在多参数MRI中检测腹腔皮层淋巴结
Tejas Sudharshan Mathai1, Thomas C Shen1, Daniel C Elton1
1Imaging Biomarkers and Computer-Aided Diagnosis Laboratory, Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, USA.
概括
一个新的计算机辅助检测管道在多参数MRI扫描中准确识别淋巴结. 这种方法提高了检测率,有助于癌症分期和减少放射科医生的工作量.
科学领域:
- 放射学和医学成像学 医学成像学
- 人工智能在医学中的应用
- 在瘤学瘤学.
背景情况:
- 在多参数MRI (mpMRI) 中精确的淋巴结 (LN) 定位对于淋巴腺病的评估和转移性疾病的分期至关重要.
- 在mpMRI中识别LN是具有挑战性的,因为它们在不同的序列 (例如,T2脂肪抑制,扩散权重成像) 中的外观和信号强度各不相同.
- 放射科医生可能会因为工作流需求和成像复杂性而错过转移性LNs.
研究的目的:
- 开发和评估计算机辅助检测 (CAD) 管道,用于在mpMRI研究中识别良性和恶性淋巴结.
- 提高淋巴结检测的效率和准确性,以及随后的癌症分期测量.
主要方法:
- 利用动态头部 (DyHead) 神经网络在各种mpMRI扫描仪和协议中检测LN.
- 采用了选择性增强技术,Intra-Label LISA (ILL),将T2脂肪抑制和扩散权重成像序列混合在一起.
- 使用来自西门子Aera扫描仪的数据训练了该模型,并使用来自西门子Verio,西门子Biograph mMR和飞利浦Achieva扫描仪的数据测试了其在销售之外的稳定性.
主要成果:
- 通过使用ILL技术,在每体积 (FP/体积) 4个错误阳性时,达到53.5%的平均平均精度 (mAP) 和大约78%的灵敏度.
- 与现有的LN检测方法相比,在4FP/体积下显示了≥10%的mAP和≥12%的灵敏度的显著改善.
- 在不同扫描仪制造商和型号中证实了DyHead模型在销售之外的稳定性.
结论:
- 拟议的CAD管道,包括DyHead和ILL,在mpMRI中有效检测淋巴结.
- 这种自动化方法比目前的方法有了显著的改进,有助于更可靠的癌症分期.
- 代表了在医学成像中实现淋巴结自动检测,细分和分类的重大进展.
更多相关视频
12:45Robot-assisted Total Mesorectal Excision and Lateral Pelvic Lymph Node Dissection for Locally Advanced Middle-low Rectal Cancer
Published on: February 12, 2022
5.7K
08:18Analysis of Lymph Node Volume by Ultra-High-Frequency Ultrasound Imaging in the Braf/Pten Genetically Engineered Mouse Model of Melanoma
Published on: September 8, 2021
2.9K
相关概念视频
01:24Magnetic Resonance Imaging
5.1K
Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique based on a phenomenon of nuclear physics discovered in the 1930s, in which matter exposed to magnetic fields and radio waves was found to emit radio signals. In 1970, a physician and researcher named Raymond Damadian noticed that malignant (cancerous) tissue gave off different signals than normal body tissue. He applied for a patent for the first MRI scanning device in clinical use by the early 1980s. The early MRI...
5.1K
01:23Assessment of the Abdomen III: Palpation
597
Palpation is a crucial tactile examination method for assessing abdominal organs and detecting conditions like tenderness, distention, masses, or fluid. It involves both light and deep palpation techniques, each serving specific diagnostic purposes. Light palpation helps identify tenderness and other surface-level indicators, while deep palpation locates and assess abdominal masses and organ boundaries. A skilled professional can gather valuable insights through palpation, including evaluating...
597
01:19Abdominal Regions and Quadrants
8.2K
To promote clear communication, for instance, about the location of a patient's abdominal pain or a suspicious mass, anatomists and clinicians typically use imaginary lines to categorize the abdominopelvic cavity into either four quadrants or nine regions to identify organs in the cavity.
The simpler quadrants approach, which is more commonly used in medicine, subdivides the cavity with one horizontal and one vertical line that intersects at the patient's umbilicus (navel). The four...
The simpler quadrants approach, which is more commonly used in medicine, subdivides the cavity with one horizontal and one vertical line that intersects at the patient's umbilicus (navel). The four...
8.2K