Advances in Clinical Medicine
Vol. 13  No. 12 ( 2023 ), Article ID: 76810 , 4 pages
10.12677/ACM.2023.13122653

CT影像组学在胸部疾病诊疗中的 应用新进展

吐尔孙阿依·米吉提

新疆医科大学第一临床医学院,新疆 乌鲁木齐

收稿日期:2023年11月7日;录用日期:2023年12月1日;发布日期:2023年12月11日

摘要

气胸(pneumothorax)系肺组织及脏层胸膜破裂或胸壁及壁层胸膜被穿透空气进入胸膜腔,形成胸膜腔积气和肺组织的压缩。在无创伤或人为情况下,肺组织及脏层胸膜自发破裂,空气进入胸膜腔,导致肺组织受压,引发一系列综合征称为自发性气胸(Spontaneous pneumothorax, SP)。自发性气胸又分为原发性自发性气胸和继发性自发性气胸。原发性自发性气胸(Primary spontaneous pneumothorax, PSP)指肺脏实质或脏层胸膜在无外源性或介入性因素影响以及无基础性肺疾病条件下,自行发生破裂,引起气体在胸膜腔蓄积,男性发病率18~28/10万,而女性发病率要低的多,大约为1.2~6.0/10万;继发性自发性气胸是指继发于肺脏各种疾病,如慢性肺结核、弥漫性肺间质纤维化、肺癌等。原发性自发性气胸的临床症状比较典型,多在休息时发病,症状多为伴或不伴呼吸困难的突发性胸痛。自发性气胸(SP)是临床上比较常见的呼吸系统疾病,呼吸困难、胸痛等是该疾病主要的临床表现,若不及时接受治疗,则可能会损害患者的肺功能,诱发皮下气肿、纵膈气肿、血气胸等并发症,给患者的日常生活质量和工作状态带来不便。严重时还会危及患者的生命,给患者的生命健康安全造成极大的威胁。

关键词

自发性气胸,CT,影像组学,纹理分析

New Progress in the Application of CT Imaging Omics in the Diagnosis and Treatment of Chest Diseases

Mijiti Tuersunayi

The First Clinical Medical College, Xinjiang Medical University, Urumqi Xinjiang

Received: Nov. 7th, 2023; accepted: Dec. 1st, 2023; published: Dec. 11th, 2023

ABSTRACT

Pneumothorax refers to the rupture of the lung tissue and visceral pleura, or the penetration of air into the pleural cavity of the chest wall and parietal pleura, resulting in pneumothorax and compression of lung tissue. Without trauma or artificial circumstances, the lung tissue and visceral pleura spontaneously rupture, air enters the pleural cavity, leading to compression of the lung tissue, which leads to a series of syndromes called spontaneous pneumothorax (SP). Spontaneous pneumothorax can be divided into primary spontaneous pneumothorax and secondary spontaneous pneumothorax. Primary spontaneous pneumothorax (PSP) refers to the spontaneous rupture of the lung parenchyma or visceral pleura without the influence of exogenous or interventional factors and without basic lung diseases, resulting in the accumulation of gas in the pleural cavity. The incidence rate of men is 18~28/100,000, while that of women is much lower, about 1.2~6.0/100,000; Secondary spontaneous pneumothorax refers to various diseases secondary to the lungs, such as chronic pulmonary tuberculosis, diffuse pulmonary interstitial fibrosis, lung cancer, etc. The clinical symptoms of primary spontaneous pneumothorax are typical, which usually occur at rest. The symptoms are usually sudden chest pain with or without dyspnea. Spontaneous pneumothorax (SP) is a common disease in clinic. The main clinical manifestations of SP are dyspnea and chest pain, which bring inconvenience to the quality of daily life and working status of patients. If patients with spontaneous pneumothorax do not receive treatment in time, they may damage their lung function, induce subcutaneous emphysema, mediastinal emphysema, hemopneumothorax and other complications, and even endanger their lives when serious, posing a great threat to their health and safety.

Keywords:Spontaneous Pneumothorax, CT, Imaging Histology, Texture Analysis

Copyright © 2023 by author(s) and Hans Publishers Inc.

This work is licensed under the Creative Commons Attribution International License (CC BY 4.0).

http://creativecommons.org/licenses/by/4.0/

1. 引言

原发性自发性气胸好发于青壮年,该好发人群往往对自身健康充满自信,活动范围广,活动环境甚至可能较复杂,而原发性自发性气胸发病突然,一般没有前期症状,在缺乏必要的医疗条件的环境下,或者因缺乏足够的重视而未及时采取及时的医学干预时,PSP可能危及生命安全。CT纹理特征分析(CT texture analysis, CTTA)是一种新型的CT图像处理技术,可通过定量分析病灶局部图像摘的像素灰度、灰度值的分布模式和变化规律来获取常规CT图像难以识别的信息,从而反映图像的本质差异 [1] 。基于HRCT的影像组学通过分析PSP患者发病前胸部HRCT纹理特征,在其发病前做出准确的预测,通过早期干预,可减少不必要的风险,旨在为广大人民群众的生命安全保驾护航。

2. 影像组学的临床应用

2.1. 影像组学的概念

影像组学的概念最早由荷兰学者在2012年提出,其强调的深层次含义是指从影像(CT、MRI、PET等)中高通量地提取大量影像信息,实现肿瘤分割、特征提取与模型建立,凭借对专海量影像数据信息进行更深层次的挖掘、预测和分析来辅助医师做出最准确的诊断。影像组学可直观地理解为将视觉影像信息转化为深层次的特征来进行量化研究。随着技术的发展,当代医学成像在显示肺野内细微结构上的改变有着明显的优势,并且可以利用后处理技术对肺野内的解剖改变进行定量化分析,使CT成像变得越来越定量化。新兴的CT纹理分析法就是一个典型的例子 [2] 。纹理分析已经成为一种工具,用来探索人类视觉上无法探索的图像所包含的数据量。放射学是一种利用数据特征化算法从医学图像中提取大量特征的方法。这些特征被称为放射学特征,具有揭示疾病特征的潜力 [3] 。而纹理是指在图像某一局部区域上,各个像素由于不同的空间排列,所产生的一种具有特定模式或形态的明暗变化,这种变化可以通过图像处理方法中的纹理分析技术来客观度量 [3] [4] [5] ,即将非结构化数据(如图像)转换为结构化数据(如成像特征),利用后处理技术定量地描述临床有用的成像表型 [2] 。这种方法不需要人为干预,解决了上述各种限制。

2.2. 影像组学对疾病诊断的协助作用

近几年以来,影像组学开始从多方面用于研究更多疾病。Alfonso Reginelli等 [6] 通过纹理分析研究描述放射组学在肺结节表征中的作用。放射组学是计算机辅助诊断的延伸,将CT纹理分析(CTTA)的定量成像数据与临床信息重叠,提高了个性化医疗决策的能力和精度 [7] [8] 。他们回顾性分析了87例非小细胞肺癌患者的CT图像,研究各放射组学特征与肿瘤体积之间的相关性。得出的结论是,单独使用放射组学特征或结合病理数据和CT形态密度评估有可能获得预测的肿瘤特征 [9] [10] [11] 。从而,更一致的测量结果可以减少临床医生在做出重要治疗决定时的不确定性,从而为患者带来更好的结果。

Wibmer等研究报道:通过T2WI与ADC图像纹理分析前列腺周围带五个纹理特征对鉴别前列腺良性及恶性病灶均具有显著性差异。前列腺移形带ADC图中的五个特征值与T2 WI DWI-ADC T2WI中的两个特征值能够鉴别良恶性前列腺病灶。这些特征值用以自动计算前列腺Gleason分级 [12] 。基于纹理分析的机器学习能够鉴别Gleason-6分(3 + 3)与GS-7分患者(准确率达93%),以及鉴别Gleason-7分患者间(3 + 4 vs. 4 + 3)准确率达92% [13] 。

Yoshiharu Ohno等 [14] 开发并引入了一个基于机器学习(ML)的CT纹理分析软件,用于对肺部疾病患者的薄层胸部CT影像进行全自动CT纹理分析。他们认为,与传统图像分析相比,机器学习(ML)和计算机辅助诊断(CAD)的优势在于,它有潜力在没有人类监督的情况下识别临床重要的成像变量,包括至关重要的视觉上无法检测的CT变量。他们回顾性的收集89名患者的36,690片切片,通过三名放射科医生以及软件对其进行验证。最终,他们的研究结果表明,基于ML的CT纹理分析可以提高放射科医师对COPD、SP、间质性肺病或感染性疾病患者胸部薄层CT影像学表现评估时观察者之间的一致性和整体的鉴别准确性。此外,该软件在提高评估的放射结果和放射科医生读取的结果之间的共识上同样具有良好的潜力。这些事实可能为研究者鉴别SP或感染性疾病与正常肺、评估疾病严重程度和肺部疾病患者治疗效果的CT特征评估提供较少的差异。

2.3. 影像组学对疾病预后的预测价值

研究显示,肺癌瘤内异质性通过纹理分析被读取后,这些图像纹理特征能极强地预测早期肺癌患者的预后,提高对肺腺癌肺叶切除术后是否复发评估的准确率 [13] 。该研究发现,CT图像特征值与肺腺癌患者生存时间显著相关,可用来预测肿瘤患者转归 [15] 。轻度认知障碍(MCI)中仅有20%会进展成为AD (进展型MCI):通过全自动分割软件分割受试者T1WI图像中海马结构,并提取海马的图像纹理特征值。特征值越高代表海马结构内异质性越强。能够有效鉴别进展型MCI与稳定型MCI [16] 。

PSP虽然较为常见,但发病突然,一般没有前期症状,当双侧同时发病或形成张力性气胸时,或者在缺乏必要的医疗条件的环境下,PSP可能危及生命安全 [17] 。然而,目前关于影像组学在PSP的预测领域的应用研究鲜有公开报道。本研究拟通过对PSP患者HRCT图像进行纹理分析,探讨其在预测PSP发病的临床价值,从而达到早期医学干预的目的,为今后原发性自发性气胸的准确预测提供科学依据。

文章引用

吐尔孙阿依·米吉提. CT影像组学在胸部疾病诊疗中的应用新进展
New Progress in the Application of CT Imaging Omics in the Diagnosis and Treatment of Chest Diseases[J]. 临床医学进展, 2023, 13(12): 18860-18863. https://doi.org/10.12677/ACM.2023.13122653

参考文献

  1. 1. 唐彩银, 李瑗, 张继, 等. CT纹理分析在肾脏透明细胞癌分级的临床应用[J]. 医学理论与实践, 2019, 32(21): 3416-3418, 3409.

  2. 2. Mir, A.H., hanmandlu, M. and Tandon, S.N. Texture Analysis of CT Images. IEEE Engineering in Medicine and Biology Magazine, 14, 781-786. https://doi.org/10.1109/51.473275

  3. 3. Corrias, G., Micheletti, G., Barberini, L., Suri, J.S. and Saba, L. (2022) Texture Analysis Imaging “What a Clinical Radiologist Needs to Know”. European Journal of Radiology, 146, Article ID: 110055. https://doi.org/10.1016/j.ejrad.2021.110055

  4. 4. Tuceryan, M. and Jain, A.K. (1998) Texture Analysis. In: Chen, C.H. and Pau, L.F., Eds., Handbook of Pattern Recognition and Computer Vision, World Scientific Publishing Co., 207-248. https://doi.org/10.1142/9789812384737_0007

  5. 5. Uppaluri, R., Mitsa, T., Sonka, M., Hoffman, E.A. and McLennan, G. (1997) Quantification of Pulmonary Emphysema from Lung Computed Tomography Images. Ameri-can Journal of Respiratory and Critical Care Medicine, 156, 248-254. https://doi.org/10.1164/ajrccm.156.1.9606093

  6. 6. Reginelli, A., Belfiore, M.P., Monti, R., Cozzolino, I., Costa, M., Vicidomini, G., Grassi, R., Morgillo, F., Urraro, F., Nardone, V. and Cappabianca, S. (2020) The Texture Analysis as a Predictive Method in the Assessment of the Cytological Specimen of CT-Guided FNAC of the Lung Cancer. Medi-cal Oncology, 37, Article Number: 54. https://doi.org/10.1007/s12032-020-01375-9

  7. 7. Brunese, L., Mercaldo, F., Reginelli, A. and Santone, A. (2020) An Ensemble Learning Approach for Brain Cancer Detection Exploiting Radiomic Features. Computer Methods and Programs in Biomedicine, 185, Article ID: 105134. https://doi.org/10.1016/j.cmpb.2019.105134

  8. 8. Brunese, L., Mercaldo, F., Reginelli, A. and Santone, A. (2020) Formal Methods for Prostate Cancer Gleason Score and Treatment Prediction Using Radiomic Biomarkers. Magnetic Resonance Imaging, 66, 165-175. https://doi.org/10.1016/j.mri.2019.08.030

  9. 9. Gillies, R.J., Kinahan, P.E. and Hricak, H. (2015) Radiomics: Im-ages Are More than Pictures, They Are Data. Radiology, 278, 563-577.

  10. 10. Müller, N.L., Staples, C.A., Miller, R.R. and Abboud, R.T. (1988) Density Mask. An Objective Method to Quantitate Emphysema Using Computed Tomography. Chest, 94, 782-787. https://doi.org/10.1378/chest.94.4.782

  11. 11. Bankier, A.A., Maertelaer, V.D., Keyzer, C. and Gevenois, P.A. (1999) Pulmonary Emphysema: Subjective Visual Grading versus Objective Quantification with Macro-scopic Morphometry and Thin-Section CT Densitometry. Radiology, 211, 851-858. https://doi.org/10.1148/radiology.211.3.r99jn05851

  12. 12. Wibmer, A., Hricak, H., Gondo, T., et al. (2015) Haralick Texture Analysis of Prostate MRl: Utility for Differentiating Non-Cancerous Prostate from Prostate Cancer and Differen-tiating with Different Gleason Scores. European Radiology, 25, 2840-2850. https://doi.org/10.1007/s00330-015-3701-8

  13. 13. Fehr, D., Veeraraghavanm H., Wibmer, A., et al. (2015) Auto-matic Classification of Prostate Cancer Gleason Sores from Multi-Parametric Magnetic Resonance Images. Proceedings of the National Academy of Sciences of the United States of America, 112, E6265-E6273.

  14. 14. Ohno, Y., Aoyagi, K., Takenaka, D., Yoshikawa, T., Ikezaki, A., Fujisawa, Y., Murayama, K., Hattori, H. and Toyama, H. (2021) Machine Learning for Lung CT Texture Analysis: Improvement of Inter-Observer Agreement for Radiological Finding Classifica-tion in Patients with Pulmonary Diseases. European Journal of Radiology, 134, Article ID: 109410. https://doi.org/10.1016/j.ejrad.2020.109410

  15. 15. Grove, O., Berglund, A.E., Schabath, M.B., Aerts, H.J.W.L., Dekker, A., et al. (2021) Correction: Quantitative Computed Tomographic Descriptors Associate Tumor Shape Com-plexity and Intratumor Heterogeneity with Prognosis in Lung Adenocarcinoma. PLOS ONE, 16, e0248541. https://doi.org/10.1371/journal.pone.0118261

  16. 16. Sørensen, L., Igel, C., Hansen, N.L., et al. (2016) Early Detec-tion of Alzheimer’s Disease Using MRI Hippocampal Texture. Human Brain Mapping, 37, 1148-1161. https://doi.org/10.1002/hbm.23091

  17. 17. 王朝, 邹卫. 原发性自发性气胸病因研究进展[J]. 临床肺科杂志, 2015, 20(6): 1120-1122, 1126.

期刊菜单