1
|
Lee H, Lee J, Jang J, Hwang I, Choi KS, Park JH, Chung JW, Choi SH. Predicting hematoma expansion in acute spontaneous intracerebral hemorrhage: integrating clinical factors with a multitask deep learning model for non-contrast head CT. Neuroradiology 2024; 66:577-587. [PMID: 38337016 PMCID: PMC10937749 DOI: 10.1007/s00234-024-03298-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/06/2023] [Accepted: 01/25/2024] [Indexed: 02/12/2024]
Abstract
PURPOSE To predict hematoma growth in intracerebral hemorrhage patients by combining clinical findings with non-contrast CT imaging features analyzed through deep learning. METHODS Three models were developed to predict hematoma expansion (HE) in 572 patients. We utilized multi-task learning for both hematoma segmentation and prediction of expansion: the Image-to-HE model processed hematoma slices, extracting features and computing a normalized DL score for HE prediction. The Clinical-to-HE model utilized multivariate logistic regression on clinical variables. The Integrated-to-HE model combined image-derived and clinical data. Significant clinical variables were selected using forward selection in logistic regression. The two models incorporating clinical variables were statistically validated. RESULTS For hematoma detection, the diagnostic performance of the developed multi-task model was excellent (AUC, 0.99). For expansion prediction, three models were evaluated for predicting HE. The Image-to-HE model achieved an accuracy of 67.3%, sensitivity of 81.0%, specificity of 64.0%, and an AUC of 0.76. The Clinical-to-HE model registered an accuracy of 74.8%, sensitivity of 81.0%, specificity of 73.3%, and an AUC of 0.81. The Integrated-to-HE model, merging both image and clinical data, excelled with an accuracy of 81.3%, sensitivity of 76.2%, specificity of 82.6%, and an AUC of 0.83. The Integrated-to-HE model, aligning closest to the diagonal line and indicating the highest level of calibration, showcases superior performance in predicting HE outcomes among the three models. CONCLUSION The integration of clinical findings with non-contrast CT imaging features analyzed through deep learning showed the potential for improving the prediction of HE in acute spontaneous intracerebral hemorrhage patients.
Collapse
Affiliation(s)
- Hyochul Lee
- Interdisciplinary Program in Cancer Biology, Seoul National University College of Medicine, Seoul, 03080, Republic of Korea
- Department of Radiology, Seoul National University Hospital, 101 Daehak-Ro, Jongno-Gu, Seoul, 03080, Republic of Korea
| | - Junhyeok Lee
- Interdisciplinary Program in Cancer Biology, Seoul National University College of Medicine, Seoul, 03080, Republic of Korea
- Department of Radiology, Seoul National University Hospital, 101 Daehak-Ro, Jongno-Gu, Seoul, 03080, Republic of Korea
| | - Joon Jang
- Department of Biomedical Sciences, Seoul National University, Seoul, 03080, Korea
| | - Inpyeong Hwang
- Department of Radiology, Seoul National University Hospital, 101 Daehak-Ro, Jongno-Gu, Seoul, 03080, Republic of Korea.
- Department of Radiology, Seoul National University College of Medicine, 101 Daehak-Ro, Jongno-Gu, Seoul, 03080, Republic of Korea.
- Artificial Intelligence Collaborative Network, Department of Radiology, Seoul National University Hospital, Seoul, 03080, Republic of Korea.
| | - Kyu Sung Choi
- Department of Radiology, Seoul National University Hospital, 101 Daehak-Ro, Jongno-Gu, Seoul, 03080, Republic of Korea.
- Artificial Intelligence Collaborative Network, Department of Radiology, Seoul National University Hospital, Seoul, 03080, Republic of Korea.
| | - Jung Hyun Park
- Department of Radiology, Seoul Metropolitan Government Seoul National University Boramae Medical Center, Seoul, 07061, South Korea
| | - Jin Wook Chung
- Department of Radiology, Seoul National University Hospital, 101 Daehak-Ro, Jongno-Gu, Seoul, 03080, Republic of Korea
- Department of Radiology, Seoul National University College of Medicine, 101 Daehak-Ro, Jongno-Gu, Seoul, 03080, Republic of Korea
- Artificial Intelligence Collaborative Network, Department of Radiology, Seoul National University Hospital, Seoul, 03080, Republic of Korea
| | - Seung Hong Choi
- Interdisciplinary Program in Cancer Biology, Seoul National University College of Medicine, Seoul, 03080, Republic of Korea
- Department of Radiology, Seoul National University Hospital, 101 Daehak-Ro, Jongno-Gu, Seoul, 03080, Republic of Korea
- Department of Radiology, Seoul National University College of Medicine, 101 Daehak-Ro, Jongno-Gu, Seoul, 03080, Republic of Korea
- Artificial Intelligence Collaborative Network, Department of Radiology, Seoul National University Hospital, Seoul, 03080, Republic of Korea
- Center for Nanoparticle Research, Institute for Basic Science (IBS), Seoul, 08826, Republic of Korea
| |
Collapse
|
2
|
Oh JW, Park CW, Moon KC, Park JS, Jun JK. The relationship among the progression of inflammation in umbilical cord, fetal inflammatory response, early-onset neonatal sepsis, and chorioamnionitis. PLoS One 2019; 14:e0225328. [PMID: 31743377 PMCID: PMC6863554 DOI: 10.1371/journal.pone.0225328] [Citation(s) in RCA: 18] [Impact Index Per Article: 3.6] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/13/2019] [Accepted: 11/01/2019] [Indexed: 12/18/2022] Open
Abstract
OBJECTIVES No information exists about whether fetal inflammatory-response(FIR), early-onset neonatal sepsis(EONS) and chorioamnionitis(an advanced-stage of maternal inflammatory-response in extraplacental membranes) continuously increase according to the progression of inflammation in umbilical-cord(UC). The objective of current-study is to examine this-issue. METHODS Study-population included 239singleton pregnant-women(gestational-age[GA] at delivery: 21.6~36weeks) who had inflammation in extraplacental membranes or chorionic plate (CP) and either preterm-labor or preterm-PROM. We examined FIR, and the frequency of fetal inflammatory-responses syndrome(FIRS), proven-EONS, suspected-EONS and chorioamnionitis according to the progression of inflammation in UC. The progression of inflammation in UC was divided with a slight-modification from previously reported-criteria as follows: stage0, inflammation-free UC; stage-1: umbilical phlebitis only; stage-2: involvement of at least one UA and either the other UA or UV without extension into WJ; stage-3: the extension of inflammation into WJ. FIR was gauged by umbilical-cord-plasma(UCP) CRP concentration(ng/ml) at birth, and FIRS was defined as an elevated UCP CRP concentration at birth(≥200ng/ml). RESULTS Stage-0, stage-1, stage-2 and stage-3 of inflammation in UC were present in 48.1%, 15.5%, 6.7%, and 29.7% of cases. FIR continuously increased according to the progression of inflammation in UC(Kruskal-Wallis test,P<0.001; Spearman-rank-correlation test,P<0.000001,r = 0.546). Moreover, there was a significant and stepwise increase in the frequency of FIRS, proven-EONS, suspected-EONS and chorioamnionitis according to the progression of inflammation in UC(each for P<0.000005 in both chi-square test and linear-by-linear-association). Multiple logistic-regression analysis demonstrated that the more advanced-stage in the progression of inflammation in UC(i.e., stage-1 vs. stage-2 vs. stage-3), the better predictor of suspected-EONS (Odds-ratio[OR]3.358, 95%confidence-interval[CI]:1.020-11.057 vs. OR5.147, 95%CI:1.189-22.275 vs. OR11.040, 95%CI:4.118-29.592) and chorioamnionitis(OR6.593, 95%CI:2.717-15.999 vs. OR16.508, 95%CI:3.916-69.596 vs. OR20.167, 95%CI:8.629-47.137). CONCLUSION FIR, EONS and chorioamnionitis continuously increase according to the progression of inflammation in UC among preterm-gestations with inflammation in extraplacental membranes or CP. This finding may suggest that funisitis(inflammation in UC) is both qualitatively and quantitatively histologic-counterpart of FIRS, and a surrogate-marker for chorioamnionitis.
Collapse
Affiliation(s)
- Jeong-Won Oh
- Department of Obstetrics and Gynecology, Division of Maternal Fetal Medicine, Seoul National University College of Medicine, Seoul, Korea
| | - Chan-Wook Park
- Department of Obstetrics and Gynecology, Division of Maternal Fetal Medicine, Seoul National University College of Medicine, Seoul, Korea
- * E-mail:
| | - Kyung Chul Moon
- Department of Pathology, Seoul National University College of Medicine, Seoul, Korea
| | - Joong Shin Park
- Department of Obstetrics and Gynecology, Division of Maternal Fetal Medicine, Seoul National University College of Medicine, Seoul, Korea
| | - Jong Kwan Jun
- Department of Obstetrics and Gynecology, Division of Maternal Fetal Medicine, Seoul National University College of Medicine, Seoul, Korea
| |
Collapse
|