โก Quick Summary
This study developed an AI-powered model called ProgSwin-UNETR to predict prognosis in patients with hepatocellular carcinoma (HCC) undergoing transarterial chemoembolization (TACE). The model demonstrated an impressive AUC of 0.92, significantly outperforming traditional methods in risk stratification.
๐ Key Details
- ๐ Dataset: 543 arterial phase CE-MRI scans from 181 HCC patients
- ๐งฉ Features used: Multi-time-point arterial phase CE-MRI data
- โ๏ธ Technology: Deep learning model based on the Swin Transformer architecture
- ๐ Performance: AUC of 0.92, accuracy of 0.86
๐ Key Takeaways
- ๐ค AI model ProgSwin-UNETR effectively stratifies HCC patients into four distinct risk groups.
- ๐ Significant performance with an AUC of 0.92, indicating high predictive accuracy.
- ๐ก Outperformed traditional radiomics-based classifiers and mRECIST criteria in risk stratification.
- ๐ Longitudinal assessments were conducted at three time points: before treatment and after two TACE sessions.
- ๐ GradCAM++ visualizations provided insights into critical imaging regions influencing predictions.
- ๐ Multivariate Cox regression confirmed the model as an independent prognostic factor.
- ๐ Potential for personalized treatment strategies in HCC management.
๐ Background
Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality worldwide. The prognosis for patients undergoing transarterial chemoembolization (TACE) can vary significantly based on individual responses to treatment. Traditional imaging techniques often fall short in providing timely and accurate prognostic information. The integration of artificial intelligence into medical imaging presents a promising avenue for enhancing prognostic capabilities and personalizing treatment plans.
๐๏ธ Study
This retrospective study involved the collection of 543 arterial phase CE-MRI scans from 181 HCC patients who underwent TACE. The researchers aimed to develop and validate an AI model, ProgSwin-UNETR, utilizing multi-time-point imaging data to stratify prognosis effectively. The model was trained and evaluated using rigorous methodologies, including fourfold cross-validation and comparative analysis against traditional classifiers.
๐ Results
The ProgSwin-UNETR model achieved an accuracy of 0.86 and an AUC of 0.92 for the four-class prognosis stratification task. Notably, it was the only approach that provided statistically significant risk stratification across the entire cohort and within both TACE-alone and TACE + MWA subgroups (p < 0.005). The model’s outputs were further validated through multivariate Cox regression analysis, establishing it as a robust independent prognostic factor (p = 0.01).
๐ Impact and Implications
The findings from this study highlight the transformative potential of AI in the field of oncology, particularly in the management of HCC. By utilizing advanced imaging techniques and deep learning algorithms, healthcare providers can achieve more accurate prognostic assessments, leading to improved patient outcomes. This model not only enhances our understanding of disease progression but also paves the way for personalized treatment strategies tailored to individual patient needs.
๐ฎ Conclusion
The development of the ProgSwin-UNETR model marks a significant advancement in the use of AI for predicting prognosis in HCC patients undergoing TACE. With its high accuracy and robust performance, this model holds promise for enhancing clinical decision-making and optimizing treatment pathways. Continued research and validation in diverse patient populations will be essential to fully realize the potential of AI in improving cancer care.
๐ฌ Your comments
What are your thoughts on the integration of AI in cancer prognosis? We would love to hear your insights! ๐ฌ Leave your comments below or connect with us on social media:
Artificial intelligence for multi-time-point arterial phase contrast-enhanced MRI profiling to predict prognosis after transarterial chemoembolization in hepatocellular carcinoma.
Abstract
PURPOSE: Contrast-enhanced magnetic resonance imaging (CE-MRI) monitoring across multiple time points is critical for optimizing hepatocellular carcinoma (HCC) prognosis during transarterial chemoembolization (TACE) treatment. The aim of this retrospective study is to develop and validate an artificial intelligence (AI)-powered models utilizing multi-time-point arterial phase CE-MRI data for HCC prognosis stratification in TACE patients.
MATERIAL AND METHODS: A total of 543 individual arterial phase CE-MRI scans from 181 HCC patients were retrospectively collected in this study. All patients underwent TACE and longitudinal arterial phase CE-MRI assessments at three time points: prior to treatment, and following the first and second TACE sessions. Among them, 110 patients received TACE monotherapy, while the remaining 71 patients underwent TACE in combination with microwave ablation (MWA). All images were subjected to standardized preprocessing procedures. We developed an end-to-end deep learning model, ProgSwin-UNETR, based on the Swin Transformer architecture, to perform four-class prognosis stratification directly from input imaging data. The model was trained using multi-time-point arterial phase CE-MRI data and evaluated via fourfold cross-validation. Classification performance was assessed using the area under the receiver operating characteristic curve (AUC). For comparative analysis, we benchmarked performance against traditional radiomics-based classifiers and the mRECIST criteria. Prognostic utility was further assessed using Kaplan-Meier (KM) survival curves. Additionally, multivariate Cox proportional hazards regression was performed as a post hoc analysis to evaluate the independent and complementary prognostic value of the model outputs and clinical variables. GradCAMโ+โโ+โwas applied to visualize the imaging regions contributing most to model prediction.
RESULTS: The ProgSwin-UNETR model achieved an accuracy of 0.86 and an AUC of 0.92 (95% CI: 0.90-0.95) for the four-class prognosis stratification task, outperforming radiomic models across all risk groups. Furthermore, KM survival analyses were performed using three different approaches-AI model, radiomics-based classifiers, and mRECIST criteria-to stratify patients by risk. Of the three approaches, only the AI-based ProgSwin-UNETR model achieved statistically significant risk stratification across the entire cohort and in both TACE-alone and TACEโ+โMWA subgroups (pโ<โ0.005). In contrast, the mRECIST and radiomics models did not yield significant survival differences across subgroups (pโ>โ0.05). Multivariate Cox regression analysis further demonstrated that the model was a robust independent prognostic factor (pโ=โ0.01), effectively stratifying patients into four distinct risk groups (Class 0 to Class 3) with Log(HR) values of 0.97, 0.51, -0.53, and -0.92, respectively. Additionally, GradCAMโ+โโ+โvisualizations highlighted critical regional features contributing to prognosis prediction, providing interpretability of the model.
CONCLUSION: ProgSwin-UNETR can well predict the various risk groups of HCC patients undergoing TACE therapy and can further be applied for personalized prediction.
Author: [‘Yao L’, ‘Adwan H’, ‘Bernatz S’, ‘Li H’, ‘Vogl TJ’]
Journal: Radiol Med
Citation: Yao L, et al. Artificial intelligence for multi-time-point arterial phase contrast-enhanced MRI profiling to predict prognosis after transarterial chemoembolization in hepatocellular carcinoma. Artificial intelligence for multi-time-point arterial phase contrast-enhanced MRI profiling to predict prognosis after transarterial chemoembolization in hepatocellular carcinoma. 2025; (unknown volume):(unknown pages). doi: 10.1007/s11547-025-02043-6