Preprints with The Lancet is a collaboration between The Lancet Group of journals and SSRN to facilitate the open sharing of preprints for early engagement, community comment, and collaboration. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early-stage research papers that have not been peer-reviewed. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. The findings should not be used for clinical or public health decision-making or presented without highlighting these facts. For more information, please see the FAQs.
Predicting Shunt-Dependent Post-Traumatic Hydrocephalus in the Early Stage after Decompressive Craniectomy: A Radiomics Method Based on CT
Background: Shunt-dependent post-traumatic hydrocephalus (SDPTH), as one of the significant complications following decompressive craniectomy (DC) for traumatic brain injury (TBI), adversely impacting neurological recovery. Early detection and intervention can enhance the long-term prognosis for these patients. This study aims to develop and validate a CT radiomics model for accurately predicting the occurrence of SDPTH.
Methods: This retrospective study analyzed data from 209 patients who underwent DC due to TBI. The patients were divided into a training set (n=173) and a validation set (n=36) based on their treatment centers. Eight features were selected using the Least Absolute Shrinkage and Selection Operator (Lasso) regression method. To address data imbalance, the synthetic minority over-sampling technique (SMOTE) was applied. Subsequently, five commonly used machine learning models were developed based on these features. The performance of these models was thoroughly evaluated using metrics such as the area under the receiver operating characteristic curve (AUC), accuracy, specificity, sensitivity, and F1 score.
Findings: Among the five radiomics models, the ExtraTrees model demonstrated the best performance. It achieved an AUC of 0.93 on the training set, with a sensitivity of 0.88, specificity of 0.82, accuracy of 0.84, and an F1 score of 0.85. Similarly, on the external validation set, it maintained an AUC of 0.93, with a sensitivity of 0.86, specificity of 0.83, accuracy of 0.83, and an F1 score of 0.67. These results reflect the model's robust predictive performance and strong generalization capability.
Interpretation: This study demonstrates that a CT radiomics model can effectively predict the occurrence of SDPTH in patients undergoing DC for TBI. The model aids clinicians in identifying high-risk patients and developing personalized treatment plans.
Funding: This work was supported by National Natural Science Foundation of China(No.81801225).
Declaration of Interest: We declare no competing interests.
Ethical Approval: Ethical approval was obtained from the ethics committees of both centers, and informed consent was acquired from all patients.
Wu, Hongbin and Wang, Ziyu and Chen, Jianping and Lu, Shenghua and Yang, Mengshi and Zhuang, Yuan and Liu, Baiyun and Gao, Guoyi and Zhang, Yupeng and Tian, Runfa, Predicting Shunt-Dependent Post-Traumatic Hydrocephalus in the Early Stage after Decompressive Craniectomy: A Radiomics Method Based on CT. Available at SSRN: https://ssrn.com/abstract=5024382 or http://dx.doi.org/10.2139/ssrn.5024382
Download This Paper