Deep learning for volumetric assessment of traumatic cerebral hematoma

Diyou CHEN; Xinyi SHI; Pengfei WU; Li ZHAN; Wenbing ZHAO; Jingru XIE; Liang ZHANG; Hui ZHAO

doi:10.16016/j.2097-0927.202311033

您当前的位置：

首页 >

文章列表页 >

Deep learning for volumetric assessment of traumatic cerebral hematoma

Clinical Medicine | 更新时间：2026-01-09

- Deep learning for volumetric assessment of traumatic cerebral hematoma
- Journal of Army Medical University Vol. 46, Issue 19, Pages: 2225-2235(2024)
- 作者机构：
  
  1.400042 重庆，陆军特色医学中心放射科
  2.400042 重庆，陆军特色医学中心军事交通伤防治研究室
  3.401135 重庆，重庆理工大学两江人工智能学院
- 作者简介：
- 基金信息：
  
  the Innovation and Development Joint Fund of Natural Science Foundation of Chongqing(CSTB2023NSCQ-LZX0155)
- DOI：10.16016/j.2097-0927.202311033
  CLC： R319;R445.2;R743.34
- Received：07 November 2023，
  
  Revised：2024-05-10，
  
  Published：15 October 2024
- 稿件说明：
移动端阅览
Diyou CHEN, Xinyi SHI, Pengfei WU, et al. Deep learning for volumetric assessment of traumatic cerebral hematoma[J]. Journal of Army Medical University, 2024, 46(19): 2225-2235.
DOI：

Diyou CHEN, Xinyi SHI, Pengfei WU, et al. Deep learning for volumetric assessment of traumatic cerebral hematoma[J]. Journal of Army Medical University, 2024, 46(19): 2225-2235. DOI： 10.16016/j.2097-0927.202311033.

摘要

目的

基于Trans-UNet模型构建创伤性脑出血(traumatic intracerebral hemorrhage

TICH)体积定量的深度学习方法，并与公式法进行比较。

方法

收集我院放射科于2018年5月至2023年5月收治的141例创伤性脑出血患者的CT资料，建立基于Trans-UNet模型的深度学习方法，以影像归档与通讯系统(picture archiving and communication system

PACS)人工勾画为金标准，对比其与公式法(共10种)在创伤性脑出血体积定量的准确性、一致性和测量耗时差异。

结果

PACS人工勾画的创伤性脑出血中位体积为1.167 mL，单个患者的测量中位时间为135 s。深度学习方法与PACS人工勾画的体积百分误差中位数为3.59%，Spearman相关系数为0.999(

＜0.001)，单个患者测量的中位时间仅为4.38 s；而公式法中，最低体积百分误差中位数为16.451%，最高Spearman相关系数为0.986(

＜0.001)，单个患者测量的最低中位时间为20 s。两类方法在体积百分误差和测量时间上的差异均具有统计学意义(

均＜0.001)。

结论

本研究所构建的深度学习方法相较于公式法，在测量准确性和时效上均表现更优。

Abstract

Objective

To develop a deep learning method for volumetric assessment of traumatic intracerebral hemorrhage (TICH) using the Trans-UNet model and to compare its performance with traditional formula-based methods.

Methods

CT data from 141 TICH patients admitted to Army Medical Center of PLA between May 2018 and May 2023 were collected. A deep learning method based on the Trans-UNet model was established. Manual delineation

via

picture archiving and communication system (PACS) was served as the gold standard for comparing the accuracy

consistency

and time efficiency of our method against 10 different formula-based methods for measuring the amount of TICH.

Results

The median volume of TICH

as manual delineation

via

PACS

was 1.167 mL

with a median measurement time of 135 s per patient. The median percentage error in volume between the deep learning method and manual delineation

via

PACS was 3.59%. Spearman correlation coefficient was 0.999 (

0.001)

and a median measurement time was only 4.38 s per patient. In contrast

in the formula-based methods

the lowest median percentage error in volume was 16.451%

the highest Spearman correlation coefficient was 0.986 (

0.001)

and the lowest median measurement time was 20 s for a single patient. The statistical differences were observed in percentage error in volume and measurement time between the 2 types of methods (all

0.001).

Conclusion

Our developed deep learning method for volumetric assessment of TICH is superior to the formula-based methods in terms of measurement accuracy and time efficiency.

关键词

Keywords

references

PEREL P, ROBERTS I, BOUAMRA O, et al . Intracranial bleeding in patients with traumatic brain injury: a prognostic study[J ] . BMC Emerg Med, 2009, 9: 15. DOI: 10.1186/1471-227X-9-15.

KAUVAR D S, WADE C E. The epidemiology and modern management of traumatic hemorrhage: US and international perspectives[J]. Crit Care, 2005, 9(Suppl 5): S1-S9. DOI: 10.1186/cc3779.

SUBAIYA S, ROBERTS I, KOMOLAFE E, et al . Predicting intracranial hemorrhage after traumatic brain injury in low and middle-income countries: a prognostic model based on a large, multi-center, international cohort[J ] . BMC Emerg Med, 2012, 12: 17. DOI: 10.1186/1471-227X-12-17.

HEMPHILL J C 3rd, GREENBERG S M, ANDERSON C S, et al . Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association[J ] . Stroke, 2015, 46(7): 2032-2060. DOI: 10.1161/STR.0000000000000069.

HEMPHILL J C, BONOVICH D C, BESMERTIS L, et al . The ICH score: a simple, reliable grading scale for intracerebral hemorrhage[J ] . Stroke, 2001, 32(4): 891-897. DOI: 10.1161/01.str.32.4.891.

LOPRESTI M A, BRUCE S S, CAMACHO E, et al . Hematoma volume as the major determinant of outcomes after intracerebral hemorrhage[J ] . J Neurol Sci, 2014, 345(1/2): 3-7. DOI: 10.1016/j.jns.2014.06.057.

SHAHANGIAN B, POURGHASSEM H. Automatic brain hemorrhage segmentation and classification algorithm based on weighted grayscale histogram feature in a hierarchical classification structure[J]. Biocybern Biomed Eng, 2016, 36(1): 217-232. DOI: 10.1016/j.bbe.2015.12.001.

CHEN D, BIAN L, HE H Y, et al . Evaluation of traumatic subdural hematoma volume by using image segmentation assessment based on deep learning[J ] . Comput Math Methods Med, 2022, 2022: 3830245. DOI: 10.1155/2022/3830245.

JACOBS B, BEEMS T, VAN DER VLIET T M, et al . Computed tomography and outcome in moderate and severe traumatic brain injury: hematoma volume and midline shift revisited[J ] . J Neurotrauma, 2011, 28(2): 203-215. DOI: 10.1089/neu.2010.1558.

ZHU Y, QI S H, WU H M, et al . New measuring method for the volume of intracerebral hematoma[J ] . Neurosurg Q, 2015, 25(2): 183-188. DOI: 10.1097/wnq.0000000000000021.

YANG W L, FENG Y L, ZHANG Y Y, et al . Volume quantification of acute infratentorial hemorrhage with computed tomography: validation of the formula 1/2ABC and 2/3SH[J ] . PLoS One, 2013, 8(4): e62286. DOI: 10.1371/journal.pone.0062286.

MUSCHELLI J, SWEENEY E M, ULLMAN N L, et al . PItcHPERFeCT: primary intrac ranial hemorrhage probability estimation using random forests on CT[J ] . Neuroimage Clin, 2017, 14: 379-390. DOI: 10.1016/j.nicl.2017.02.007.

GREWAL M, WIERSMA J, WESTERVELD H, et al . Automatic landmark correspondence detection in medical images with an application to deformable image registration[J ] . J Med Imaging, 2023, 10(1): 014007. DOI: 10.1117/1.JMI.10.1.014007.

ZHAO B, JIA W B, ZHANG L Y, et al . 1/2SH: a simple, accurate, and reliable method of calculating the hematoma volume of spontaneous intracerebral hemorrhage[J ] . Stroke, 2020, 51(1): 193-201. DOI: 10.1161/STROKEAHA.119.026951.

WANG C W, JUAN C J, LIU Y J, et al . Volume-dependent overestimation of spontaneous intracerebral hematoma volume by the ABC/2 formula[J ] . Acta Radiol, 2009, 50(3): 306-311. DOI: 10.1080/02841850802647039.

YUSHKEVICH P A, PIVEN J, HAZLETT H C, et al . User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability[J ] . Neuroimage, 2006, 31(3): 1116-1128. DOI: 10.1016/j.neuroimage.2006.01.015.

CHEN J N, LU Y Y, YU Q H, et al . TransUNet: transformers make strong encoders for medical image segmentation[EB/OL ] . 2021: 2102.04306. http ://arxiv.org/abs/2102.04306v1

BOUSABARAH K, RUGE M, BRAND J S, et al . Deep convolutional neural networks for automated segmentation of brain metastases trained on clinical data[J ] . Radiat Oncol, 2020, 15(1): 87. DOI: 10.1186/s13014-020-01514-6.

OKTAY O, SCHLEMPER J, LE FOLGOC L, et al . Attention U-net: learning where to look for the pancreas[EB/OL ] . 2018: 1804.03999. https://arxiv.org/abs/1804.03999v3.

ZHOU Z W, SIDDIQUEE M M R, TAJBAKHSH N, et al . UNet: redesigning skip connections to exploit multiscale features in image segmentation[J ] . IEEE Trans Med Imag, 2020, 39(6): 1856-1867. DOI: 10.1109/TMI.2019.2959609.

XIAO H G, RAN Z Q, MABU S, et al . SAUNet++: an automatic segmentation model of COVID-19 lesion from CT slices[J ] . Vis Comput, 2023, 39(6): 2291-2304. DOI: 10.1007/s00371-022-02414-4.

HUANG H M, LIN L F, TONG R F, et al . UNet 3: a full-scale connected UNet for medical image segmentation[C ] //ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Barcelona, Spain. IEEE, 2020: 1055-1059. DOI: 10.1109/ICASSP40776.2020.9053405.

HE K M, ZHANG X Y, REN S Q, et al . Deep residual learning for image recognition[C ] //2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA. IEEE, 2016: 770-778. DOI: 10.1109/CVPR.2016.90.

SUN K, XIAO B, LIU D, et al . Deep high-resolution representation learning for human pose estimation[EB/OL ] . 2019: 1902.09212. http://arxiv.org/abs/1902.09212v1

POWERS A Y, PINTO M B, TANG O Y, et al . Predicting mortality in traumatic intracranial hemorrhage[J ] . J Neurosurg, 2019, 132(2): 552-559. DOI: 10.3171/2018.11.JNS182199.

XIAO H G, SHI X Y, XIA Q L, et al . DFMA-ICH: a deformable mixed-attention model for intracranial hemorrhage lesion segmentation based on deep supervision[J ] . Neural Comput Appl, 2024, 36(15): 8657-8679. DOI: 10.1007/s00521-024-09545-w.

PENG K, LI Y L, XIA Q L, et al . MSMCNet: Differential context drives accurate localization and edge smoothing of lesions for medical image segmentation[J ] . Comput Biol Med, 2023, 167: 107624. DOI: 10.1016/j.compbiomed.2023.107624.

MACINTOSH B J, LIU Q H, SCHELLHORN T, et al . Radiological features of brain hemorrhage through automated segmentation from computed tomography in stroke and traumatic brain injury[J ] . Front Neurol, 2023, 14: 1244672. DOI: 10.3389/fneur.2023.1244672.

INKEAW P, ANGKURAWARANON S, KHUMRIN P, et al . Automatic hemorrhage segmentation on head CT scan for traumatic brain injury using 3D deep learning model[J ] . Comput Biol Med, 2022, 146: 105530. DOI: 10.1016/j.compbiomed.2022.105530.

KELLOGG R T, VARGAS J, BARROS G, et al . Segmentation of chronic subdural hematomas using 3D convolutional neural networks[J ] . World Neurosurg, 2021, 148: e58-e65. DOI: 10.1016/j.wneu.2020.12.014.

JAIN S, VYVERE T V, TERZOPOULOS V, et al . Automatic quantification of computed tomography features in acute traumatic brain injury[J ] . J Neurotrauma, 2019, 36(11): 1794-1803. DOI: 10.1089/neu.2018.6183.

WEBB A J, ULLMAN N L, MORGAN T C, et al . Accuracy of the ABC/2 score for intracerebral hemorrhage: systematic review and analysis of MISTIE, CLEAR-IVH, and CLEAR Ⅲ[J ] . Stroke, 2015, 46(9): 2470-2476. DOI: 10.1161/STROKEAHA.114.007343.

XU X H, CHEN X L, ZHANG J, et al . Comparison of the Tada formula with software slicer: precise and low-cost method for volume assessmen t of intracerebral hematoma[J ] . Stroke, 2014, 45(11): 3433-3435. DOI: 10.1161/STROKEAHA.114.007095.

贾永军, 于楠, 于勇, 等. 基于深度学习的计算机辅助诊断系统对脑出血血肿体积的测量研究[J]. 中国中西医结合影像学杂志, 2021, 19(2): 180-183. DOI: 10.3969/j.issn.1672-0512.2021.02.022.

JIA Y J, YU N, YU Y, et al . Accuracy of deep learning-based computer aided diagnosis system for measuring the intracranial hematoma volume[J ] . Chin Imag J Integr Tradit West Med, 2021, 19(2): 180-183. DOI: 10.3969/j.issn.1672-0512.2021.02.022.

田风选,常健博, 陈亦豪, 等. 基于卷积神经网络的硬膜下和硬膜外血肿分割方法的一致性评价[J]. 中国现代神经疾病杂志, 2021, 21(3): 192-196. DOI: 10.3969/j.issn.1672-6731.2021.03.011.

TIAN F X, CHANG J B, CHEN Y H, et al . Agreement evaluation of an automatic segmentation algorithm for quantifying subdural/epidural hemorrhage volume using convolution neural network[J ] . Chin J Contemp Neurol Neurosurg, 2021, 21(3): 192-196. DOI: 10.3969/j.issn.1672-6731.2021.03.011.

李涛, 陈凯, 刘耀辉, 等. 基于改进U-net网络的脑血肿CT图像分割与体积测量研究[J]. 湘南学院学报(医学版), 2022, 24(2): 1-6. DOI: 10.16500/j.cnki.1673-498x.2022.02.001.

LI T, CHEN K, LIU Y H, et al . Segmentation and volume measurement of intracerebral hemorrhage in CT images based on improved U-NET network[J ] . J Xiangnan Univ Med Sci, 2022, 24(2): 1-6. DOI: 10.16500/j.cnki.1673-498x.2022.02.001.

窦末未, 刘远健, 陈凯, 等. 人工智能对外伤性脑出血的诊断及定量临床应用研究[J]. 医学信息, 2023, 36(15): 57-63. DOI: 10.3969/j.issn.1006-1959.2023.15.010.

DOU M W, LIU Y J, CHEN K, et al . Study on diagnosis and quantitative clinical application of artificial intelligence in traumatic intracerebral hemorrhage[J ] . J Med Inf, 2023, 36(15): 57-63. DOI: 10.3969/j.issn.1006-1959.2023.15.010.

姜雪, 吕发金, 李昱仑, 等. 医学影像人工智能技术评价脑出血准确性研究[J]. 中国医学计算机成像杂志, 2021, 27(1): 1-5. DOI: 10.19627/j.cnki.cn31-1700/th.2021.01.001.

JIANG X, LV F J, LI Y L, et al . The accuracy of medical imaging artificial intelligence technology in evaluation of cerebral hemorrhage[J ] . Chin Comput Med Imag, 2021, 27(1): 1-5. DOI: 10.19627/j.cnki.cn31-1700/th.2021.01.001.

白珊玮, 韦鑫, 周治明, 等. 双源CT虚拟平扫技术联合人工智能在急性颅内出血诊断中的应用价值[J]. 陆军军医大学学报, 2023, 45(12): 1320-1327. DOI: 10.16016/j.2097-0927.202210168.

BAI S W, WEI X, ZHOU Z M, et al . Application value of dual-source virtual non-contrast CT combined with AI in diagnosis of acute intracranial hemorrhage[J ] . J Army Med Univ, 2023, 45(12): 1320-1327. DOI: 10.16016/j.2097-0927.202210168.

FONSECA J, LIU X Y, OLIVEIRA H P, et al . Mortality prediction using medical time series on TBI patients[J ] . Comput Methods Programs Biomed, 2023, 242: 107806. DOI: 10.1016/j.cmpb.2023.107806.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Integrative model combining deep learning, clinical and radiomic features enhances EGFR mutation prediction in non-small cell lung cancer

Predictive performance of CT images-based 3D ResNet18 model for identifying lung tuberculosis drug resistance

Advances in deep learning and MRI co-assisted genotyping of brain gliomas

Ultrasound thyroid nodule segmentation algorithm based on wavelet transform and CNN-Transformer

Automatic segmentation of female urine control anatomical elements and related structures in MRI images based on deep learning

Related Author

YANG Taotao

CHEN Wei

CHEN Cancan

YAN Wanying

WANG Dawei

XIONG Kunlin

SUN Zhiyuan

WANG Xianqi

Related Institution

Department of Radiology，First Affiliated Hospital， Army Medical University （Third Military Medical University）

Infervision Medical Technology Co.， Ltd

Department of Radiology， Daping Hospital， Army Medical University

Department of Diagnostic Radiology， Eastern Theatre General Hospital

Department of Medical Imaging， Chongqing Public Health Medical Center

⁰