使用头骨CT预测跨超声波插入损失:一种深度学习方法
Ning Wang1, Han Li1, Jinpeng Liao1
1School of Physics, Engineering and Technology, University of York, UK.
Ultrasonics
|February 4, 2026
概括
深度学习使用CT扫描准确地预测了通过头骨的超声波信号损失. 这种方法比传统的模拟更快,可以精确控制跨膜超声波疗法.
科学领域:
- 生物医学工程 生物医学工程
- 神经科学是一个神经科学.
- 医疗成像医学成像
背景情况:
- 穿超声波 (tUS) 提供非侵入性大脑调制,但由于头骨诱导的超声波衰减而面临挑战.
- 插入损失 (IL) 量化了这种信号退化,对于有效的tUS传输至关重要.
- 目前的IL预测方法是计算密集型,对变化敏感.
研究的目的:
- 根据头骨结构特征开发一种快速而准确的方法来预测IL.
- 为了研究头骨解剖学和超声波衰减之间的相关性.
- 在tUS应用中利用深度学习进行高效的IL预测.
主要方法:
- 从20个人类头骨样本收集了220kHz,650kHz和1000kHz的IL数据.
- 通过使用头骨CT扫描开发了一种修改的基于Inception的双路径神经网络 (mDPI-Net).
- 将mDPI-Net的性能与同质的伪光谱方法和异质的模拟进行比较.
主要成果:
- 在精度方面,mDPI-Net显著优于同质方法 (峰值压力误差:26.6%与34.3%).
- mDPI-Net显示了与复杂模拟相比的准确性 (IL偏差:2.47 dB与1.69 dB相比).
- 计算效率大大提高,预测时间从15分钟/样本减少到0.5秒/样本.
结论:
- 头骨CT扫描包含内在的结构信息,对于IL预测至关重要.
- 像mDPI-Net这样的深度学习模型为IL预测提供了计算效率高,准确的方法.
- 这种技术具有实时手术前IL评估的潜力,提高了tUS治疗的精度.
相关概念视频
Sutures of the Skull
10.5K
The human skull is composed of several bones that come together to protect the brain and support the structures of the face. The junctions where these bones meet are called sutures.
Sutures are immobile joints between adjacent bones of the skull. The narrow gap between the bones is filled with dense, fibrous connective tissue that unites the bones. The long sutures located between the skull bones are not straight but instead follow irregular, tightly twisting paths. These twisting lines tightly...
Sutures are immobile joints between adjacent bones of the skull. The narrow gap between the bones is filled with dense, fibrous connective tissue that unites the bones. The long sutures located between the skull bones are not straight but instead follow irregular, tightly twisting paths. These twisting lines tightly...
10.5K
Line Loss
545
The different configurations of source-load connections include wye (star) and delta connections. The relationship between line and phase voltages and currents varies depending on the configuration. When the source is supplying power, it is transmitted through the wires to the load, and during this transmission, some power is absorbed by the wires, leading to line loss.
Line loss impacts power delivery efficiency in a balanced three-phase circuit. The symmetry in such a circuit simplifies the...
Line loss impacts power delivery efficiency in a balanced three-phase circuit. The symmetry in such a circuit simplifies the...
545
Overview of the Skull
7.9K
The cranium (skull) is the skeletal structure of the head that supports the face and protects the brain. It is subdivided into the facial bones and the brain case, or cranial vault. The facial bones underlie the facial structures, form the nasal cavity, enclose the eyeballs, and support the teeth of the upper and lower jaws.
The cranial vault surrounds and protects the brain and houses the middle and inner ear structures. This cavity is bounded superiorly by the rounded top of the skull, which...
The cranial vault surrounds and protects the brain and houses the middle and inner ear structures. This cavity is bounded superiorly by the rounded top of the skull, which...
7.9K
Predicting Molecular Geometry
46.0K
VSEPR Theory for Determination of Electron Pair Geometries
46.0K
Reducing Line Loss
390
In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss in...
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss in...
390
Prediction Intervals
3.4K
The interval estimate of any variable is known as the prediction interval. It helps decide if a point estimate is dependable.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
3.4K


