景牙科放射图中,自动牙细分对于诊断准确性至关重要.
- 用各种编码器骨干评估基于U-Net的深度学习模型对于优化性能和效率至关重要.
研究的目的:
- 系统地对U-Net深度学习模型进行基准测试,用于全景牙科放射图中的牙自动细分.
- 分析不同编码器骨干 (ResNet,EfficientNet,DenseNet,MobileNetV3-Small) 的细分精度和计算成本之间的权衡.
主要方法:
- 在塔夫斯牙科数据库 (1000张图像) 上评估了预先训练有素的ImageNet编码器家族的U-Net模型.
- 采用了五重交叉验证策略.
- 用子系数和IoU测量细分性能;通过参数数量和GFLOPs评估计算效率.
主要成果:
- 整体细分质量很高 (Dice:0.9168-0.9259),由于骨干复杂度增加,回报率下降.
- EfficientNet-B0实现了高精度 (Dice: 0.9244 ± 0.0011) 和低计算成本 (5.98 GFLOPs) 的近乎最佳平衡.
- 虽然EfficientNet-B7具有最高的名义准确性,但与EfficientNet-B0和B4.4相比,差异在统计上并不显著.
结论:
- 较大的深度学习模型并不总是为牙细分提供优异的性能.
- EfficientNet-B0被认为是最实用的模型,提供接近和的准确性,模型大小和计算需求显著减少.
- 这些发现强调了选择计算效率高的模型的重要性,而不会影响牙科成像应用中的诊断准确性.
相关概念视频
Uncertainty in Measurement: Accuracy and Precision
100.9K
Scientists typically make repeated measurements of a quantity to ensure the quality of their findings and to evaluate both the precision and the accuracy of their results. Measurements are said to be precise if they yield very similar results when repeated in the same manner. A measurement is considered accurate if it yields a result that is very close to the true or the accepted value. Precise values agree with each other; accurate values agree with a true value.
100.9K
Comparing the Survival Analysis of Two or More Groups
583
Survival analysis is a cornerstone of medical research, used to evaluate the time until an event of interest occurs, such as death, disease recurrence, or recovery. Unlike standard statistical methods, survival analysis is particularly adept at handling censored data—instances where the event has not occurred for some participants by the end of the study or remains unobserved. To address these unique challenges, specialized techniques like the Kaplan-Meier estimator, log-rank test, and...
583
Automatic Processing and Automatic Social Behavior
245
Automatic processing refers to the cognitive operations that occur without conscious intent or awareness, playing a fundamental role in shaping social cognition and behavior. These processes enable individuals to navigate complex social environments efficiently by relying on mental shortcuts and pre-existing knowledge structures known as schemas. One of the most influential mechanisms underlying automatic processing is priming, which subtly activates mental representations through exposure to...
245
Improving Translational Accuracy
14.9K
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
14.9K
Improving Translational Accuracy
3.7K
3.7K
Balancing Redox Equations
61.9K
Electrochemistry is the science involved in the interconversion of electrical and chemical reactions. Such reactions are called reduction-oxidation, or redox reactions. These important reactions are defined by changes in oxidation states for one or more reactant elements and include a subset of reactions involving the transfer of electrons between reactant species. Electrochemistry as a field has evolved to yield sufficient insights on the fundamental principles of redox chemistry and multiple...
61.9K


